1. Alchemical-Body White(Siddha Traditions) MRML

THE ALCHEMICAL BODY

Siddha Traditions in Medieval India

DAVID GORDON WHITE

THE UNIVERSITY OF CHICAGO PRESS / CHICAGO AND LONDON

for Catherine David Gordon White is associate professor of religious studies at the University of California, Santa Barbara. He is the author of Myths of the Dog-Man (1991), also published by the University of Chicago Press.

The University of Chicago Press, Chicago 60637 The University of Chicago Press, Ltd., London C1996 by The University of Chicago All rights reserved. Published 1996 Printed in the United States of America 05 04 03 02 01 00 99 98 97 96 1 2 3 4 5 ISBN: 0-226-89497-5 (cloth)

Library of Congress Cataloging-in-Publication Data

White, David Gordon. The alchemical body : Siddha traditions in medieval India / David Gordon White. p. cm. Includes bibliographical references and index. ISBN o-226-89497-5 (cloth : alk. paper) 1. Siddhas. 2. Alchemy-Religious aspects- Tantrism. 3. Yoga, Hatha. 4. Tantrism. 5. Natha sect. I. Title. BL1241.56.W47 1996 294.5'514-dczo 96-16977 CIP

®The paper used in this publication meets the minimum requirements of the American National Standard for Information Sciences-Permanence of Paper for Printed Library Materials, ANSI Z39.48- 1984.

CONTENTS

PREFACE / ix

NOTE ON TRANSLITERATION / x

ABBREVIATIONS / xvii

ONE Indian Paths to Immortality / 1

TWO Categories of Indian Thought: The Universe by Numbers / 15

THREE The Prehistory of Tantric Alchemy / 48

FOUR Sources for the History of Tantric Alchemy in India / 78

FIVE Tantric and Siddha Alchemical Literature / 123

SIX Tantra in the Rasārnava / 171

vili Contents

SEVEN PREFACE Corresponding Hierarchies: The Substance of the Alcbemical Body / 184

EIGHT Homologous Structures of the Alchemical Body / zi8

NINE The Dynamics of Transformation in Siddha Alchemy / 264

TEN In the new age India of the 199os, it has become popular, even fashionable, Penetration, Perfection, and Immortality / (o to have the name of a tantrika, a kind of all-purpose sexologist, medicine man, and shaman, in one's little black book of phone numbers. This same phenomenon has brought with it the appearance, preceding the title page EPILOGUE The Siddha Legacy in Modern India / u5 of books on magic and tantra, of "disclaimers" to the effect that said book does not guarantee the results of the techniques it is treating and that its editors are not responsible for unhappy side-effects of said techniques NOTES / 35 when they are practiced in the privacy of one's home. The present work carries no such disclaimer because it in no way purports to be a "how-to"

SELECTED BIBLIOGRAPHY / book for realizing immortality. Nor is this a study in the history of Indian medicine or science: a great number of Indian scholars and scientists as well as a growing number of western authors have written excellent works INDEX / 55 on the matters I will be treating from these perspectives, incorporating into their writings comprehensive overviews of Indian chemistry, human physiology, pharmacology, and therapeutics. The present work is rather a history-of-religions study of the medieval Siddha traditions of Hindu alchemy and hatha yoga, which formed two im- portant fields of theory and practice within the vast current of Indian mys- ticism known as tantra. It is the religious and, more specifically, tantric features of these interpenetrating traditions that I will be treating in these pages, from both a historical and a phenomenological perspective. In the main, this will be a study of the language of mystic experience and expres- sion, and it will be from the standpoint of language that I will chart out the theoretical, symbolic, and analogical parameters of the alchemical and hathayogic disciplines within their broader tantric and Hindu contexts. And, working from the semantic and symbolic fields of meaning that the alchemical material generates, this study will also look at a much wider array of Hindu and Indian phenomena through "alchemical eyes." This will furthermore be a scholarly work, nearly entirely divorced from

xi Preface Preface

any ground of personal mystical experience. Apart from a short period of schooling in batha yoga undertaken in Benares in 1984-85, I have never for which these words and images are so many signposts is to generate a

experienced anything that one could qualify as a genuine master-disciple symbolic lexicon from the multitude of intersecting words and images the

relationship. I have never levitated, read other people's minds, or even seen sources offer.7 Rather than exposing the doctrines of any single school,

auras. This being the case, it may well be that I belong to the great mass movement, or exegetical tradition, this study seeks to lay bare the words,

of those who "must go on blundering inside our front-brain faith in Kute images, and logic that a wide swath of the Hindu, and particularly Saiva

Korrespondences, hoping that for each psi-synthetic taken from Earth's (and tantric), population always already assume to be the case prior to giv-

soul there is a molecule, secular and more or less ordinary and named, over ing voice to their doctrines.

here-kicking endlessly among the plastic trivia, finding in each Deeper This is, in the main, a study of a pervasive Indian worldview from a

Significance and trying to string them all together like terms of a power tantric and alchemical perspective. Now, if we follow Douglas Brooks

series hoping to zero in on the tremendous and secret Function whose when he maintains that Hindu tantrism has been treated as "an unwanted

name, like the permuted names of God, cannot be spoken ... to make stepchild in the family of Hindu studies";8 and if, as Betty Dobbs has writ-

sense out of, to find the meanest sharp sliver of truth in so much repli- ten with regard to its rejection of alchemy as so much fuzzy mysticism,

cation."! that "modern science, like adolescence, denies its parentage,"9 then the

Ultimate reality is beyond my reach, either to experience or express. I subject of this study has a troubled family life. Perhaps it is the stepparents'

nonetheless hope that these pages may serve to bridge a certain gap be- and adolescents' judgment that one ought to question here.

tween raw experience and synthetic description, and thereby contribute to an ongoing tradition of cultural exchange that is at least as old as the This book is the fruit of twelve years of research begun at the University

Silk Road. of Chicago, where I began to translate the banīs, the mystic vernacular poems of Gorakhnath, under the direction of Professor Kali Charan Bahl.

In reading these pages, the reader may come to experience a sensation of The use of metaphor and imagery in these poems reminded me of similar

vertigo, as the horizon of one mystic landscape opens onto yet another language from an alchemical work entitled the Rasarnava (The Flood of

landscape, equally vast and troubling in its internal immensity. It may be Mercury), which I had attempted to translate (with mitigated success) for

that these landscapes,2 with their dizzying multitudinous levels of self- a self-styled French mystic a few years earlier.

interpretation, may inspire analysis by psychologists of both the armchair It was on the basis of these first tentative identifications that I embarked

and professional varieties.3 I believe, however, that the most useful western in earnest on the present research, going to India in 1984-85 under the

companion to the present study is the work of the French philosopher Gas- auspices of a grant from the American Institute for Indian Studies. It had

ton Bachelard entitled The Poetics of Space.+ Bachelard's work is a phenome- been my intention, in undertaking my research tour, to find a living yogin-

nological study of literary depictions of the experience of space, from cellar alchemist and to sit at his feet until I had solved all the riddles the Rasār-

to attic, from Chinese boxes to the interiors of seashells. In these pages I nava and the Gorakh Bani had posed for me. This endeavor was a total

will endeavor to follow just such a phenomenological approach, pointing failure. There were no alchemists to be found in the places in which I

out homologies where the sources would seem to indicate connections in- sought them out (although I did meet a number of amateurs and charla-

ternal to the traditions themselves, without attempting to force the textual tans), and the few Nath Siddhas who struck me as genuine practitioners of

data into any preconceived model.5 the hatha yoga taught by Gorakhnath made it clear that they would be will-

I treat alchemical and tantric discourse as self-referential, as part and ing to divulge their secrets to me only after a long period of discipleship.

parcel of a self-enclosed network of specifically Indian symbols and signs, Being a westerner in a hurry, I spurned this path, in spite of its many

my assumption being that the words and images of these traditions are attractions, for that of the textualist. First in India, under the guidance of

always referring, before all else, to other words and images.6 Therefore, a number of professors of Ayurvedic studies, and later in Europe and the

the best way to formulate a theory concerning the nature of the experiences United States, then in India and Nepal once again, under the auspices of a grant from the Council for the International Exchange of Scholars, I fol-

NOTE ON TRANSLITERATION

Unless otherwise noted, all transliterations from the Sanskrit follow stan- dard lexicographical usage, with the following exceptions: (1) toponyms still in use are transliterated without diacritics (thus Srisailam and not Śrīśailam), except in the case of sites identified with deities (thus Kedār- nāth and not Kedarnath); (2) names of authors and editors from the colo- nial and postcolonial periods are transliterated without diacriticals; (3) the term Nath is transliterated in its modern Hindi form as opposed to the Sanskritic Nātha; (4) proper names of historical Nath Siddhas are translit- erated with the -näth suffix, as opposed to the Sanskritic-natha (however, see note 8 to chapter r).

ABBREVIATIONS

ĀK Ånandakanda of MBbT Matrkabbeda Tantra. Mahabhairava. MSL Maharāja Mān Singh Library, ASL Anup Sanskrit Library, Mehrangarh Fort, Jodhpur, Lalgarh Palace, Bikaner, India. India. NNA Nepal National Archives, BbP Bhūtiprakarana of the Kathmandu, Nepal. Goraksa Samhitā. NSC Ramlal Srivastav, ed., "Nātb GAU Gujarat Ayurved University, Siddh Carit" Visesank (Yog Jamnagar, India. Vani, special issue no. I for HT Sanjukta Gupta, Dirk Jan Hoens, and Teun Goudriaan, 1984). Gorakhpur:

Hindu Tantrism. Handbuch Gorakhnath Mandir, 1984 Ocean der Orientalistik, 2.4.2 The Ocean of Story Being C. H.

(Leiden: Brill, 1979) Tawney's Translation of

HTSL Teun Goudriaan and Somadeva's Kathā Sarit Sāgara

Sanjukta Gupta, Hindu (or Ocean of Streams of Story),

Tantric and Sākta Literature ed. N. M. Penzer, 1o vols.

(Wiesbaden: Harrassowitz, (London: Chas. J. Sawyer, Ltd., 1924-28). 1981). HYP Hatbayogapradīpikā of RA Rasārņava.

Svātmarāma RAK Rasārņavakalpa.

KCKT Kākacandīśvara Kalpa Tantra. RC Rasendracūdāmani

KCM Kakacandeśvarīmata. RHT Rasabrdaya Tantra of

KbV Khecari Vidyā of Ādinātha. Govinda. Kaulajnānanirnaya of RM Rasendra Mangala of Srīman Matsyendranāth. Nāgārjuna. KM Kubjikāmata RPS Rasaprakāša Sudbākara of KPT Kaksaputa Tantra of Siddha Yasodhara. Nāgārjuna. RRÅ Rasaratnakara of Nityanātha.

xvili Abbreviations

RRS Rasaratnasamucchaya of ȘSS Şatsāhasra Sambitā. ONE Vagbhatta II Rasopanisat. SV "Siddh Vandanām." RU RV Rg Veda. SuT Svacchanda Tantra. Tantrāloka of Abhinavagupta. Indian Paths to Immortality SDS Sarvadarsana Samgraba of TÀ

Madhava VRÅ Varņaratnākara SSP Siddba Siddbanta Paddbati of YRM Yogaratnamala Goraksanātha YSA Yogisampradāyāviskrti

The emperor Aurangzeb issued a firman to Anand Nath, the abbot of Jakhbar, an obscure monastery in the Punjab, in 1661 or 1662:

The letter sent by Your Reverence has been received along with two tolabs of quicksilver. However, it is not so good as Your Reverence had given us to understand. It is desired that Your Reverence should carefully treat some more quicksilver and have that sent, without un- necessary delay. A piece of cloth for a cloak and a sum of twenty-five rupees which have been sent as an offering will reach (Your Rever- ence). Also, a few words have been written to the valiant Fateh Chand to the effect that he should always afford protection.'

The greatest Mogul persecutor of Hinduism in history offers his pro- tection to a Hindu abbot named Nath in exchange for twenty grams of treated mercury. What is the story behind this curious missive?

I. Sexual Fluids in Medieval India

Some time around the sixth century A.D., a wave of genius began to sweep over India, a wave that has yet to be stilled. This wave, which took the form of a body of religious thought and practice, has been interpreted in a number of different ways by Indians and westerners alike. What some have called madness and abomination, others have deemed a path to ecstasy or the sublime. Such have been the evaluations of this phenomenon, which has, over some fourteen hundred years, never ceased to enthuse and con- found. The Indians who innovated this body of theory and practice called it tantra, "the warp (of reality)." The word has a most ancient pedigree. Its root, tan, means "to stretch," as one would a thread on a loom (also called

2 3 Chapter One Indian Paths to Immortality

tantra) or, in Vedic parlance, a body (tanu) to be sacrificed on an altar not historical figures but rather demigods and intermediaries between the within the ritual framework (tantra).2 Those persons who followed the way human and the divine. Cults of these semidivine Siddhas go back to at least of tantra were called tantrikas, and their written and orally transmitted the beginning of the common era; they and their peers the Vidyādharas works the Tantras. Indian tantrism,3 in its Hindu, Buddhist, and Jain varieties, did not (Wizards) are a standard fixture of Indian fantasy and adventure literature throughout the medieval period.10 Central to these cults was their "popu- emerge out of a void. It was on the one hand influenced by cultural inter- actions with China, Tibet, central Asia, Persia, and Europe, interactions lar" soteriology, which had little in common with the "authorized" soteri-

which had the Silk Road and medieval maritime routes and ports as their ologies of Vedic and classical Hinduism. The worlds of the Siddhas and Vidyādharas were the closest homologue India has known to popular west- venue. Much more important, however, were the indigenous Indian roots of tantrism, which was not so much a departure from earlier forms of Hin- ern notions of heaven as a place of sensual gratification and freedom from

duism as their continuation, albeit in sometimes tangential and heterodox the human condition. Those capable of acceding to these atmospheric lev-

ways. This book explores the uniquely Indian foundation of tantrism. els remained there, liberated from the fruits of their acts (karma) and for-

More specifically, this book is an inquiry into those Hindu sectarian groups ever exempted from the lower worlds of rebirth (samsara) but not divested

that have come to be known as the Siddhas, which, appropriating traditions of their individuality as is the case with the impersonal workings of release into the Absolute (moksa). A precursor of the Puranic notion of the "sev- that were more ancient than those of tantrism itself, did not in fact fully flower until the twelfth and thirteenth centuries. As a loosely structured enth heaven" of brabmaloka or satyaloka, the world of the Siddhas was a

religious community identified with a particular body of practice, the Sid- place that endured even beyond the cyclic dissolution (pralaya) at the close

dhas have had greater staying power than the täntrikas and continue to of a cosmic eon (kalpa). This popular tradition, whose reflection is found

form a visible part of the Indian religious landscape. in the lower hierarchies of the Hindu and Buddhist pantheons, in adven- ture and fantasy literature, and in humble shrines to these anonymous As a common noun, siddha* means "realized, perfected one," a term gen- demigods, lay beyond the pale of brahmanic control and legitimation. So erally applied to a practitioner (sādhaka, sādhu) who has through his prac- tice (sādhana) realized his dual goal of superhuman powers (siddhis, "real- too would the medieval Siddha movements, which appropriated for them-

izations," "perfections") and bodily immortality (jīvanmukti). As a proper selves, with certain modifications, the preexisting Siddha soteriology. The most important innovation of these medieval Siddha traditions (the noun, Siddha becomes a broad sectarian appellation, applying to devotees Nath and Rasa Siddhas in particular) was the concrete and coherent of Siva in the Deccan (Mahesvara Siddhas),5 alchemists in Tamil Nadu (Sit- method they proposed for the attainment of the Siddha world and Siddha tars), a group of early Buddhist tāntrikas from Bengal (Mahāsiddbas, Sid- status. This is what had been lacking in the earlier Siddha cults: the belief dhācāryas), the alchemists of medieval India (Rasa Siddhas)6 and, most espe- system was there, but the notions of how to reach that blessed abode were cially, a mainly north Indian group known as the Nath Siddhas.7 These last two groups greatly overlapped one another, with many of the vague at best. Certain traditions maintained that it could be reached through travel," others through the miraculous intervention of the Sid- most important Nath Siddhas-Gorakh, Matsyendra, Carpati, Dattātreya, Nāgnāth, Ādināth,8 and others-being the authors (if only by attribution) dhas one propiated,12 others through more serendipitous means. The later medieval Siddha movements proposed the following working principle: or transmitters of a wide array of revealed yogic and alchemical teachings. The medieval Nath Siddhas and Rasa Siddhas further interacted with a mere humans could, through their tantric, yogic, and alchemical practice, climb the ladder of being and accede to the ranks of the semidivine Sid- third group. This was the paścimamnāya (Western Transmission), a Sakta dhas. In this new perspective one could, by perfecting oneself, transform sect devoted to the worship of the goddess Kubjika which, based mainly perfected role models into colleagues. A trace of the notion of a primordial in Nepal, also incorporated tantric,9 yogic, and alchemical elements into ontological difference between those born perfect and those who made its doctrine and practice. themselves perfect (not unlike the difference between old money and new) A major point of convergence between these three groups, within the remains in works which categorize the Siddhas into the three oghas broader tantric matrix, was their cult of the Siddhas who were for them (streams)-the divine, the perfected, and the human-but the dividing line

4 5

Chapter One Indian Paths to Immortality

between them was a dotted one that could be crossed through a systematic ergy, generally called the kundalini, and a male principle, identified with body of esoteric practice.13 Siva, both of which were located within the subtle body. An intricate meta- Apart from this common heritage, a second point of convergence be- physics of the subtle body-its relationship to the brute matter of the gross tween the Nath Siddhas, Rasa Siddhas, and the Western Transmission lies body as well as to the universal divine life force within, the bipolar dynam- in their common body of mystic doctrines and practices involving sexual ics of its male and female constituents, etc .- was developed in every tan- fluids-male and female sexual fluids, to be sure, but ever so much more. tric school.20 Since the time of the Vedas, rasa-the fluid element found in the universe, It was especially within two tantric sects, the Western Transmission and sacrifice, and human beings-has been more or less identified by Indians the Yoginī Kaula (transmitted by Matsyendra), that a practical concomitant with the fount of life. All fluids, including vital fluids in humans, plant to this speculative-and in some cases gnoseological or soteriological- resins, rain, the waters, and the sacrificial oblation, are so many manifesta- metaphysics came to be elaborated. This was hatha yoga, the "method of tions of rasa."+ So too, since at least the dawn of the common era,'5 Indians violent exertion," whose system of the six cakras ("wheels [or circles] of have known that the miracle of conception occurs through the union of transformation") became the centerpiece of the doctrine and practice male and female vital fluids, semen and uterine blood. With early tantrism, of the Nath Siddhas-who claim their origins in the person and teachings these procreative fluids came to be conceived as "power substances" for of Matsyendranath.21 For the Nath Siddhas, the siddhis and jivanmukti were the worship of and ultimately the identification with gods and goddesses the direct results of the internal combination and transformation of sexual whose boundless energy was often portrayed as sexual in nature. Nearly fluids into amrta, the divine nectar of immortality. always, the god in question was some form of Siva, the god whose worship Matsyendranath and the founders of the Western Transmission were in the form of a linga (phallus) dates from at least the second century B.C.16 not alone, however, in their persistent emphasis on the sexual fluids as The way to becoming a "second Siva"-for this has nearly always been the (generally internalized) power substances, rather than simply as by- goal of tantric practice in its various forms-was, in early tantrism, real- products of a transubstantiating experience of bliss. At about the same time ized through the conduit of a horde of wild goddesses (which the tāntrikas as their hathayogic systems were being elaborated, the matter of sexual identified with their human consorts), generally known as yoginīs. These fluids was being broached from a novel and rather unexpected angle by a "bliss-starved" goddesses, attracted by offerings of mingled sexual fluids, third group. These were the Rasa Siddhas, the alchemists of medieval In- would converge into the consciousness of the practitioner, to transform dia, whose doctrines are best summed up in a classic aphorism from the him, through their limitless libido, into a god on earth.17 foundational Rasārņava: yathā lohe tathā debe, "as in metal, so in the body."22 Following the brilliant tenth- through eleventh-century reconfiguration In a universe that was the ongoing procreation of the phallic god Siva of Trika Kaulism by Abhinavagupta and others, most of the messy parts of and his consort the Goddess,23 a pair whose procreative activity was mir- tantric practice (at least outward practice) were cleaned up, aestheticized, rored in the fluid transactions and transformations of human sexuality, in and internalized in different ways.18 For the later "high" tantric schools, a universe whose every facet reflected the fundamental complementarity the cult of the yoginis and the ritual production, offering, and consumption of the male and female principles, the mineral world too had its sexual of sexual fluids were continued, but only within the restricted context of valences and fluids. In the case of the Goddess, her sexual emission, her the "secret practice" of an inner circle of initiates. Outwardly, however, seed, took the form of mica, while her uterine or menstrual blood was ritual sexuality had undergone a paradigm shift. Sexual fluids themselves identified with sulfur. There are a number of reasons for these identifica- were no longer the way to godhead; rather, it was in the bliss of sexual tions, not the least of which are chemical: mica and sulfur are important orgasm that one realized god-consciousness for oneself.19 reagents in the purification and activation of the mineral homologue to In certain cases, all such transactions involving sexual fluids became divine semen. This is mercury, and if there ever was an elective affinity to wholly internalized and incorporated into the so-called subtle body be found at the interface between chemistry and theology, this is it. For (sūksma sarīra). Here, all humans were viewed as essentially androgynous what a miraculous mineral mercury is! Mercury is a shining liquid, amaz- with sexual intercourse an affair between a female serpentine nexus of en- ingly volatile, seemingly possessed of a life of its own: what better homol-

6 7 Chapter One Indian Paths to Immortality

ogy could one hope to find for the semen of a phallic god? But this is not sience and attendant sorrows. On the other, for persons still trapped in this all. Mercury's chemical behavior as well is nothing short of miraculous, world, a good Siddha is hard to find. and as such it stands, in the words of an early twentieth-century scholar Hindu tantrism disappeared as a major sectarian phenomenon a number alchemist as the "central idea upon which the whole structure of the Hindu of centuries ago, a victim of its own excesses.27 These excesses were pri- Chemistry is erected: viz., the fact that mercury can be made to swallow, marily of two orders. The first and best documented is nonetheless less by special processes, a considerable quantity of gold or other metals, with- important than the second. This excess was one of bad publicity. In seek- out any appreciable increase in the weight of the swallowing mercury."24 ing to truly live out their principles of nondifference-between god and Mercury, which when "swooned" drives away disease, "killed" revives itself, and "bound" affords the power of flight,25 is the presence in the creature, elite and preterite, squalor and grandeur, the exalted and the demented-many täntrikas, openly indulging in cross-caste adultery, co- mineral world of the sexual essence of the Absolute. As such, it is as prophagy, and all manner of other purity violations and antisocial behavior all-absorbing as Siva who, at the end of cyclic time, implodes the entire (or at least openly claiming to do so), were simply revolting to the general universe into his yogic body, thereby transforming existence into essence. public. The second excess, which truly sounded the death knell of tantrism This is precisely what occurs in alchemical reactions. A "seed" (bija) of as an important religious movement, came as the result of a sea change gold or silver is planted in mercury (whose powers of absorption have been in tantric theory and practice. Following Abhinavagupta, tantrism became massively enhanced through a series of treatments in sulfur, mica, and transformed into an elite mystic path that was all too complicated, refined, other mainly "female" elements), which then becomes possessed of a and cerebralized for common people to grasp. The man on the street could "mouth" capable of "swallowing," of absorbing into itself, according to not recognize himself in its discourse. It bore too little resemblance to his the alchemical scriptures, millions, even billions and trillions, of times its experience as a mortal being inhabiting a body doomed to age and die, mass in base metals. These are thereby transmuted into gold, and in a tra- entangled in the meantime within a network of family and social relations; dition in which "gold is immortality,"26 that's saying a mouthful. All that wielding plowshares, hammers, and the like; living, loving, and dying on remains is for the alchemist to swallow the mercury in question to himself the trampled earth of a village his people had inhabited for hundreds of become a second Siva, an immortal superman (Siddha) whose every bodily years. The thirty-six or thirty-seven metaphysical levels of being were in- secretion becomes transmutative and transubstantiating. In tandem with his work in the laboratory, the Hindu alchemist also engages in the prac- comprehensible to India's masses and held few answers to their human concerns and aspirations. tice of hatha yoga, as well as a certain number of erotico-mystical tantric For the Näth Siddhas, whose institutionalized sectarian orders (sampra- operations involving the sexual fluids that he and his female laboratory dāyas) mainly grew out of earlier and more heterodox Saiva orders (the assistant generate in order to catalyze reactions between divine sexual fluids in their mineral forms. In the end, all is a continuity of sexual Pāśupatas and Kāpālikas in particular), the "brahmanization" of tantrism and its departure from the realm of the concrete into that of the sublime fluids. came as a boon of sorts. For whereas "high" tantrism was now mainly of- fering transcendence of the world, the Nath Siddhas' path continued to offer concrete and relatively accessible power in the world. For the masses, 2. Tāntrikas, Siddhas, and Yogis as well as for kings whose concerns were often more this-worldly than those of brahman metaphysicians, the Näths and many of their fellow Sid- The sole surviving heirs to this medieval legacy are the Nath Siddhas, who dhas became the supernatural power brokers of medieval India. The Sid- continue to be revered, on a popular level at least, as India's masters of dhas, yogins and alchemists that they are, have always been technicians of yoga and wizards of alchemy, the last living guides along the secret paths the concrete: specialists in the concrete transmutation of base metals into to supernatural power and bodily immortality. Theirs is a powerful legacy. gold and the concrete transformation of mortal, aging man into a per- On the one hand, they are perfected immortals who have chosen to remain fected, immortal superman, masters of the natural processes rather than in the world of men, moving through it even as they transcend its tran- mere victims of or bystanders to them. Theirs has always been a path to

8 9 Chapter One Indian Patbs to Immortality

mastery and raw, unadulterated power-mastery over the forces of nature, Vedic, devotional, and "high" tantric religion. On the other hand, they are including the inexorable processes of aging and death, and dominion over defined by certain features of their sectarian affiliations and practices: heirs the temporal powers of even the greatest kings and armies. to the heterodox Pāśupatas and Kāpālikas of an earlier age, they are devo- In this context, the Näth Siddhas have often been cast in the role of tees of terrible forms of Siva (usually Bhairava) who besmear themselves kingmakers, elevating untested boys to the thrones of kingdoms through- with ashes, leave their hair uncut, and continue to adhere to the practices out medieval south Asia, at times bringing down mighty tyrants in the pro- cess.28 But India has also always been a country of villages, and the Nath of "primitive" tantrism. As such, their "yoga" is more closely identified, in

Siddhas, whose backgrounds have generally been humble, have long been the jaundiced eyes of their critics, with black magic, sorcery, sexual perver- sion, and the subversion of alimentary prohibitions than with the practice the special champions of village India. These are cowherd boys who, hav- ing performed minor miracles at an early age, were initiated into a Siddha of yoga in the conventional sense of the term. In recent times, "Yogi" has been most specifically applied to the Nath Siddhas, who are widely known order and grew up to be immortal, hail-stopping "god men." Many are as Kānphata (Split-eared, for the very visible earrings they wear in holes the accounts I have heard, from traveling salesmen, university students, bored through the thick of their ears, the hallmark of the order) Yogis or monastic novices, and village plowboys alike, of these perfected beings who have dotted and even defined the religious landscape of village India Jogis-a term that they themselves eschew. This book, then, is about those tantric movements and sects which with their awesome, death-defying presence. called themselves and continue to call themselves Siddhas, but which The Näth Siddhas' persistent popular success, coupled with their gen- were-following the "brahmanization" of the tantrism with which they erally humble social backgrounds, the relative accessibility of their path, had interacted throughout their early development-branded as Yogis by and the this-worldly focus of their practices and goals, has long made them the Hindu orthodoxy. That these Yogis were alchemists is borne out by no the object of scorn and censure on the part of India's social, cultural, and less a person than Marco Polo who, describing a group of ciugi (Jogis) religious elites-the upper castes, urban intelligentsia, and cosmopolitan whom he had encountered on the Malabar coast of India at the close of the literati whose religious proclivities have tended more towards refined and thirteenth century, attributed their superhuman life spans of 150 to 200 cosmeticized orthodoxy or cerebralized tantrism. Indeed, the Nath Sid- years to their ingestion of an elixir composed of mercury and sulfur.30 dhas have long been accused of being charlatans or mere conjurers-an Some five hundred years later, Marco Polo's observations are seconded by accusation that India's street magicians have long used to their advantage, the French traveler Francois Bernier, a Catholic man of letters, when he posing as yogins or tantrikas in their performances.29 It is in this context notes that the Yogis "know how to make gold and to prepare mercury so that we must introduce yet another important term to our lexicon. This is admirably that one or two grains taken every morning restore the body to the term Yogi (Jogi, in vernacular parlance), which has to a certain extent perfect health."31 supplanted the terms "tāntrika," "Kāpālika," and "heretic" in orthodox What links these two accounts in a most startling way is that their Hindu discourse. While yogi[n] is nothing more than an adjectival or pos- descriptions of Yogis both seem to define these figures as alchemists. Yogis sessive form of the term yoga, used to designate a practitioner of yoga, the were healthy, had good digestion, and lived for hundreds of years because term came to take on a sectarian and often pejorative connotation in medi- they ingested mercury and sulfur as part of their daily regime. Here, let eval India, a connotation which has remained operative down to the pres- ent day. "Yogi" or "jogi" has, for at least eight hundred years, been an all- us also recall that the firman Aurangzeb sent to the Nath Siddha abbot of the Jakhbar monastery was a request for treated mercury. These data, set purpose term employed to designate those Saiva religious specialists whom against the backdrop of the vast wealth of yogic literature-as well as a orthodox Hindus have considered suspect, heterodox, and even heretical in their doctrine and practice. On the one hand, the Yogis are defined (like sprinkling of alchemical works-produced by such illustrious Nath Sid- dhas as Gorakhnath and Matsyendranath, can lead to only one conclusion. the tantrikas of an earlier time) by their nonconformity to and exclusion The Siddhas, the Yogis, of medieval India were both alchemists (Rasa Sidd- from orthodox categories: they are that troubling aggregate of sectarian has) and pioneers of hatha yoga (Nath Siddhas). Yoga and alchemy were groups and individuals whose language and behavior subvert the canons of complementary, interpenetrating disciplines for the medieval Siddhas.

IO II

Chapter One Indian Paths to Immortality

The Rasa Siddhas and Nath Siddhas, if they were not one and the same There were two ways in which sacrifice saved one from death. The first people, were at least closely linked in their practice. The balance of this of these was the mechanism of sacrifice itself. According to brahmanic the- book is devoted to proving this thesis. ory, the body one inhabited in life was in fact a loan from the gods or, more precisely, from Yama, the Lord of the Dead. As such, sacrifice was nothing other than a payment on a loan; failure to pay (i.e., offer sacrifice) resulted 3. The Quest for Immortality: The Vedic Legacy in repossession (i.e., death). In order that his debtors might keep up on their payments, Yama, Vedic Hinduism's cosmic "repo man," threw in a The altogether human aspiration to be possessed of a body not subject to piece of land with the body he loaned: this was the parcel on which the the trammels of death finds its earliest Indian expression in the ca. 1200 sacrificer installed his household (garbapatya) fire."7 In this context, the B.C. Rg Veda, in which a poet pleads "Deliver me from death, not from English term "mortgage" (literally "dead pledge") for "the conveyance of nondeath."32 Here, the Vedic term amrta is a polyvalent one, at once signi- real or personal property by a debtor to a creditor as security for a debt"38 fying nondeath (a-mrta), immortality, the immortals (the gods), the world to be repaid within a fixed period of time takes on a new fullness of mean- of the immortals (heaven)-and nectar or ambrosia (which is the Greek ing. Under the terms of Yama's and the gods' contract, no human could cognate of amrta), the draft of immortality, by which the gods remain im- occupy a body for more than one hundred years, since such would have mortal. It is this final gloss that is the most pregnant with meaning for the been tantamount to (divine) immortality. Thus, once again, "a hundred later traditions I treat in these pages. In the Vedic context, the gods win years is tantamount to immortality." and maintain their eternal life by offering soma, the miraculous herb of The second way in which the sacrifice saved humans from death lay in immortality, as a sacrificial oblation among themselves. the nature of the oblation itself. Soma, the divine nectar of immortality, Here, the rich Vedic (and Indo-European) mythology of the theft of was, in the time of the Vedas, considered (or fantasized) to be accessible to soma33-from either the atmospheric Gandharvas or the rival Asuras (anti- humans, whence such hymns as: "We have drunk the Soma; we have be- gods)-is given a particularly sacrificial gloss in the priestly tradition of come immortal; we have gone to the light; we have found the gods .... the brahmanic literature. It is not enough to simply possess the soma-or Far-famed Soma, stretch out our lifespans so that we may live .... The any sacrificial oblation for that matter-to benefit from it. Rather, as the drop that we have drunk has entered our hearts, an immortal inside gods first discovered, it is by offering or surrendering the sacrifice to an- mortals."39 other (god) that its benefits accrue to the sacrificer. The brutish Asuras, Like fire (agni), soma is both a substance and the god identified with that unable to fathom this secret, each offered the oblation into his own mouth fluid oblation .* Early on, however, Soma the god became identified with and so failed to win (the benefits of) the world of the sacrifice.3 the moon (Indu, in Vedic parlance), which was considered to be a drop In the Vedic present, humans who have now learned the secret of sacri- (indu) of nectar (amrta), of soma, shining in the heavens.+1 But the moon, fice come to reap its benefits by offering the sacrificial oblation (idealized this drop of nectar, was nothing other than divine seed (retas),+2 which was as soma regardless of the oblatory material) to the gods. The oblation sus- identified by analogy with vital fluids both animal and vegetable (rasa),*3 as tains the gods and maintains their immortality; moreover, the fruit of the well as with the vivifying rains and waters (ap), which were so many medi- sacrifice that accrues to the human sacrificer also takes the form of a certain cines or remedies (bhesaja) for all that ails mortal man.# order of immortality.35 In addition to fulfilling to the more or less mundane It is the fluid element (rasa), then, that the Vedic theoreticians conceived aspirations of the brahmanic sacrificer-wealth in cows, faithful wives, as the support of all life and indeed of nondeath for humans as well as gods. sons, etc .- the principal fruit of the sacrifice was a mitigated immortality As I have already noted, however, the potential of the fluid oblation could for a "full life span" (visvyus) of one hundred years. Therefore, in order to not be activated or realized without the dynamic of sacrifice, which also live a full life, one had to sacrifice constantly, "for a hundred years is tanta- brought two other elements into play: these were fire (agni), divinized as mount to immortality."36 Agni, the god of fire; and wind (vayu), the active element of exchange,

12 I3 Chapter One Indian Patbs to Immortality

which conveyed the essence of the sacrificial oblation from the world of was the installation of a golden image of a man (hiranya-purusa) beneath a humans to the divine realm. This trinity of elements (and gods) was com- corner of the altar emplacement. With this ritual, the adhvaryu priest in- plemented by another conceptual triad, which served and continues to toned, "He is Prajapati, he is Agni, he is made of gold for gold is light and serve as the ground for the network of homologies and analogies that are fire is light; gold is immortality and fire is immortality. He is a man for the framework of the entire sweep of Indian symbol systems. This is the triad constituted by the human being (microcosm), the mediating mecha- Prajāpati is the Man." 47 Let us note here that the thermal energy that transforms the body of nism of the sacrifice (mesocosm), and the universe as a whole (macro- Prajāpati into gold (and in other myths of this sort, the entire created uni- cosm)-which is often conceived as the body of a universal man or god.45 A passage from the Satapatha Brahmana (9.5.1.11) plays on all of these verse in all its parts) is an inner fire or heat that is kindled through religious austerities. Within a few centuries of the composition of this Brāhmana interrelations: "When he [the sacrificer] has offered in the fire, he drinks [soma]; for that [fire altar] is his divine body, and this [the sacrificer's body] text, a revolution in Indian thought would issue into the notion that hu-

is his [the fire god Agni's] human one." mans too could internalize the sacrifice and thereby entirely bypass the

As I will show, this threefold structure, in combination with the triad of mechanism of external sacrifice. This inward turn, which would ground the entire gnostic and nondualist project of the Upanisads, also sowed the fluid-fire-wind, comes to inform both medical and yogic models of diges- seeds for the innovation of a body of techniques for internal bodily trans- tion, conception, metabolism, and bodily regeneration, as well as alchemi- formation-i.e., for the practice of hatha yoga. Here one's bodily fluids, and cal models of the chemical reactions between the fluid element mercury semen in particular, become identified with the oblation, the heat of inner (called rasa) and the fiery element sulfur, transmutation in the laboratory, austerities with fire, and breath with the dynamic element of wind.48 and the transubstantiation of the human body. It is in the Atharva Veda in particular that we find the most important Already in Vedic traditions we find embryonic notions of this interplay foundations for the later medical and alchemical traditions, which sought between the human, divine, and sacrificial-and mineral-realms. In a to extend (indefinitely) the life span of human beings. Indeed, it is in this hymn of praise to odana, the sacrificial porridge, the Atharva Veda (11.3.1- text that one finds the greatest preponderance of healing hymns involving 2,7-8) states that "of this porridge Brhaspati is the head, Bráhman the the use of charms and herbal remedies to restore the ailing patient to mouth, heaven and earth the ears, sun and moon the eyes, the seven seers health. At the center of this practice stood the healer (bhisaj) who was also the in- and out-breaths ... dark metal its flesh, red metal its blood, tin its ash, gold its complexion."46 a possessed "shaker" (vipra) and an inspired master of incantation (kavi).19

Later, the Satapatha Brāhmana (6.1.3.1-5) puts a mineral twist on one of Part physician, part shaman, part sorcerer, the atharvan priest was viewed as both powerful and dangerous by Vedic society. For this very reason, its many accounts of the creation of the universe through the self-sacrifice perhaps, his heir, the itinerant Ayurvedic physician (cārana-vaidya) was also of the cosmic god-man Prajapati: regarded with suspicion by "good" brahmanic society.50 Verily, Prajapati alone was here in the beginning. He desired "May I That the Hindu medical tradition (Ayurveda, the "science of longevity") exist, may I reproduce myself." He toiled, he heated himself with in- is the self-conscious heir to the Atharvavedic synthesis is clearly evinced in

ner heat. From his exhausted and overheated body the waters flowed the Caraka Samhita, the textual cornerstone of this tradition: "The physi- forth ... from those heated waters foam arose; from the heated foam cian [vaidya] ... should manifest his devotion to the Atharva Veda ... be-

there arose clay; from the heated clay, sand; from the heated sand, cause the Veda of the atharvans has discussed medicine [cikitsā] by way of grit; from the heated grit, rock; from the heated rock, metallic ore; prescribing donations, propitiatory rites, offerings, auspicious rites, obla- and from the smelted ore, gold arose. tions, observance of rules, expiations, fasting, and mantras; and because it indicates that medicine improves the quality of life."51 The Vedic ritual in which the exhausted and decomposed body of the Within the Indian medical science of Ayurveda, the term employed for creative self-sacrificer Prajapati was restored to wholeness was called the the prestigious body of techniques devoted to rejuvenation therapy is ras- agnicayana, the "piling of [the] Fire [altar]," of which an important moment ayana, the "path of rasa," of which an important component consists in the

14 Chapter One

application of herbal remedies, inherited in part from the Atbarva Veda. TWO This same term, rasyana, is also used by the Rasa Siddhas to designate their alchemical "Work in two parts," with its dual emphasis on transmuta- tion and bodily transubstantiation. In this alchemical context, rasa is a term Categories of Indian Thought: for the fluid metal mercury, the mineral hierophany of the vital seed of the The Universe by Numbers phallic god Siva.

1. Microcosm, Macrocosm, and Mesocosm

Of all the conceptual constructs I treat in this book, rhe most pervasive and persistent by far is that which treats of the multivalent relationships or homologies obtaining between the individual and the world, or the micro- cosm and the macrocosm. Spanning the history of ideas the world over, three broad strategies for describing this relationship have predominated. These are the monist (which maintains that creature, creation, and creator are essentially one), the dualist (all is two), and the atomist (all is many)- with a myriad of permutations, qualifications, and recombinations on these three basic organizing principles. Although the sacrificial worldview of the Vedas was a dualistic one, it was one that nonetheless allowed for a breakthrough or transfer to occur -via the sacrifice-between man in the world and the gods in heaven, between the human world order and divine cosmic order (which together formed a whole called rta). As a transfer mechanism or template between the two orders of being, between the human (adbyatman) and the divine (adbidevatā), the sacrifice became possessed of an ontological status of its own. That which pertained to the sacrifice (yajña), to that pivot between the human and divine worlds without which neither could survive, was termed adhiyajna.' This tripartite configuration, undoubtedly the most pervasive structure to be found in the Indian world of ideas, has come to be applied to a myriad of domains, across a wide array of religions, philoso- phies, and scientific disciplines, including those of yoga and alchemy. I term the three members of this configuration-of human + mediating structure + divine-as microcosm, mesocosm, and macrocosm.2 Over time, the mechanism of sacrifice itself came to take precedence over both the humans who enacted it and the gods to whom it was offered, and so we find, in the tenth- through eighth-century B.C. body of reve-

15

16 17 Chapter Two Categories of Indian Thought

lation known as the Brahmanas (Priestly Books), the notion that the sacri- the existence of God that illuminated western savants have proferred over fice (or the ritual of sacrifice) is all that truly matters in the universe. Hu- the centuries. Number and proportion become the very foundation of the mans and gods become tributary to the sacrifice in this period, and the sole good, the true, and the beautiful, in the petals of the lotus, just as they do gods granted any importance are precisely gods of sacrifice. These are (1) in their western reflection, the Secret Rose. Purusa/Prajāpati, the "Man" or "Lord of Creatures" whose primal (self-) Much of this chapter will be devoted precisely to this matter of numeri- sacrifice, which "created" the universe, stands as the model for every sacri- cal progression, from duality to ternarity and thence to pentads, the num- fice that has followed; (2) Agni, "(Sacrificial) Fire," and (3) Soma, the ber sixteen, and the staggering figures given in Hindu reckonings of the "Fluid" god of the sacrificial oblation. In a sense, this brahmanic triad of duration of the cosmic eons (kalpas). Out of all these dilations of number, sacrificial gods is itself a reworking of the triune Vedic universe, with two however, there emerges a single bipolar* dynamic that has played itself out static elements (gods and humans, oblation and fire) being mediated by an in the form of four interrelated temporal cycles that Hindus have em- active third element (the enacted sacrifice). ployed, over the centuries, to situate the microcosmic individual within (or Throughout the history of Indian thought, no set of concrete elements without) the macrocosmic flow of time. A series of charts will illustrate has been as pervasive as this sacrificial triad-of fluid, fire, and air; of rasa, these interrelated cycles. agni, and vayu. Although the three members of this triad have, according A second leitmotif concerns the dynamics of the sacrificial and later sys- to their specific fields of application, taken the form of moon-sun-wind, tems. The aggregate of microcosm-mesocosm-macrocosm would not be semen-blood-breath, or mercury-sulfur-air, they have always borne the an interesting or useful one were an exchange not possible among the three same valences as they did in their original Vedic context. Much of this levels. This is the exchange, nay the transformation, that is effected book will be devoted to describing the ways in which fluid, semen, moon, through sacrifice, that most exalted of human activities, in which men "do and mercury on the one hand, and fire, blood, sun, and sulfur on the what the gods did in the beginning."5 As I have already indicated, the sacri- other-always mediated by the active element of air, wind, and breath- ficial world order was dualistic: there was a sharp break between the human have interacted with one another through the "sacrificial" structure of order and the divine, cosmic order, which only sacrifice could bridge. This microcosm-mesocosm-macrocosm, across a dozen interpenetrating ritual it did as if magically: a pot broken in this world, that is, in the sacrificial and belief systems and some three thousand years of cultural history. context, becomes a whole pot in that world of the gods.6 The metaphysics Two other features of the Vedic synthesis which have persisted through that flowed from this system therefore assumed the building blocks of real- time need also to be mentioned here. The first of these is a fascination ity to be discrete and impermeable. Its dynamic was one of differentiation with number. If, as the Brahmanic sources assert, the sacrifice in all its parts and reintegration. This dualistic approach, which finds early expression in is identical to the universe in all its parts, then it is necessary to enumerate the Rgvedic "Hymn of the Man" (10.90), is restated time and again in later all of those parts, and "cross-list" them with other parts. More than this, texts, sometimes taking on sexual valences (to describe a universe in which the number of parts in a given whole-for example, the 4 X 11 syllables all is ultimately two), such as in a Brhadāranyaka Upanisad myth which de- of the tristubh meter-has a significance which is independent of that ag- picts Prajāpati as splitting into male and female halves to incestuously rein- gregate of parts for which it is the numerical index. Thus, it was not un- tegrate "himself" through all manner of human and animal forms.7 This is common for the priestly commentators on the ritual to wax poetic on the the mythic foundation of Samkhya, literally the "enumerating" philosophy, "eleven-ness" of the number eleven, and so on.3 This Vedic fascination the earliest of the Indian philosophical systems. becomes a veritable obsession in tantrism, in which we witness nothing Out of this dualist system, or perhaps in response to it, there emerged less than an explosion of numbers, categories, and numbers as categories. another current of thought, this a mystic and monistic one which, on the In the tantric case, the hallucinating proliferation of number-based homol- contrary, assumed a continuity of being, extending unbroken from the su- ogies-between microcosm, mesocosm, and macrocosm-appears, in the preme absolute down to the lowest forms of inert matter. Because all being, final analysis, to serve to reassure the tantric practitioner of the efficacy of every being, was emanated from a primal and ultimate source, it thereby his ritual acts-something akin, perhaps, to the many numerical proofs for participated in some way in the very Being of that Absolute. This gno-

18 19 Chapter Two Categories of Indian Thought

seological doctrine, first promulgated in the Aranyakas (the ca. seventh- century B.c. "Forest Books") and the classic Upanisads (the ca. sixth- exists between the subtle and gross body. To the five gross elements that

century B.C. traditions of "Placing in Equivalence") maintained that all are the building blocks of the gross body and the universe correspond the

bodies, but especially all souls (atman), participated in the nature of the five subtle elements (as well as the five senses of grasping and of percep-

absolute or universal soul (brabman). If "atman is brabman in a pot [the tion). More than this, the universal macrocosm and the human microcosm

body]," then one need merely break the pot to fully realize the primordial are essentially composed of a "layering" of the elements. Armed with the

unity of the individual soul with the plenitude of Being that was the Abso- knowledge that the body is nothing more than a series of overlays of the

lute. To know it was to be it.8 This early monism is known by the name of five hierarchized elements, Indian mystic thought innovated a concrete

Vedānta, because it is broached in the Upanisads, that corpus which con- technique for the return of being into essence, for the resorption of the

stitutes the "end (anta) of the Veda." As we will see later in this chapter, it human microcosm into its divine source. By first "imploding" the lower

was likely the concrete experience of yoga that gave rise to this mystical gross elements into the higher, the practitioner could thence implode, via

and monistic vision. All apparent oppositions-between god and man, be- the subtle ether, the gross into the subtle and by degrees telescope the sub-

tween male and female, etc .- here become consumed as it were in the fires tle back into its essential source, the individual soul (atman) which, as the

of yogic austerities (tapas) conceived as the internalization of the sacrifice. Upanisadic gnosis never tired of reiterating, was identical to the universal

The notion of transfer (from one plane to another) becomes metamor- Soul, the absolute brabman. Later, Hindu Tantra would carry this reason-

phosed into one of transformation (one plane primordially and ultimately ing to its logical conclusion, conceptually imploding body, individual soul,

is-the same as-the other), with the human body itself becoming the seat and divine Soul into One:

of the sacrifice and the human soul the indwelling Absolute. These two Ultimately the conscious bits of the universe, like stones, are also dynamic systems, of dualist differentiation and reintegration and monist God and hence consciousness, but a consciousness that has decided emanation and participation, inform, singly or in combination, all of the to conceal itself (atmasamkoca) ... The world of the Tantric, then, is Indian traditions that pass in review in these pages. They are vital to an ultimately all God, but it contains a vast range of things, from things understanding of any and every Indian metaphysical system. as gross as stones to things as subtle as God ... Looking from God If we are to understand the dynamics of these systems, two further no- downwards, we have the range from conscious to unconscious, the tions are absolutely essential here. The first concerns the nature of the range from simple to complex, and the range from subtle to gross. "body" that transmigrates from the corpse of a deceased person to a world These three ranges are co-ordinate; in fact they are different aspects of intermediate afterlife, identified in Upanishadic Hinduism with the of the same thing. Moreover, movement down the scale is precisely moon. This body, termed the "body of enjoyment" (bhoga-śarīra), is an what happens when God creates the universe.10 elaboration on the subtle body (sūksma śarīra), i.e., the "living being" (jīva) that, according to nondualist thought, mediates between the eternal but wholly intangible soul and the gross body (sthūla sarira) composed of the 2. Ternarity and Bipolarity in Veda and Āyurveda five elements: earth, water, air, fire, and ether. As I demonstrate, the subtle body-especially when it is "clothed" in the body of enjoyment-becomes Two practical disciplines that grew out of the Vedic matrix to interact with lunar at certain points in its cyclic existence, filling out and diminishing, in Hindu and Buddhist philosophical and mystic traditions well before the the unending course of births and deaths, around the "core" of the immor- tal soul. Like the moon itself, this transmigrating body is also "fluid"; like beginnings of the tantric age were Ayurveda and the body of physical and meditative techniques known as yoga. Both traditions have their origins in the moon, the subtle body is composed of incremental digits or members the Vedas, both emerge as systems of thought and practice in or around (kalās)° that come into being and pass away, to be renewed yet again. the sixth century B.c., and both continue to share common methods and The second basic notion concerns the five elements to which I referred goals down to the present day. When we go back to the very first use of a moment ago. In the individual, it is via these elements that a correlation term yoga, in the Rg Veda (7.86.8; 10.166.5), we find that it is part of a

20 Chapter Two Categories of Indian Thought

compound: yoga-ksema means "harmonious adjustment."" As we will see, stances there is in the world, that much [diversity] there is in man; however the principal aim of Ayurvedic practice is also adjustment: sama-yoga is the much there is in man, that much there is in the world."20 harmonious conjunction of microcosmic and macrocosmic "climates." Also When applied to the medical tradition proper, these concepts define the growing out of the Vedic matrix is the metaphor of cooking for such trans- physician's craft: the treatment of imbalances between the bodily micro- formative processes as sacrifice, cremation, digestion, aging, and the yogic cosm and the universal macrocosm, i.e., of diseases. Critical to the Ayur- austerities. As in Vedic sacrifice, so in yoga and Ayurveda: the body is to vedic understanding of the body and its metabolic functions is its concep- be "cooked to a turn" (paripakvā).12 tualization of the process of digestion. This is a seven-step process, in The extent to which the yogic and Ayurvedic perspectives interacted which the food one eats is serially "burned" or refined, over the seven "fires with other nascent Indian traditions is incalculable. The organizing prin- of digestion" (dhatvagnis) into the seven bodily constituents (dhātus).21 In ciples of the sixth-century B.C. teachings of the Buddha on suffering and order, these are chyle (rasa),22 blood (rakta), flesh (māmsa), fat (medas), bone its cessation were essentially medical;13 conversely, one finds Buddhist ter- (asthi), marrow (majja), and finally, in men, semen or sperm (śukra, retas).23 minology as well as an inductive method that is clearly Buddhistic14 in the In women, the seventh dhatu is uterine or menstrual blood (sonita, artava), most venerable textual pillar of Ayurveda, the Caraka Sambita (compiled or, after childbirth, breast milk.24 between the third century B.c. and the fourth century A.D.).15 This same Ayurveda further conceives of bodily metabolism in terms of an inter- source contains what is perhaps the earliest complete expression of Sām- action between the dosas, the three "morbid states"25 or humors, of phlegm khya philosophy that has come down to us.16 Towards the end of this same (slesma), bile (pitta), and wind (vata). While we may see in the humors a period, such Upanisads as the Praśna (3.6), and Maitrī (6.22, 30; 7.11.2-5) reapplication of the Vedic rasa-agni-vayu triad, we should note that they were tentatively charting the yogic body and practicing yoga as a means to also anticipate one of India's most important metaphysical constructs, that concretely experiencing the absolute in ways that were deeper than reason of the three "strands" (gunas) of manifest being, the discussion of which is could know.17 It is not our intention here to trace a chain of transmission first broached in the classic Upanisads. According to Sāmkhya, the three of ideas from one tradition to another, but rather to present them as vari- strands-white sattvā, red rajas, and black tamas-remain in a state of equi- ations on a sensus communis that existed in India from a very early time, librium for so long as the universe persists in a nonmanifest state. It is an even before the recorded teachings of the Buddha, the proto-Sāmkhya of unexplained disturbance in their equilibrium that triggers a gradual fall Caraka, and the dualist Samkhya and the monistic Vedanta of the early into manifestation, which is cast as the self-reproduction of the original Upanisads. Here, we follow the French physician and historian of Indian materiality (prakrti).26 medicine Jean Filliozat, when he states that "the Samkhyan concepts, like The Ayurvedic dosas are of the same order as the gunas in the sense that, those of Ayurveda, are part and parcel of the intellectual baggage common for so long as a human being is not exposed to the outside world (when in to all Indian thinkers."18 the womb, for example), it enjoys a perfect balance of dosas.27 When, how- According to Ayurveda, the human being is composed of a triad of psy- ever, it becomes exposed to the outside world, the dosas fall out of balance chic being (sattva), soul (ätman), and body (sarīra), which arises from yet is and the individual becomes subject to health disorders. It is in these cases distinct from purusa, the transcendent self. Ayurveda is thus a dualistic sys- that the Ayurvedic physician must intervene to restore the lost equilibrium tem. Like all living creatures, the human being is possessed of natural char- between the bodily microcosm and the universal macrocosm, a macrocosm acteristics (prakrti), which are sixfold: the five gross elements (mabābhūtas) whose climatic changes are governed by the interactions of moon (rasa), presided over by the atman: "Earth is that which is solid in man, water is sun (agni), and wind (väyu).28 As we will show, the physician's most power- that which is moist, fire is that which heats up, air is breath, ether the empty ful weapon against such humoral imbalances is the sequestering of his pa- spaces, bráhman is the inner soul (atman)"19 As such, the human self is an tient in an edifice identified with that womb within which his dosas had exact replica of the macrocosm: "Indeed, this world is the measure (sam- originally been in perfect balance. mita) of the man. However much diversity of corporeal forms and sub- The identification of the three dosas with the three components of sacri-

22 z3 Chapter Two Categories of Indian Thought

fice is also explicit in Āyurveda. The body's inner fire (antarāgni) is rep- medicines, and clinical techniques appropriate to a specific locus in time resented by the fiery liquid that is bile; phlegm is "lunar" (saumya) fluid, and space (rtu-sātmya), is able to ensure the continuity of the good life. and the microcosmic wind bears the same name as it does in the macro- Ultimately, one may conceive of the physician's practice as one of con- cosm: vāyu (or vāta). Each of these three dosas is further subdivided into fronting time-both the time that ages mortals and saps them of life (time five types. In the case of the windy humor, the five breaths are known as: as death) and the time that takes the form of seasonal or temporal changes präna, the in-breath that enters the body via the nostrils and sustains the which have more immediate effects on the human organism. In Vedic sac- body; udāna, the out-breath that is the vehicle of speech; samana, the breath rifice, "articulating activity" (rtu) is the foundation of a universe that is that moves food through the digestive organs and kindles the fires of diges- well-ordered, regulated, or "articulated" (rta). Thus the Vedic god of sacri- tion; apana, the breath that voids the lower body of excretions; and vyāna, fice Prajāpati, who is also the year, is reordered, rearticulated, and recon- the breath that circulates throughout the body, vehiculating the inner fluids structed through the human activity of sacrifice. and producing bodily locomotion.29 With the last of the Brahmanas, the Gopatha Brahmana, these notions Similarly, there are five "lunar" phlegms and five "fiery" biles, the first become applied to the treatment of disease: "It is at the turning points of which is the fire of digestion (pacaka). In the process of digestion, the between seasons that afflictions arise; therefore, sacrifices are performed at product of each cooking process is of a more refined nature, but also of a the turning points between seasons."32 In Ayurveda, however, the situation lesser volume than that dhatu from which it is produced.30 This is perceived becomes altered if not reversed: time is of two orders, both "objective" and as a long and even dangerous process. Over the period of twenty-eight "subjective," with the physician able to act on the latter alone, i.e., the in- days required for the "raw" food one consumes to be fully transformed ternal evolution of the patient's disease, his imbalance of the three humors. into semen, the human microcosm is subject to a certain number of dan- External time takes the form of the changing of the seasons, which, when gers. These dangers are of two orders. The first of these concerns only they are excessively unbalanced, constitute an excessive conjunction (ati- males directly. This is the sex drive, which inclines men to lose, in a single yoga) or insufficient conjunction (a-yoga) of time (kāla). Kāla, that all- stroke, all that a full month of digestion has provided them: I return to this conquering deity of the Atharva Veda (19.53.1-9), is capable of destroying special problem at the end of this book. The second of these, of a more life in this way; the physician pits himself against Time's excessive or insuf- general order, is the nature of the universal macrocosm or ecocosm, the ficient conjunctions by adjusting the microcosm to the macrocosm: this is many climatic changes of which constitute so many threats to the human called sama-yoga.33 microcosm and so many challenges to the Ayurvedic physician. In Āyurveda, the excessive manifestations of time that the physician The vaidya is armed with two powerful weapons for battling the hu- must most often combat are the three "extreme" seasons of the Indian year, moral imbalances that so threaten the harmonious functioning of the di- seasons that correspond to the three dosas. To winter correspond accumula- gestive and metabolic processes. First, he may calm (samana) the overac- tions and disturbances in the phlegm; the hot season is identified with ex- cumulation of a given dosa internally, through dietary regimen and the use cesses of wind; and the rainy season with excessive bile.34 More important of pharmaceutical preparations, the scientific development of which owes for our concerns than this tripartite division, however, is the bipolar struc- much to the alchemical tradition in India. Should this fail, he may take ture underlying it. In models that were being developed simultaneously in recourse to external purifications (sodhana) through clinical therapy which Upanishadic and Ayurvedic circles, the round of the seasons and the cycles in an unexpected way constitutes an adaptation of a number of yogic prin- of life and death were ultimately reducible to a single dynamic: this was an ciples and techniques. We now turn to the conceptual foundations that ongoing tug-of-war, between sun and moon, in which the prize was mois- undergird these two Ayurvedic techniques. ture, in the especial form of vital fluids. So the year was divided into two In a groundbreaking study written twenty years ago,31 Francis Zimmer- semesters: the fiery (agneya), in which a blazing sun, which rose higher in mann set forth the Ayurvedic principles for adjusting the human micro- the sky with every passing day, drained (adana, "captation") the fluid life cosm to the universal macroclimate or ecocosm. Here, he outlined the principle out of all living creatures, and the lunar (saumya), during which ways in which the Ayurvedic physician, by employing dietary regimens, the moon, relatively higher in the sky than the lowering sun, poured (vi-

24 25 Chapter Two Categories of Indian Thought

sarga, "release") more moisture into the world (through the rains of which The importance of this myth, and of the conjunction of the disease of it is the presumed source) than the sun was able to draw out.35 The first of these two semesters roughly corresponds to the period be- royal consumption and the height of the hot season, lies in the dire effects

tween the winter and summer solstices, during which the sun's angle of of sun and fire (agni) on moon and fluid (rasa). Here the parallel, between

elevation increases and the point of sunrise moves northward along the the fiery (agneya) and lunar (saumya) semesters of the year on the one hand,

horizon with each passing day; the second approximates the period be- and the waning and waxing fortnights of the lunar month on the other, is

tween the summer and winter solstices, when the sun's angle of elevation transparently evident. In both cases, that half of a temporal cycle character-

decreases and its apparent movement along the horizon follows a southerly ized by dessication (of the ecosystem, of the moon's rasa) is associated with

course. In India, these are known as the northern course (uttarāyana) and the heat of the sun and death. Royal consumption is most likely to occur

southern course (daksināyana), respectively: their "turning points" (sam- at the end of the solar semester, just as the disappearance of King Moon

kranti) fall on 14 January and (approximately) 14 July, respectively.36 The occurs at the end of a dark fortnight during which he has dissipated himself

northern course is, throughout much of the Indian subcontinent, a long by exposing himself to the draining heat of the sexual embrace of the starry

period of increasing dessication, culminating in the blazing dry heat of woman Rohinī. King Moon's loss of rasa is manifested in the latter half of

the hot season; the southern course begins with the torrential rains of the his monthly cycle, by the waning of the moon, by its diminution, by one

summer monsoon and ends with the gentler winter monsoon. The latter digit (kala) on each succeeding night. And, at the end of its dark fortnight,

is a life-giving season in which creatures thrive; the former-in which the moon, completely dissipated, disappears.41

wind and heat are combined-can prove deadly to all. The latter is a time The moon can and does, however, return to wholeness, over the fifteen

of life, even immortality; the former a time of death.37 days of the waxing fortnight; and versions of the myth of King Moon found

It is precisely with the culmination of the northern course that the "on- in Ayurvedic sources demonstrate his rehabilitation (as well as that of hu-

tological disease" of rājayaksma, "royal consumption," is said to occur in mans smitten with the same ailment) through elixir therapy, rasāyana. In-

humans. According to both medical38 and literary3° convention, the king deed, one of the earliest medical references we have to the internal use of

who allows himself to become debauched in the clutches of too many pas- mercury-also called rasa, the vital fluid of the god Siva-prescribes it as

sionate women also falls prey to "royal consumption": as a result, his king- a treatment for increasing the production of male semen (sukravrddhi). The

dom, sapped of all its rasa, withers and dies. In Hindu mythology, the pro- duration of the treatment is most significant: "like the moon (śasanka), the bodily dhātus are replenished over fifteen days."42 totypical king to suffer from royal consumption is the moon itself, the This connection, between the gradual replenishment, even rejuvena- same moon that is responsible for revivifying a dessicated world at the end of each hot season, the same moon whose substance, whose fluid rasa, has tion, of a dissipated moon and that of dissipated human bodies takes us to

been identified, since at least the time of the Taittirīya Sambitā (2.3.5.2) the heart of the two crowning disciplines of the Indian medical tradition.

with semen. In his mythic loss and recovery of his rasa, Candra, King These are rasāyana, elixir or rejuvenation therapy; and vājīkaraņa, sexual

Moon, is married to the twenty-seven (or twenty-eight) daughters of rehabilitation therapy. Both of these branches of Ayurveda assume that

Daksa, who are the stars that make up the naksatras, the lunar mansions youthful vigor is primarily a matter of good digestion which, when overly

through which the moon passes in its waxing and waning phases. On each troubled by "excessive manifestations of time," must be restored through

night, he dallies and makes love to his wives, but it is in the embrace of his more radical treatments than special dietary regimens or purification

favorite, Rohinī, that he passes the most time. It is here that Candra, at techniques. As we have noted, the end product of digestion is, in males,

that point at which he is "closest to the sun," spends himself completely in semen-semen that is homologized with the rasa of the moon, soma,

the clutches of his starry wife, and the moon disappears. His rasa, his vigor, the nectar of immortality. As such, semen is called saumya, lunar, like the

his semen completely dried up, the moon must perform a soma (which is semester characterized by an outpouring of vitalizing moisture into the

both a name for and the stuff of the moon, the rasa par excellence) sacrifice ecosystem. As such, microcosmic semen is subject to many of the same

in order to recover his lost rasa, and so the cycle begins anew.40 dangers as the macrocosmic moon. On the one hand, it takes either a lunar fortnight (as indicated above) or a lunar month (the same time it takes a

26 17

Chapter Two Categories of Indian Thought

woman's body to produce an ovum) for the food males ingest to become On successive days in the first week, he vomits, purges, and emits fully transformed into semen." On the other, it requires a prodigious fluids ... His hair, nails, teeth, and skin fall off and are soon regener- quantity of food to produce a single gram of semen.# These perceived ated. But before the regeneration begins, he is a ghastly sight ... dangers to the very survival of the male sexual fluid are compounded by an Very soon the skin of the man begins to grow normal ... For some Ayurvedic identification of female uterine blood with the fiery (agneya) sun time, the person thus has to be nursed through all the growing stages that drains the ecocosm of all its vitalizing moisture in the first semester of of a new-born babe, oil smearing, bathing, feeding, and putting to the year:45 the "lunar" semen a male is capable of producing is but a drop sleep in [a] soft bed. Before the close of the first month, the person in the fiery maelstrom of his partner's sexual fluid. begins to develop a new set of teeth ... After some more time ... In this context, the venerable Ayurvedic treatments for the replen- his old hair is removed and then he develops a luxuriant growth of ishment of vital male fluids have come of late to enjoy a renaissance all jet black hair. Later still ... exposure is practised as in the case of a across India. Ayurvedic gupta rog (the "secret" or "hidden affliction") clin- child. [He is] carefully and gradually [taken] from the innermost ics, which specialize in the treatment of "sexual disorders" whose prime room of the kuti into the second enclosure, then into the third, and symptom is the birth of daughters, have been mushrooming across India finally into the open sun and air, all in the course of about thirty days. for the past several decades. The techniques these modern clinics employ, After nearly three months or more ... he may be in a fit condition however, and the medicines they dispense are essentially the same as those to go about in the world.+7 employed over the past two millennia for such ailing kings (royal patrons of the medical and alchemical authors) as King Moon. In the classic Ay- This Ayurvedic technique bears striking parallels to sacrificial dīksā, in urvedic sources, rasayana, the seventh branch of Ayurveda, is a holistic which the initiate is sequestered in and reborn out of a closed initiation approach to increasing bodily longevity through the use of plant- and hut. It also reminds us of the common perspective of early Āyurveda and mineral-based elixirs in combination with clinical therapeutic techniques. Sāmkhya, in which a fall into manifestation, triggered by a disturbance in Intimately associated with rasāyana is the eighth and final branch of Indian the equilibrium of the dosas or gunas, is reversed through a return to the medical science. This is vājīkarana, treatments for increasing male vigor "womb" of nonmanifestation. Finally, it appears to work from the same and virility,46 which the fathers of Ayurveda quite sensibly placed after ra- principles and assumptions as do a number of yogic techniques for the syana, reasoning that a long and healthy life was a necessary precondition generation of inner heat (tapas) and the production of a new transcendent to a long and happy sex life. This near identification of virility with longev- self.48 ity is a fundamental one in India, linking together the bodily processes of It is here, at the level of the replenishment and maintenance of vital digestion, semen production and retention, conception, and reproduction. fluids, and most particularly the vital fluid that is semen, that the disci- Here, the most elaborate and prestigious body of rasayana therapy once plines of Äyurveda and hatha yoga intersect: the same semen that the physi- again refers back to the myth of King Moon. This is the treatment known cian identifies with male virility and vitality is the sine qua non of yogic as kutīpravesa, "entering into the hut," in which the patient is sequestered practice: semen is the raw material and fuel of every psychochemical trans- within the triply enclosed (trigarbha) innermost chamber of a hut called formation the yogin, alchemist, or tantric practitioner undergoes, transfor- the "womb of the womb" (garbha-garbham). Any resemblance to the female mations through which a new, superhuman and immortal body is "con- reproductive system is altogether intentional here, with the rebirth of the ceived" out of the husk of the mortal, conditioned, biological body. In the patient being portrayed quite literally: royal consumption myth, we saw that King Moon lost his rasa, his semen, through sexual contact with fiery women close to the sun. In the hatha- The soma plant brought into the kuti ... [is] cut with a gold imple- yogic tradition, some of whose descriptions of the subtle body may be ment, and [its] milk collected in golden cup. With one dose of soma traced as far back as the sixth-century B.C. Chandogya Upanisad (8.6.1-2, 6), milk in the kuti, a person passes through several severe states of we find a similar homology, one that becomes a commonplace in later yo- cleansing in about ten days, and then begins to grow as quickly again. gic traditions. The yogin's lower abdomen (the solar plexus) is the place of

18 29 Chapter Two Categories of Indian Thought

the female, sanguineous sun, which provides the heat necessary to trig- religious calendar of the hot season in north India, which is more or less gering the yogic process, but which can also, like Time (Maitri 6.15), framed by the vernal equinox and the summer solstice. In rural Madhya wholly consume the body, causing aging, disease, and death. The head, and Pradesh, nearly all of the major religious festivals of this period are con- more specifically the cranial vault, is the locus of the sun's counterpart, the cerned with ritually "cooling down" goddesses who are considered to be cooling moon, a moon whose rasa is nothing other than semen that has malevolent, dangerous, bloodthirsty causes of disease. "Hot" diseases, with been carried upwards by the yogic process and so been transmuted into their attendant fevers arise, for example, from the anger of the smallpox nectar, amrta, which is equivalent to soma, the draft of immortality.49 goddess Sitalā (whose name means "cooling," a reference to her ideal na- We should not conclude from these observations that an ideal world ture, which manifests itself when she has been appeased through ritual), would be one in which there were no sun, fire, women, or uterine blood at who is identified or associated with the "Seven Sisters," seven varieties of all. Sacrifice, the medical year, and the rhythms of yoga are all bipolar sys- smallpox. The heat of these goddesses' fury is ritually distributed into tems in which two opposed principles interact, constructively and destruc- wheat seedlings which, after eight days of sprouting, are carried in pots on tively, after the fashion of the up-and-down motion of a firing rod, piston, women's heads before being immersed in water; it is also transformed into or camshaft, to produce a cycle characterized by alternations between fluid types of possession ("which comes on like a fever"), whose "seizures" are and fire, dry and rainy seasons, male and female offspring, and a death that calmed by the imposition of cooling nīm (margosa) leaves. These "hot" leads to immortality. Indeed, the genius of each of these three systems- goddesses are further appeased through songs of praise and the offering of the sacrificial, Ayurvedic, and yogic-lies in the particular way with which blood sacrifices.51 each comes to terms with this bipolar fact of life. Whereas the Vedic sacri- All of these treatments-elite and popular, ritual, medical, yogic, erotic, ficer sought to regulate macrocosmic time as a means to ensuring cosmic and mythological-are ultimately grounded in a body of metaphysical as- and social order and the Ayurvedic physician strives to adjust microcosmic sumptions that date back, in some cases, to the time of the classical Upani- time to excesses in macrocosmic time in order to ensure bodily health, the sads. Since at least the beginning of the common era, there have existed yogin claims to be capable of imploding these two temporal orders into two more or less parallel models for depicting the bipolar interaction of one another as a means to transcending both and freeing himself from time fluid and fire in the microcosm and macrocosm. The first of these is the (and every other natural and cultural constraint) altogether. Ayurvedic model of the seasons in their interactions with the dosas. The In this, the hathayogic synthesis appears to subscribe to the same basic second is a yogic model, which depicts yogic withdrawal from and return working principle as the erotico-mystic practices of Hindu tantrism, i.e., to mundane consciousness in terms of the interaction between the sun lo- that one must, if he is to transcend the human condition, work through, cated in the lower abdomen and the moon located in the cranial vault of rather than against, the overwhelming energy of the feminine. In Tantra, the subtle body. This model, which respects the Ayurvedic seasonal para- sexual intercourse, the abandonment of male semen (the sacrificial offer- digms of solar captation and lunar release of vital fluids, can be effected ing) into the fiery maw of the female sexual organ, is identified as a sacri- only through the "regressive practice" (ulata sadhana) that is the hallmark fice, the benefits of which accrue to the sacrificer. During the act, the male of hathayogic practice. I return to this model later in this chapter: suffice practitioner will therefore recite, "Om, thou the Goddess, resplendent by it to say here that it literally stands all other models "on their heads." There the oblation of dharma and non-dharma, into the fire of the self, using the is also a third model which, dating from the earliest Upanisads, is chrono- mind as sacrificial ladle, along the path of the susumna, I who am engaging logically prior to those of both the Ayurvedic and yogic syntheses. This in harnessing the sense organs, constantly offer this oblation."50 model is based not on the dynamics of heat convection and fluid transfer, Techniques for palliating or channeling the effects of this bipolar dy- but rather on a photic opposition between darkness and light. This is, in namic are by no means restricted to these elite traditions. In fact, the same fact, the earliest systematic statement of the doctrine of karma and re- sorts of principles and techniques also ground the cycles of the ceremonial birth-or, alternatively, liberation from rebirth-to be found in all of In- year in modern-day popular Hinduism. This is particularly evident in the dian literature. Here, in describing the fate of the soul after death, these

30 31 Chapter Two Categories of Indian Thought

sun at summer solstice full moon sun at summer solstice new moon

N N July 14 July 14 rājayakşma

immortality death-laden existence uttarāyana daksiņāyana devayāna pitryāna ādāna saumya waxing moon waning moon āgneya visarga lowering sun lowering sun

S optimal health

January 14 January 14

sun at winter solstice new moon sun at winter solstice full moon

Figure 2.1. Upanisadie model of the year Figure 2.2. Ayurvedic model of the year

sources state that those for whom liberation is promised go into the fire of On the lunar orb, [the dead, in compensation for their accumulated the funeral pyre and thence into the day, the bright lunar fortnight, and merits] obtain a body of aqueous nature for the enjoyment of plea- the sun (along the path of the gods, the devayana). Those, however, who sures. The liquid elements employed in the funerary rites combine must suffer rebirth go into the smoke of the funeral pyre and thence into with other elements, and reach the heavenly regions. There, they the night, the dark lunar fortnight, and the moon (along the path of the amalgamate with the structure of the moon and become the [new] ancestral fathers, the pitryana), where, after they have been "eaten" by the bodies of those who have performed sacrifices, etc. [during their gods, are "rained" down to earth again, and become part of the food cycle earthly existence]. When the final oblation [is] made into the funer- and thereby the cycle of rebirth.52 ary fire, and when the body [is] entirely consumed, the humors that In this last case, the notion arises that some portion of the transmigrat- arise from it fuse with the rising smoke and envelop the [subtle body ing human body is basically fluid and lunar in nature. In his commentary of the] sacrificer. Upon reaching the moon, they form the primary on this passage of the Chandogya Upanisad, Sankara, the great eight-century matter of [his] new body, in the same way as straw and clay [serve to A.D. synthesizer of nondualist Advaita Vedanta philosophy, coherently de- construct the ephemeral "statue" of a divinity].53 lineates for the first time the relationship between the soul, the subtle body, When this body of enjoyment has exhausted the merits that permitted the body of enjoyment, and the moon. it to enjoy its temporary lunar sojourn, it "melts" to once again become a

32 33

Chapter Two Categories of Indian Thought

vehicle for the subtle body. Returning to its prior state of an amorphous This deconstruction of a perceived unity called "self" was a central con- vapor, it fuses with the clouds, falls to earth with the rains, enters into the cern of Buddhism, for which the reality of this world was nothing more or sap of plants and thence into the blood and seed of the animals and humans less than an ever-changing configuration of five "heaps" or "aggregates" who eat those plants. Here, it enters, as it has done an infinite number of (khandhas). Rather than there being bodies inhabited by individual souls times before, into an embryo formed of the mixture of blood and seed thirsting for reintegration into a universal soul, as the Upanisads taught, produced in sexual intercourse to once again become a living creature.54 the Buddha showed existence to be nothing other than a series of evanes- The "fluid" or "thermodynamic" and lunisolar Ayurvedic model of the cent recombinations of appearances, sensations, conceptions, mental for- year together with its mirror image, the "photic" but nonetheless lunisolar mations, and consciousness. The extinction (nirvana) of suffering con- Upanisadic model are presented in chart form in figures 2.1 and 2.2. I pres- sisted in dissociating the five aggregates from any notion of "self," for this ent the yogic model later in this chapter.55 was what trapped one in existence. Ultimately, however, the Buddha was working from the same basic problematic as were proto-Samkhyan and proto-Vedantic thought in the 3. Physical and Metaphysical Factors of Five early Upanisads. Buddhism denied the reality of self in order to emphasize an ethical attitude towards an impermanent world characterized by suffer- Like the fingers on the hand, the number five in India has long been "good ing born of ignorance. Samkhya affirmed the existence of a plurality of to think with." Even if its multiple pentads are expansions on the primal selves and the reality of the world for so long as spirit (purusa) remained sacrificial triad-of fire (agni), oblation (rasa), and the wind (vāyu) that confused with nature or original materiality (prakrti). Vedānta asserted the conveyed the smoke and aroma of the offering to the gods-the Hindu identity of the individual soul (atman, jīvatman) with the universal soul cosmos has been, for at least three thousand years, a fivefold one.56 It is the (brahman, paramatman) while denying the reality of the phenomenal Brähmanas that, even as they continue the Vedic discourse of the triune world.59 Yet underlying these divergent philosophies, there remained a universe, first elaborate the concept of the universe as fivefold. This they common ground that no subtlety of argumentation could efface. This was do most especially in their speculations concerning the piling of the great the concrete experience of the human body-and, as we shall see, most fire altar (agnicayana), which is composed of five layers of bricks. These five particularly the very concrete yogic experience of the body-in its relation layers, the Satapatha Brahmana tells us, are the five bodily constituents of to the external world.60 Thus, despite the important points of divergence the god Prajapati, as well as the five seasons, the five directions, etc.57 The between the metaphysics of these three schools, the identity of microcosm brahmanic identification of this sacrificial god with the year and with food and macrocosm-already evoked in Brahmanic speculations on the identi- would in turn give rise, in the early Taittirīya Upanisad (2.1), to a hierarchi- fication of the fire altar with the body of the cosmic man (Purusa-Prajapati) cal representation of the five elements in their relationship to the universal and the universe-was constantly undermining the philosophical distinc- and microcosmic man: "From this ätman verily ether arose; from ether air; tions drawn between them. In the end, "wet" experience would win out from air fire; from fire, water; from water, earth; from earth, herbs; from over "dry" philosophical speculation.61 herbs, food; from food, semen; from semen, Man." In the early Upanisads, the cosmogonic metaphysics that was generated This notion of the physical universe as an aggregate of the five elements to link the individual to the universe and its absolute source was nothing is one that informs, as we will see, not only the three great metaphysical other than a reverse account of the stages by which the yogin withdraws systems of ancient and classical India (Samkhya, Vedanta, and the Buddhist from the external world to realize the absolute within.62 For the monistic Dhamma), but also permeates commonsense discourse on the nature of Vedantins, this was a process of resorption that was implied in the primal life and death. The perennial expression pañcatvam gamana, "going to the cosmic emanation: yoga was a remounting of those stages through which fiveness [the five elements]," is particularly disclosive of this worldview. absolute poured itself out to form our manifold, manifest universe. This Indians have employed this term since at least the time of the epics as a universe was a continuum, a single reverberation, out of the primal essence euphemism for death, the dissolution of bodily integrity.58 that was brahman, down into subtle (suksma) and gross (sthula) forms of life

34 35 Chapter Two Categories of Indian Thought and matter. All was interconnected, both structurally and materially; and prised twenty-five metaphysical categories, five hierarchical series of five. because all being contained a trace of the absolute brabman (in the form of the individual soul, the atman) within, all was potentially, and thereby This system was taken as given by Vedantic thought in its description of emanated (as opposed to differentiated) reality and, as I now show, by tan- virtually, one with the universal essence. Yogic practice, meditation and tric Buddhism and Hinduism in their maps of a theistic universe, in which insight (jñäna) were the means to realizing, in the gnoseological sense of the impersonal Purusa or brábman was replaced by personalized Buddhas, the word, this inner potential. or by forms of Siva and the Goddess. Although monist, Vedanta constructed its model of an emanated uni- In the Hindu case, the metaphysical categories or cosmic principles (tat- verse upon a Samkhyan prototype, which was dualist. According to Sām- tvas) of Sämkhya were deified, early on, in the form of the twenty-five faces khya, the manifold universe in all its parts was essentially a mistake. That of Mahāsadāsiva.66 Because its was an "exploded" metaphysics that denied is, Spirit (Purusa), deluded (for reasons that are never clearly explained) any primal essence, absolute or universal, Buddhism resisted the Samkhyan into identifying itself with a rather scatterbrained Mother Nature (Prakrti), model for several centuries; but here too, yogic experience eventually pre- moves her to undergo internal modifications that serially generate lower vailed. Already in Asvaghosa's (A.D. 80) theory of "suchness" (tatbata), the and lower strata of being. Rather than a process of emanation, this is one unbridgeable gap that the Buddha taught between existence (samsāra) and of differentiation: in spite of its temporary loss of discrimination, Purusa its cessation (nirvana) was beginning to yield to the irresistible force of is wholly spirit and therefore not of the same stuff as Prakrti or the lower differentiates with which he identifies himself. Purusa, spirit, and Prakrti, yogic experience. A few centuries later, it would collapse completely with the Mahayana notion that the Dharmakaya-the "Buddha body" com- the world, are proximate, though never in actual contact. Thus the various levels of metaphysical reality remain disconnected materially, with their posed of the body of the Buddha's teachings-was an absolute or universal soul, a Buddhist equivalent of the Vedantin's brahman, with which the prac- structural sequentiality being the result of Purusa's "fall" into (identifying titioner entered into mystic union.67 Once the inviolate gap between sam- itself with these ever-descending mutations of) Prakrti.63 sāra and nirvāna had been breached, the familiar corresponding hierarchies In spite of the fundamental differences between them, the basic model of the Indian cosmos came rushing in through the back door, as it were. employed by these two metaphysical systems is one and the same. The Thus, while logicians like Santideva and Dinnāga were devising hair- shape of reality is composed of five sets of five categories, headed by the "preternatural" pentad of purusa, prakrti, buddbi (intellect), abamkāra (ego), splitting arguments by which to interpret the world as a void entity,68 Bud- dhist täntrikas were deifying and hypostasizing the Buddha into five and manas (mind). Below this pentad are four parallel pentads, which inter- emanated Buddhas or Tathāgatas: Amitabhā, Vairocana, Amoghasiddhi, act with and interpenetrate one another. These are the five buddhīndriyas Ratnasambhava, and Aksobhya.69 These five primal Buddhas were subse- (sense-capacities: hearing, feeling, seeing, tasting, smelling); the five kar- quently equated with the five elements,70 the basic concept being that cos- mendriyas (action-capacities: speaking, grasping, walking, excreting, gen- mic expansion, the multiplication of the absolute into fundamental forces, erating), the five tanmatras (subtle elements: sound, touch, form, taste, could be represented by lineages of gods just as easily as by metaphysical smell) and the five mabābhūtas (gross elements: ether, air, fire, water, categories. Thus, each of the five primal Buddhas presided over five lin- earth).64 It is at the level of the twenty lower tattvas that another constant eages of five bodhisattvas which, added to the transcendent Dharmakāya, feature of the Indian worldview emerges. This is the pattern of corre- generated a total of twenty-five divine beings, the same as the number of sponding hierarchies that proliferates throughout a wide variety of differ- Samkhyan categories. Thus we read in the Jnanasiddhi: "Since they have ent yet parallel disciplines. In the case of the Samkhyan categories, there is the nature of the five Buddhas, the five constituents of the human personal- a point-for-point correspondence between the members of each level of ity are called jīnas (conquerers): and the five dhātus (elements) correspond each of the four parallel groups. Thus, the buddhindriya of hearing corre- to the Buddha's saktis ... Therefore our body is a Buddha body."71 sponds to the karmendriya of speaking, the tanmatra of sound, and the ma- This complete turnabout, effected before the seventh century, gave rise habbūta of ether, and so on, down to the correspondence between olfactory to Buddhist tantra. Hindu tantrism as well came to generate clan or family sensing, procreation, odor, and earth.65 The Samkhyan system thus com- lineages (kulas) of divinities that were so many deifications of the Sam-

36 37 Chapter Two Categories of Indian Thought

khyan categories, adding, as had the Buddhists, transcendent categories the fifteen nights are also taken to be the spokes of a wheel, with its felly, (totaling as many as thirty-seven in the Trika Kaula synthesis) to under- the sixteenth, identified with the atman. This transcendent sixteenth unit, score the superiority of its pantheon of supreme deities. In the Hindu case, the amrta-kala, "fills out," renders perfect, the fifteen units of the mundane the supreme god Siva and his consort Sakti (or the Goddess by one of her lunar fortnight.74 Let us dwell for a moment upon the Hindu concept of many other names) are transparent appropriations, but now with a sexual the kala, the lunar digit, and most especially on the sixteenth, immortal reinterpretation, of the Samkhyan pair of Purusa and Prakrti. digit (amrta-kalā) of the moon as it applies to the yogic body. These tantric reworkings of the Samkhyan categories would have a myr- In Ayurvedic parlance, the term kalā is used as a synonym for dhātu75 or iad of practical applications in the fields of hatha yoga and alchemy, both of for the invisible supports (dharas) of or divisions between each of the seven which stressed the importance of the five elements. In the former system, dhatus.76 This usage probably originates from the original concept of sac- the five lower cakras were identified with the five elements, while in al- rificial initiation (diksā) in the Vedic tradition. Here, when the sacrificer's chemy, mercury, the semen of Siva, was viewed as incorporating the five body is made anew, the embryo of his sacrificial body is symbolically gen- elements in itself. I discuss these applications in detail in later chapters. erated by adding layer upon layer of bodily constituent: semen, blood, caul, placenta, etc.77 Indeed, the original sacrificial body-that of Purusa- Prajāpati who is reconstructed time and again in brahmanic speculation- 4. The Lunar Cipher of Sixteen in Veda, Yoga, and Tantra is said to be divided into sixteen kalās, which are each of his eight dhātus, taken twice.78 An Upanisadic identification of Prajapati's body with the In the royal consumption origin myth related above, the moon was revived moon would appear to reinforce this interpretation. Indeed, sixteen is also and replenished in its bright fortnight through the offering of a soma sacri- a key figure in Indian reckonings of a woman's menstrual cycle. In the fice. Soma is the fluid essence of the moon, which, in the sacrificial context, words of the Yajnavalkya Smrti, "sixteen nights are the season of a woman must be bought. With what does King Moon buy back his vital fluids? ... from the appearance of menses, sixteen nights is for a woman the sea- With a red cow, whose name, robini, is the same as that of the starry woman son, i.e. the time for the conception of the fetus." As with the moon, which who was the original cause of his woes.72 The Brahmanas offer another fills out through an addition of layers (digits), so with the (re)constitution explanation for the moon's ability to wax anew, indeed, for its immortality. of sacrificial and embryonic bodies, as well as the ovum.79 Elsewhere, in Of the moon's sixteen digits, the last is said to descend to earth to dwell in Saiva Siddhänta ritual, the body parts of God, which the devotee superim- animals and plants: "Now this King Soma, the food of the gods, is none poses upon his own body through the recitation of mantras, are called kalās.80 other than the moon. When he is not seen that night either in the east or A similar perspective appears to underlie a popular ritual observance west, then he visits this world; and here he enters into the waters and the found throughout modern north India: the "sixteen Mondays vow," during plants. And since during that night he dwells (vasati) [here] at home (amā), which one fasts and offers coarse wheat-flour cakes to Siva on sixteen suc- that is called the new moon (amāvāsya)." "Then, on the night of the new cessive Mondays. As in English, Monday means "moon-day" (soma-vāra) moon, entering by means of its sixteenth digit (kala) into all that is pos- in the modern Indo-Aryan languages; this ritual observance may be inter- sessed of breath, it [the moon] is reborn in the morning."73 preted as a practice by which to fill out the moon-through the intermedi- What is this sixteenth, immortal digit of the moon? Clearly, it is to be ary of Siva, the god whose semen is identified with soma-in successive identified with the immortal soul, which persists through an infinite series phases, until it is sixteen kalās full.81 of rebirths and redeaths. Why the number sixteen? In all likelihood, the The original use of the term kala is found in the Rg Veda, where it was intimate association of this figure with the moon is the result of the Indian used to signify "any single portion of a whole, especially a sixteenth part."82 penchant for "rounding up" from a given set of aggregates to express the It was somewhat later that the term came to be applied to each of the six- notion of perfect wholeness. In this case, the fifteen observable nights of a teen digits of the moon, but by that time its meaning had already expanded lunar fortnight are taken as fifteen separate units, to which is added an to encompass a wide array of other sets of sixteen.83 The standard number additional unit (the whole lunar fortnight as the sum of its parts). Here, of priestly specialists (excluding the yajamana) in a brahmanic soma sacrifice

38 39 Chapter Two Categories of Indian Thought

was sixteen;84 already in the Brahmanas, the number sixteen was identified moon in the cranial vault be replenished to its fullness and shine with all as the numerical attribute of brabman, the absolute.85 When the identifi- of its sixteen digits. It is, however, impossible for these sixteen digits to be cation of the universal macrocosm, ruled by bráhman, and the corporeal regenerated without the "seed" that is the immortal digit, also known, in microcosm, ruled by atman, was made, the sixteen attributes of brabman yogic parlance, as the nivrtti kala, the digit of yogic introversion.90 translated into the aggregate of the five breaths, the five organs of action The subtle body is in fact composed of two sets of sixteen digits, the (karmendrīyas), the five organs of sensation (jnanendrīyas) and the mind,86 one solar and fiery and the other lunar and fluid. In the body of a nonprac- or into the sixteen components of the living being of which only one- titioner, it is the former set that holds sway: the sun or fire of time in the sixteenth is unmanifest (avyakta) and transmigrates.87 lower abdomen burns with all of its rwelve digits91 and thus consumes, ages Nowhere, however, does the notion of wholeness and perfection trans- the body. In addition to these twelve solar digits, there are also said to be late more fully, on both microcosmic and macrocosmic levels, than in the dynamics of the waxing and waning of the moon, in the increase and dimi- four fiery digits in the body.92 These are located in the medial channel of the subtle body, the susumņā nādī or brahmamārga (the "path of brabman"), nution of lunar digits, the kalas. The moon fills out with each lunar month to become brimful of this nectar, which it pours into the world in the form which runs up the spinal column from the lower abdomen to the cranial

of vivifying rain, the fluid source of every creature's vitality.88 In this con- vault. Yoga, and most especially hatha yoga, involves forcefully controlling,

text the invisible sixteenth kala, that digit that makes the moon fully whole, even reversing the body's natural tendencies through the combined tech-

takes on its fullness of meaning as the amrta-kala, the "digit of immor- niques of breath control, fixed postures, and meditation. What these tech- niques aim to reverse is the aging process, which yogic traditions identify tality." with the predominance in the body of the solar and fiery digits. This yogic The increase and diminution of the moon, both in its lunar fortnights battery of techniques for reversing the course of nature and time is vari- and in its relationship to the sun in the system of the solar and lunar semes- ously called ulatā sādhana (regressive practice), kāya kalpa (bodily reinte- ters of the year, is intimately tied into yogic theory and practice relative to the subtle body. This is a bipolar body that is divided into two halves at the gration), and parāvrtti (retroversion). Yogic practice reduces the influence

level of the navel. Of these, the lower half is associated with femininity, of the sixteen digits of the lower body while simultaneously increasing that

with male semen that is "prey" to blood and to the fire of the sun; and of the sixteen lunar digits of the moon located in the cranial vault. This it does, in practical terms, by raising the yogin's semen from his lower ab- the upper with masculinity, with semen that has been refined into nectar domen along the length of the medial channel until it fills out the moon in identified with the moon. The lower half is further identified with mun- his head. As it rises, this same semen is gradually transformed into amrta, dane existence, with the dispersal of the individual's life into a myriad of the stuff of the macrocosmic moon, the divine nectar of immortality which worldly concerns paralleled by the dispersal, in his lower body, of his seed: pours itself into the world in the form of vivifying rain. this is pravrtti, extroversion. The upper half, on the contrary, is identified The way in which this transformation occurs lies at the heart of hatha- with supermundane consciousness, with the reintegration of all that is nor- yogic theory and practice. Hatha yoga is the forceful channeling and control mally dispersed in mundane existence: this is nivrtti, introversion. of the vital breaths (pranas) and of the thermal energy (tapas, yogāgni) of In this system, human life itself is seen as dependent upon the main- the subtle body. Yogic transformation begins when the yogin concentrates tenance of the immortal digit of lunar nectar. Even at such times as an all of his vital breaths at the base of his medial channel, in the region of the individual is not practicing yoga, he remains alive by virtue of the sixteenth subtle sun that is burning with all of its twelve digits. This concentration digit of nectar that endures in the cranial vault, identified as the abode of of breaths opens the medial channel, the mouth of which had theretofore bráhman or Siva (both of whom are identified with the number sixteen) in remained closed, at which point the process of yogic reversal truly begins. the subtle body: this is the microcosmic homologue of the single immortal Through heroic efforts of mental concentration and physical exertion, the lunar digit that dwells in the world on new-moon nights.89 The realization yogin now initiates a controlled raising of his seed, the heat of his solar of supernatural powers and bodily immortality requires that that same fires, and his breath along the medial channel.93 At each stage of this pro-

40 41 Chapter Two Categories of Indian Thought

July 14 full sun mundane consciousness sun at summer solstice new moon

N July 14 .rājayaksma

solar pingalā in right side of body

<<< yogic rel uttarayana daksiņāyana ādāna saumya āgneya visarga

lunar ida in left << yogic absorption

side of body

optimal health S January 14

January 14 sun at winter solstice full moon full moon samadhi Figure 2.3. Yogic model of the year Figure 2.4. Ayurvedic model of the year

cess, the fire of the sun in his abdomen decreases incrementally: what the tion.9 This heat, concentrated within the infinitesimal space of the medial yogin is in fact doing is burning up the fire of death (kalagni) with the channel, effects the gradual transformation of "raw" semen into "cooked" immortalizing fire of yoga (kalagnirudra),94 and even perfected nectar, amrta; it is this nectar that gradually fills out the In its simplest terms, this channeling of vital fluid, breath, and heat en- moon in the cranial vault such that, at the conclusion of this process, the ergy is a six-stage process. These six stages are described as the piercing lunar orb, now brimming with nectar, is possessed of its full complement (bbeda) by these vital elements of six energy centers or circles of transfor- of sixteen digits. The brimming downturned moon in the cranial vault is mation called the cakras. The six cakras,"5 strung along the length of the also identified as a thousand-petaled lotus: this is the so-called "seventh" spinal column in the subtle body, are, in order, the miladhara (at the level cakra, the sabasrara. This transformation of semen into nectar wholly of the anus), svadhisthana (at the level of the genitals), manipura (navel), transforms the body, rendering it immortal. anahata (heart), visuddbi (throat), and ajñā (between the eyebrows). Because the yogin reverses all natural tendencies (inertia, being- The prodigious heat generated with the piercing of each cakra, coupled towards-death, extroversion) through his practice, he quite literally re- with the fact that upward movement is here equated with absorption, verses the flow of time; because his is a "regressive practice," it stands the allows for a homologization of each circle of transformation with a crema- conventional models of the temporal cycles on their heads. Here, we il- tion ground, the place of the final sacrifice, and a pralaya, a cosmic dissolu- lustrate with a chart these reversals-between the interplay of the macro-

42 43

Chapter Two Categories of Indian Thought

cosmic sun and moon in the visarga and adana semesters of the Ayurvedic Following their supreme fusion in hatha yoga, the lunar digits and the year on the one hand and the cyclic interplay of lunar and solar kalās in number sixteen seem to pursue separate careers in later Hindu traditions. the subtle body of hatha yoga on the other (figs. 2.3 and 2.4). On the one hand, the number sixteen becomes expanded, through an in- The application of the hathayogic model-of a full, male, immortaliz- flationary tendency quite endemic to Hinduism, into eighteen. On the ing moon in the upper body and a blazing, female, deadly sun in the lower other, kalā becomes a tantric metaphysical category more or less divorced body-becomes greatly expanded in the Puranic reckoning of the great from its prior associations with the waxing and waning moon. It is espe- cosmic time cycles, the kalpas or eons. Here, a divine, cosmic yogin trig- cially in the context of the phonematic emanation of the tantric universe, gers the creation or dissolution of the universe through his cycles of yogic in which the self-manifestation of the absolute is effected through the withdrawal from (which corresponds to cosmic dissolution) and return to garland of the fifty-one ordered phonemes (varna-mala) of the Sanskrit (which corresponds to cosmic emission) mundane consciousness.97 Here alphabet-which are at once acoustic matrices and mother goddesses as well, the phases into which eons are divided, i.e., of nonmanifestation (mātrkas)101-that the term kala is most broadly employed in nondualist and manifestation, are termed niskala and sakala, with kalas and without tantra. Here, through the "particularizing energy" of the kala, the un- kalās, respectively.98 manifest acoustic nature of the absolute, called nāda, "resonance," becomes In fact, time (kala) is always lurking in the background of any discussion condensed into the differentiated phonemes of manifest speech.102 It is of the kalās, because to each kalā there corresponds a lunar day (tithi):99 here that tantrism coopts, in its own way, the notion of the kalās as lunar with each passing kala, another day goes by. Normally, man is the victim digits: the fifteen vocalic phonemes of the Sanskrit language, together with of this relentless passing of the kaläs, the march of time, which, just as it a nasalization called the bindu, are identified with the first fifteen digits of diminishes the moon in its waning fortnight, wears down the human body the moon. The sixteenth kalā, the invisible and immortal digit that is pres- and brings man a day closer to his death. Only the yogin, through his re- ent in all that flows from it, becomes the turning point between vowels and gressive practice, can effect a return to a primordial plenitude, thereby en- consonants in the order of the emanated phonemes: this is the visarga, the abling himself to stave off and even reverse and gain back "lost time." So it surd aspirate often found at the end of Sanskrit words. As such, the six- is that as he enters deeper and deeper into his practice, the yogin turns teenth phonemic kala, the visarga is pictured as the point of intersection back the hands of time, moving backwards through his earlier life and prior or union between utterance and meaning, between language and event, existences and thereby annulling the karma, the effects of the accumulated signifiant and signifié.103 These sixteen phonemic kalās, while identified fruits of past acts, attached to them. Still deeper into his yogic trance, when with Siva, are only enlivened through his interaction with his manifest as- he has remounted the emanated order of creation back to its very source- pect, the goddess Sakti.10* As such, this group of sixteen may also be identi- a source prior to sameness and difference, life and death, male and female, fied as sixteen divine mothers.105 ego and other-he realizes a state of being that is unconditioned even In addition to its extended applications of the full complement of lunar by Time itself. The yogin then becomes a kāla-vañcaka, a "tricker" or digits in its theory and practice, tantrism also expands on the number of "skewer" of Time,100 and many are the myths of Siddhas who harrow hell digits in the full moon, raising it to eighteen. This is a process that was and give Death a thrashing for his having presumed to hold sway over one already under way prior to the time of the tantras themselves. The numeri- of their fold. For the Indian yogin, following the example of the yogic god cal attribute of the absolute brabman is raised, in the Mahābharata, from whose withdrawal from and return to mundane consciousness synchronize sixteen to eighteen.106 Likewise, in the writings of the Mahāyana philoso- the arising and passing away of the cosmic ages, time need not merely pher Asanga, as well as of certain Nath Siddhas, the number of inner move forward: there are also the options of fast forward and reverse. It is "voids" (sünyas) alternates between sixteen and eighteen.107 Finally, in in this way that the yogin, even though he employs the same bipolar system Hindu tantra, the number of kalas is raised from sixteen to eighteen, to for time reckoning as do all other Indians, relates to time in his own unique accomodate for additional metaphysical categories.108 way. Rather than being a slave to time, he is its master. He becomes a "sec- There is yet another important sphere of Hindu life in which the num- ond Siva," the master of a universe he knows to be of his own making. ber sixteen plays an important role. These are the samskāras, the Hindu

44 45 Chapter Two Categories of Indian Thought

God returns to mundane consciousness July 14 srsti a cosmic day begins full sun mundane consciousness

solar pingalā in right NO side of body <syogic re cosmic night samsāra

lunar īda in left side of body

God in samadhi January 14 pralaya full moon a cosmic night begins samādhi Figure 2.5. Yogic model of the cosmic year (Kalpa) Figure 2.6. Yogic model of the year

"sacraments" or "life-cycle rites," which, extending from conception to 5. Yogic and Cosmic Cycles cremation, respect the numerical determinism of the number sixteen.100 Over the millennia, this term has come to take on a wide range of applica- In hatha yoga, the principal motor behind the transformations of mundane tions, one of which is an alchemical one: the series of 16 + 2 processes semen into divine nectar of immortality-and of mundane mind (manas) which lead to the transformation of mercury, the "lunar" semen of Siva, to a state beyond mind (unmand)-is a pneumatic one. It is wind, the dy- into a transmuting agent, are called the samskaras."10 The alchemical sam- namic element in the ancient Vedic triad, which here, taking the form of skäras moreover retain a number of the "lunar" elements intrinsic to the controlled breathing, plays a crucial transformative role in the hathayogic number sixteen. Thus, the alchemical Bhūtiprakarana, in a clear reference system. When the breath is stable, mind and semen are stabilized; but more to the myth of the origin of royal consumption, states that the mineral rasas important, when through breath control (pranāyma) the base of the me- arose from the union of Bhairava and Rohini, the erstwhile spouse of King dial channel is opened, that same breath causes the reversal of mundane Moon. The same source goes on to state that it is only under a waxing or polarities. Rather than descending, semen, energy, and mind are now full moon that alchemical operations leading to bodily rejuvenation will forced upwards into the cranial vault, effecting total yogic integration (sa- succeed." We will return to a detailed discussion of the alchemical sam- mädbi), a reversal of the flow of time, immortality and transcendence over skaras in chapter nine. the entire created universe.

46 47 Chapter Two Categories of Indian Thought

The importance of yogic pranāyama is underscored in Indian systems of time reckoning, especially as concerns two extreme units of measurement. and breathing out. It is for this reason in particular that breath control

It is here that we find not a mere dilation of number-as has been the case plays such a paramount role in the entire yogic enterprise.

with two, three, five, and sixteen-but a veritable explosion thereof, with With these remarks, I present two final charts which compare the basic

numerical progressions raising these figures to powers of fourteen or fac- bipolar model of yogic withdrawal and return on the one hand and the

tors of trillions. Yet, as always, the same phenomena, of bipolarity and ho- kalpic cycle of divine withdrawal (dissolution, pralaya) and return (emis-

mology, lie at the root of such mathematical exultation. sion, srsti) on the other (figs. 2.5 and z.6).

We have already outlined the correspondences between the semesters of the solar year and the fortnights of the lunar month. Moving down the scale from these temporal units, we find that each of the days in a lunar month is divided into bright (day) and dark (night) halves and so on until one arrives at the basic "bipolar time unit" of human, or more particularly yogic, respiration. This is the matra ("measure"),"12 or the elapsed time of a yogic inhalation (homologized with the day, the bright lunar fortnight, and the northern path) and exhalation (homologized with night, the dark lunar fortnight, and the southern path). Ninety-six mātras constitute one ghata; sixty ghatas one day and night; thirty days one month (a day of the ancestors); twelve months one year (a day of the gods); 12 X 360 X 1000 years a mahayuga (a day of the god Brahma); and one thousand mabāyugas a kalpa (a day of the cosmic yogin Visnu). According to nearly every yogic and tantric tradition, Siva is an even greater yogin than Visnu. As such, a kalpa of Visnu is but a single matra, a divine inbreath and an outbreath for Siva-the proof of this being the garlands of skulls, of Brahmas and Visnus of innumerable past creations that he wears around his neck. In the words of the twelfth-century Virasaiva poet Basavanna,

When the ghosts read the writing on the skulls Siva wears around his neck, they know, "This one is Brahma, this one is Visnu, this one is Indra, this is Death," as they play happily with them, Siva smiles, he laughs, our god.11

In the monistic or pneumatic perspective of the yoga-based Indian gno- seologies, it is ultimately breath, breathing in and breathing out, that unites the microcosm to the macrocosm (indeed, atman can be translated as "spirit" or "breath, re-spir-ation")."* The lunar months, solar years, etc. are so many temporal mesocosms, so many levels at which the human be- comes joined to the absolute, through the bipolar dynamic of breathing in

49 The Prebistory of Tantric Alchemy

THREE path ... The monk saw the huge rock and the upside-down monas- tery and realized that the worst had happened. "Alas, Lord King," he exclaimed, " ... I fear that my boys have eaten the roasted alchemist, The Prebistory of Tantric Alchemy and unless they are quickly apprehended they will rebel against you." b. The seventh-century Indian author and playwright Bānabhatta pro- vides the earliest literary account we have of an Indian täntrika in the comic description he gives of a south Indian (Dravida) ascetic who superintended a temple of the goddess Candikā on the road to Ujjain:2 He had a tumor growing on his forehead that was blackened by con-

I. Six Alchemical Accounts stantly falling at the feet of the mother Goddess ... and was blind in one eye from a batch of invisibility salve [siddhāñjana] given him by

a. The Zawgyis or Weikzas, the alchemist-monks of Burma whose eso- a quack ... He had brought a premature fever on himself with an improperly prepared mercurial elixir [rasāyana] used as a vermifuge teric alchemical tradition dates from the fifth century A.D., are indebted at least in part to India for their knowledge of the mercurials with which they ... He had a collection of palm-leaf manuscripts containing material

are expert in "pickling" their bodies. A Burmese legend set in the time of on conjuring, tantra, and mantra, which were written in letters of

the eleventh-century king Anawrahta also betrays an Indian connection:' smoky-red lacquer. He had written the doctrine of Mahākala as such had been taught to him by an old Maha-pasupata. He was afflicted [A monk, having saved two boys from drowning] discovered that they with the condition of babbling about buried treasure and had become were ... Indians by race. He took them to his monastery and, nam- very windy on the subject of transmutational alchemy [dhātuvada] ... ing them Byat-wi and Byat-ta, he brought them up as his pupils. Years He had increased his grasp on the mantra-sädbana for becoming in- passed and the boys became fully grown young men. One day the visible, and knew thousands of wonderful stories about Srīparvata. monk found on the hill-side the body of an alchemist who had died c. Alberuni, the Muslim savant who accompanied Mahmud of Ghazni during the final stages of his experiments, and he instructed his pupils in his conquest of western India between A.D. 998 and 1030, relates a num- to carry it to the monastery and roast it. After the body had been ber of Indian alchemical legends, including the following:3 roasted, the monk said, "Look here, pupils, the roasted flesh of the alchemist is to be eaten only by the Great King of Thaton, so that he In the city of Dhara,+ the capital of Malava, which is in our days ruled will become a mighty man of endeavour and protect our country by Bhojadeva [1000-55], there lies in the door of the Government from its enemies. So I must go to the city to invite the King to din- house an oblong piece of pure silver, in which the outlines of the ner" ... In the darkness, the roasted body of the alchemist shone like limbs of a man are visible. Its origin is accounted for by the following gold, and it gave out such a sweet flavour that the two youths yearned story :- Once in olden times a man went to a king of theirs, bringing to taste the strange flesh ... "Let us just take a bite each," and they him a Rasāyana, the use of which would make him immortal, victori- cut off a tiny part of the roasted body and ate it, but as the flesh ous, invincible, and capable of doing everything he desired. He asked tasted so good they greedily went on eating until the whole body was the king to come alone to the place of their meeting, and the king finished ... The elder brother [said] " ... let us enjoy ourselves." gave orders to keep in readiness all the man required. Then, feeling gay and strong, he lifted the monastery from its foun- The man began to boil the oil for several days, until at last it ac- dations and turned it upside down. "Is that all you can do?" mocked quired consistency. Then he spoke to the king: "Spring into it and I the younger brother, and he lifted a huge rock and placed it on the shall finish the process." But the king, terrified at what he saw, had

48

50 51 Chapter Three The Prebistory of Tantric Alchemy

not the courage to dive into it. The man, on perceiving his coward- There are others quite different from these [other Hindu sectarian ice, spoke to him: "If you have not sufficient courage, and will not groups]; strange fellows these, almost constantly travelling hither and do it for yourself, will you allow me myself to do it?" Whereupon the thither; these are people who scoff at everything, and whom nothing king answered, "Do as you like." Now he produced several packets of troubles. They are people with secrets who, it is said, even know how drugs, and instructed him that when such and such symptoms should to make gold and to prepare mercury so admirably that one or two appear, he should throw upon him this or that packet. Then the man grains taken every morning restore the body to perfect health and so stepped forward to the cauldron and threw himself into it, and at fortify the stomach that it digests very well, such that it is nearly once he was dissolved and reduced into pulp. Now the king pro- impossible for them to eat their fill. ceeded according to his instruction, but when he had nearly finished the process, and there remained only one packet that was not yet f. Two hundred years after Bernier, John Campbell Oman, a relatively thrown into the mass, he began to be anxious, and to think what open-minded7 subject of the British Empire-for whom, however, western might happen to his realm, in case the man should return to life as an superiority and the propriety, even the necessity of the colonial and mis- immortal, victorious, invincible person, as has above been mentioned. sionary enterprise were never open to question-gives a report of an And so he thought it preferable not to throw the last packet into the Indian alchemist that is less flattering than those of his two western prede- mass. The consequence was that the caldron became cold, and the cessors:8 dissolved man became consolidated in the shape of the said piece of silver. A learned Sikh told me of an ... unfruitful experience he had with a

d. In the account he gives of his travels in India in the last years of the gold-making Nirmali sadhu. This man made friends with the Sikh,

thirteenth century, Marco Polo offers the following description of a group and insinuated himself into his confidence. He ... revealed, under

of persons he encountered on the Malabar Coast of India:5 the seal of secret, that he was acquainted with the occult art of transmuting metals. The Sikh ... was much excited at finding that Here these Braaman live more than any other people in the his new friend was a potent alchemist ... The transmuter of metals world .... Moreover they have among them regulars and orders of seemed to live very well, yet occasionally borrowed money, showing monks ... who are called ciugi [Yogis, Jogis] who certainly live more special favour to the Sikh in this matter ... One day the sadhu than all the others in the world, for they commonly live from 150 to showed the Sikh a common bronze ... coin, and then in his presence 200 years ... And again I tell you that these ciugi who live so long put it into a small furnace along with various leaves and roots he had ... eat also what I shall explain ... I tell you that they take quicksilver collected. After an hour or so he produced from his crucible a golden and sulphur and mix them together with water and make a drink out fac-simile of the [coin]. The Sikh, not to be taken in even by his dear of them; and they drink it and say it increases their life ... They do it friend, asked to be allowed to have it tested by a goldsmith. Permis- twice on the week, and sometimes twice each month ... and without sion was given and acted upon, with the result that the experts in the mistake those who live so long use this drink of sulphur and quick- bazaar pronounced it gold of the purest quality. The Sikh was now silver. agog to learn the important secret of gold-making, and many were the rupees he willingly lent the sadhu, in the hope that he would ac- e. At the close of the seventeenth century, the French traveler François cept him as a pupil. But the saintly man of science suddenly and un- Bernier, a Catholic man of letters, gives a more nuanced account of the expectedly decamped. "Alas," said the Sikh after he had narrated these religious practices he observed in Mogul India. While he either ridicules circumstances to me, "I lost more than sixty rupees through that or shows disgust or contempt for nearly all that he sees of Hindu religious imposter. I have since learned how he fooled me, but never a Nirmali life, one group, whom he calls, as had Marco Polo, Yogis, seems to com- sadbu has, since those days, had so much as a drop of water from mand his respect:6 my hand!"

52 53 Chapter Three The Prebistory of Tantric Alchemy

2. Religious Alchemy in India: Three Typologies and rasa sastra, we will have the occasion to refer to a wide array of medical and pharmacological sources throughout this study. The six accounts presented here epitomize the historical evolution and de- I have chosen the term magical alchemy to designate the earliest phase of volution of religious alchemy in India. The Burmese folktale and Bāna- religious alchemy for a simple reason: it is, for all intents and purposes, the bhatta's seventh-century caricature typify magical alchemy, which held the field from the second to the tenth century A.D. Al-Biruni's eleventh- stuff of fairy tales. Transmutation and bodily immortality are its stated goals, but the means to these ends are a matter of serendipity throughout century account is a legendary portrayal of an operation proper to tantric this period. Its watchword is the term rasa-rasāyana-a mercurial elixir alchemy, which enjoyed its golden age from the tenth to the fourteenth cum philosopher's stone and one of the eight magical siddhis of Mahāyāna century. Marco Polo's and François Bernier's descriptions of Yogis docu- Buddhism,14 as well as medieval Hinduism15 and Jainism16-but this is a ment the tradition of Siddha alchemy, which thrived from the thirteenth to power or object to be won or wrested from gods, demigods, or demons the seventeenth century. The decline of Siddha alchemy, already hinted at rather than produced in the laboratory. While most of the alchemical lore in the Aurangzebian firman with which this book opened, is chronicled in of this period is found in Buddhist sources, the secular literature of the Oman's nineteenth-century account. time (of which Bana's Kadambarī is a prime example) also contains its share We may begin to generate a working definition of religious alchemy by of accounts of Hindu alchemical heroes and buffoons. Indeed, the Gupta distinguishing it from what it is not. Throughout the three phases I have age was a period of great syncretism between Hinduism and Buddhism. evoked, religious alchemy interacted with, and was at times even indistin- Both the Burmese folktale and Bana's account also square with a number guishable from, other theoretical and applied sciences in which mercurial of other features of the alchemy of this period, not the least salient of which and mineral-based preparations played a central role. The most important is that it rarely seems to work. Also, as Bana's account intimates, alchemy of these (though perhaps not the earliest)9 was Ãyurveda, whose two foun- appears to be a mainly south Indian phenomenon in this period, as evi- dational works, the Caraka Samhita (ca. A.D. 100)10 and the Suśruta Sambitā denced in frequent mentions of Srisailam-Srīparvata in period sources. (ca. fourth century A.D.),1 contain references to external, therapeutic uses Since India's original fascination with alchemy most probably arose out of of mercury. Following these, the sixth- to seventh-century A.D. Astānga early contacts with a China (India was exporting Buddhism to China in Samgraba of Vägbhatta the Elder contains the earliest Indian reference to this period) whose Taoist speculative alchemical tradition had been devel- the internal use of mercury for therapeutic ends.12 oping since the second century A.D., one might conclude that such tradi- In spite of the fact, however, that Ayurvedic uses of mercury predate tions reached south India via a maritime route.17 those of the tantric and Siddha alchemical traditions that will concern us, While Bana's seventh-century description of the hydrargyriasic Dravida and although Ayurveda later incorporated many of the technical discover- ascetic also contains India's earliest literary reference to tantric manu- ies made by tantric alchemy (especially following the decline of the latter scripts, it would not be until some three centuries later that tantric alchemy, in the fourteenth century), its use of mercurial and mineral-based prepara- as I define it, actually emerged. Here, Alberuni's eleventh-century legend tions falls, for the most part, outside the purview of this study. The reason proves to be nothing other than the narrativization of an actual alchemical for this is that the goals of the Ayurvedic "mercurial science" (rasa śāstra)- operation, as described in the final verses of the Rasārnava (RA).18 Albe- i.e., mineral-based pharmacy-are essentially therapeutic (rogavāda),13 runi's legend is entirely faithful to the tantric spirit of the RA and other whereas the hallmark of religious alchemy is its dual emphasis on transmu- texts of this period. The goal of tantric alchemy is bodily immortality, tational (lohavāda) alchemy and elixir (dehavāda) alchemy, on the bodily invincibility, and transcendence of the human condition. The tantric al- transformation of the living practitioner into a perfected immortal, a Sid- chemist, like most of his brother tantrikas, seeks through his practice to dha, Vidyädhara, or a "second Siva." The Ayurvedic tradition has no such render himself godlike, a second Śiva. pretensions. The Ayurvedic physician's goal is to heal the man and not to The means to this end also distinguishes tantric alchemy from magical create a superman. Because, however, a vast wealth of religious alchemical alchemy as well as from the therapeutic uses of mercurials proper to the doctrine has remained fossilized as it were within the canons of Āyurveda Ayurvedic tradition. In the procedure of which Alberuni's text is a mytho-

54 55

Chapter Three The Prebistory of Tantric Alchemy

logization, the alchemist is instructed to first "test out" his mercury on of wholesale borrowing. The roots of the revolution that was tantric al- metals before throwing himself into his alchemical cauldron. Only mer- chemy may be traced back to the powerful impact of tantrism on Indian cury that has proven itself capable of transmuting ten million times its mystic and metaphysical speculation on the one hand and to developments mass of base metals into gold will suffice here.19 This dual focus is indeed within the medical schools on the other. In this latter context, a gradual the watchword of the tantric alchemical method, which the RA summarizes phasing out of the practice of surgery (salyatantra)-a development some with the pithy formula: yatha lohe tatha debe-"as in metal, so in the body attribute to the pervasive influence of the Buddhist ideal of noninjury .. first test [mercury] on a metal, then use it on the body."20 Moreover, (abimsa)-seems to have been counterbalanced by discoveries and inno- the two elements placed in relation here, metals (loha) and bodies (deba), vations in the field of mercurial and mineral-based medicines.2! Tantric define the two branches of the tantric alchemical synthesis: these are lo- alchemy would have inherited some portion of its science from these new bavada, "transmutational alchemy" and debavāda, "elixir alchemy." As this developments, which begin to appear both in Ayurvedic works and such verse makes clear, however, the transmutation of base metals into gold is tantric alchemical classics as the Rasabrdaya Tantra (RHT) and RA in the not an end in itself, but rather the necessary means to the ultimate end of tenth to eleventh centuries. bodily immortality. Sometime in the fourteenth century, we witness the gradual disappear- Not only are the goals of tantric alchemy consistent with those of the ance of tantric alchemy and the appropriation of its techniques and goals broader Hindu tantric tradition, but so are its means to attaining those of transmutation and transubstantiation by other Indian systems of goals. The alchemical Tantras abound in references to tantric formulae thought and practice, both old and new. As already noted, many of the (mantras) and diagrams (mandalas), as well as in descriptions of divine hier- techniques of tantric alchemy were churned back into the Ayurvedic tradi- archies, yogic and meditative techniques, sexual and ritual practices, and tion from which they had, at least in part, originated. Here the legacy of the Sakta-Saiva devotionalism that are the hallmarks of the tantric tradition tantric alchemy is rasa sāstra, Ayurvedic pharmacy, a sine qua non of Ay- as a whole. Many of the major alchemical works of the period call them- urvedic practice and a discipline that continues to be taught in Ayurvedic selves Tantras and are cast as the revealed teachings of Siva (often in his universities and colleges in India down to the present day.22 tantric Bhairava form) to some tantric form of the Goddess. For reasons I have already qualified this reapplication of transmutational and elixir that will be made clear shortly, tantric alchemy was, in the main, a Hindu alchemy to therapeutic ends as rogavada, the "medical alchemy" specific to rather than a Buddhist occult science. Alberuni's account is set in Dhāra, north Indian Ayurveda. Another development of the same order is rasaci- the capital of the great king Bhoja of the Paramara dynasty, located in west- kitsa, "mercurial medicine." Although this school retains certain of the reli- ern Madhya Pradesh-and it was indeed in western India that the greatest gious elements of tantric alchemy (its devotional cult of mercury, for flowering of tantric alchemy occurred. example), it nonetheless shifts the emphasis of the latter away from What truly sets tantric alchemy apart from magical alchemy is the rigor transmutation and bodily immortality towards therapeutic ends. The rasa- of its method and the remarkable breadth of the botanical, mineralogical, cikitsā school continued, for centuries, to thrive-and even to rival classical chemical, geographical, religious, and technical knowledge it mobilizes in Äyurveda-in south India, the eastern states, and the Sind and was ex- the pursuit of its ambitious ends. Seemingly out of nowhere, the alchemical ported along with classical Ayurveda to Tibet, China, southeast Asia, and science burst upon the Indian scene in the tenth century with a laboratory Sri Lanka.23 Yet another spin-off of the decline of tantric alchemy was the full of specialized equipment and mineral and botanical raw materials in emergence of transmutational alchemy as an independent pursuit. Here, its theoretical inventory which magical alchemy had in no way anticipated. aurifaction, the production of alchemical gold, became an end in itself, While Chinese (Taoist alchemy) and Persian (the Shi'a Jabirian school) rather than a means to the tantric end of bodily transubstantiation and traditions no doubt interacted with tantric alchemy, the Indian material is immortality. Kings interested in increasing the royal treasury are known to so specifically Indian-as much in the subcontinental provenance of its have taken a lively interest in this discipline.24 materiae primae as in its nearly exclusively Hindu religious and metaphysi- The fourth evolute of tantric alchemy was the tradition I will call Siddha cal presuppositions-as to preclude any possibility of this being a matter alchemy, which is most readily identified by its emphasis on the combined

56 57 Chapter Three The Prebistory of Tantric Alchemy

use of mercurial preparations with techniques of batha yoga for the attain- sics contained all that needed to be known to carry out the alchemical ment of immortality and a mode of being on a par with that of the divine complement to the hathayogic practices proper to a two-pronged approach Siddhas and Vidyadharas. In this, it bears certain similarities with the to bodily immortality. This is clearly the purport of the RA (1.18b) when physiological alchemy (nei-tan) of Taoist traditions.25 The complementar- it states: "Mercury and breath [control] are known as the Work in two ity of these two modes of practice is intimated in the accounts of Marco parts." It was the complementary discipline of hatha yoga that required fur- Polo and François Bernier, inasmuch as both present figures they call "Yo- ther explanation, and in this field of expertise the Nath Siddhas simply had gis" (yogins, practitioners of yoga) who are long-lived and healthy because no rivals. Nearly all of India's hathayogic classics are the works of Nath they eat mercury. An important Indian witness to this complementary em- Siddhas (with the most important of these being attributed to Goraksanā- phasis is the fourteenth-century Mādhavācārya, who devotes chapter nine tha, i.e., Gorakhnäth), and all date from the period of Siddha alchemy. Fur- of his Sarvadarsana Samgraba to the "Raseśvara Darsana," the "revealed thermore, as we will demonstrate, the language of the Nath Siddhas' hatha system of the Lord of Mercury."26 What is significant in Madhava's syn- yoga is often nothing other than a projection of alchemical discourse upon thesis is that, vhile he mainly draws on the tantric alchemical classics (the the human body. The human body is an alchemical body. RA, etc.) to expound the basic principles of the alchemical doctrine, the inferences he draws from them relate specifically to the characteristically dual emphasis of Siddha alchemy, and to its emphasis on stabilizing and 3. Magical Alchemy immortalizing the body through yoga.27 The south Indian ciugi alchemists were, in the words of Marco Polo, an The term "Siddha" is an ambiguous one, given that a great number of "order of monks," while those whom Bernier chronicles in the north were Hindu and Buddhist schools, sects, and traditions have been so identi- a itinerant order "almost constantly travelling hither and thither." Now, fied-either by themselves or (retrospectively) by others-since the Gupta while there were many monastic orders and even many orders of itinerant period. The original referent of the term was a class of demigods: in Bud- monks circulating in medieval India, there was only one such order whose dhist and Hindu traditions alike, the Siddhas shared the interface between highly mobile members enjoyed a reputation as alchemists in this period, earth and sky-mountaintops and the atmospheric region-with a horde and these were the Nath Siddhas, also known as Nath Yogis. The Siddha (gana) of semidivine beings. In the words of the fifth-century Amarakośa, alchemists were, by and large, Näth Siddhas; and because the Näth Siddhas "The Wizards (Vidyadharas), Nymphs, Dryads, Protectors, Celestial Mu- were itinerant, they made Siddha alchemy a pan-Indian phenomenon. sicians, Centaurs, Ghouls, Hidden Ones, Perfecti (Siddhas), and Beings: A significant number of alchemical works from this period are attrib- these constitute the class of the demigods."29 Gradually, however, the no- uted to Nath Siddhas, although it must be said that none-with perhaps tion arose that the world or level of the Siddhas and Vidyadharas was one two important exceptions28-are as weighty or innovative as the tantric to which humans too could accede, and so it was that throughout the alchemical classics. Two reasons may be adduced to explain this phenome- Indian middle ages, a growing pool of such Siddhas came to be shared, non. The first of these is that the Nath Siddhas were, for the most part, together with an expanding body of legend on their subject, by Hindus, not members of the Indian literati. Sons of the people-if not the salt of Buddhists, and Jains alike.30 the earth-they simply practiced their alchemy, and when they wrote about Vyāsa, a fifth-century commentator on Patañjali's Yoga Sūtras, provides it, it was more often in compact, even elliptical, poetry cribbed down in us, albeit in an apophatic way, with what may be called the "Siddha char- the vernaculars of the day than in Sanskrit. Had the Nath Siddhas written ter"; that is, a comprehensive account of the goals of the human who as- in Sanskrit, more of their alchemical works would perhaps be extant today. pires to Siddha-hood. This he does in his commentary on Yoga Sūtra 3.51, However, their forte was not writing Sanskrit, but rather living and speak- which states: "When invited by the celestial beings, that invitation should ing in the popular speech of their time. The second reason for the relative not be accepted nor should it cause vanity because it involves the possibil- paucity of Siddha alchemical works is that the old tantric alchemical clas- ity of undesirable consequences." Vyāsa's commentary reads:

58 59 Chapter Three The Prebistory of Tantric Alchemy The celestial beings residing in high regions noticing the purity of the intellect of those who have attained unalloyed truth ... try to The Asuras and the subterranean locus of their wealth seem to be con- invite them by tempting them with enjoyments available in their re- nected to an obscure passage from the eleventh-century RA, which de- gions in the following manner: "O Great Soul, come and sit here and scribes the types of persons best adapted for "Treasure Practice" (nidhi- enjoy yourself. It is lovely here. Here is a lovely lady. This elixir pre- sadhana), and "Hole in the Ground Practice" (bila-sadhana): the former are identified by the digging tools they carry, and the latter by their resem- vents death and decay. Here is a vehicle which can take you to the skies. The tree which fulfils all wishes is here ... here are the per- blance to male Asuras (who are ugly and terrible, in contradistinction to fected Siddhas and the great seers. Beautiful and obedient nymphs, their female counterparts).+0 With regard to the Serpents as custodians supernormal eyes and ears, a body of adamantine strength, all are of fabulous underground treasures, a living tradition among the self- here."31 proclaimed "Jogis" of the Sind province of Pakistan, who claim descent from the great magician and serpent master Gūga Chauhan, is highly illus- In his Jātakamālā, Aryaśūra refers to experts in demonology or "de- trative:41

monic medicine" for the exorcism of demons (bhūtavidya, the fourth limb Jogi tradition has it that seven jogis are required to catch a single of Āyurveda), as Siddhavidyās, persons versed in the occult knowledge of cobra because although the cobra strikes only two and a half times, the Siddhas.32 At first blush, there appears to be no connection between it spits out a substantial amount of venom that can be lethal even these two fields of expertise. Yet, upon closer inspection, we in fact find a from a distance. The seven jogis make seven dunghills, dig seven number of uncanny correspondences, of which many are specifically south ditches behind them, fill these with water and climb inside with only Indian, between the two. To begin, the classic work on Ayurvedic demon- their heads visible on the surface. Then they start playing their en- ology, the Kumara Tantra, is attributed to the great arch-demon of the chanting tunes on their beens to lure the cobra, silently reciting magi- south and villain of the Sanskrit Rāmāyana, Rävana.33 In the ca. twelfth- cal mantars [Sanskrit: mantras] at the same time .. century Tamil Irāmāvatāram of Kampan, the daughters of Rāvana are said Explains Punoo Jogi, "When the sheesh nang ["king cobra": San- to be the wives of the Sittars, the Tamil equivalent of the Siddhas.34 In his skrit sesa nāga] is provoked, he strikes at the first dung heap and sets commentary on Yoga Sūtra 4.1, Vyāsa glosses ausadbi, "botanicals"-which it aflame. While he strikes, the jogi submerges himself in water so Patañjali designates as one of the four means to obtaining siddhis-with that no harm comes to him. Then he lunges again, and yet again but rasayana, which he says may be found "in the dwellings of Asuras ['anti- with less force. Once he has disgorged all his poison, he can no gods,' 'demons']."35 longer fight. You should see his rage then. In one final blow, he Vyasa's association of botanicals with Asuras is echoed in later com- strikes at the treasure he has been guarding for years and destroys it mentaries, including that of the ninth-century Vacaspati Miśra, who also before finally succumbing to the seven jogis.42 That is why the jogis glosses ausadhi as rasayana into which, he says, a man may be initiated "by have never found the treasure that the sheesh nang guards." alluring Asura maidens."36 A similar reading is found in the Bhāgavata Pur- āna (5.24.13), which employs the term rasa-rasāyana to refer to the medici- The Dryads and Serpents, perhaps the most archaic divinities of India, nal herbs employed by the many denizens of the subterranean regions to were gradually absorbed, quite literally, as "props" for the high gods of Hinduism and Buddhism when these became the great religions of the ensure eternal youth and good health.37 Even today, the Sittar physicians subcontinent. In classical Hinduism, it is a Serpent that serves to support of south India qualify one class of treatment as rāksasa.38 the entire universe atop his bejeweled hood; at times when the universe is The great wealth of the subterannean Asuras39 is also enjoyed, according nonexistent (exactly half the time), he supports the sleeping god Visnu, in to epic and Puranic traditions, by the Yaksas (the Dryads of India, whose whom the universe is contained in a nonmanifest state, serving as his leader Kubera is the god of wealth in both Hindu and Buddhist traditions), couch.+3 Like Visnu, the Buddha too is sheltered by the hood of a Serpent, and the Nagas, Serpent deities, subterranean plutocrats as well, and India's and the earliest iconographic depiction of such is in fact Buddhist rather version of the archetypal dragon who lies atop hoards of buried treasure. than Hindu. This is the third-century A.D. sculpted image of the Buddha,

60 61 Chapter Three The Prebistory of Tantric Alchemy

on a Serpent couch, from Nagarjunikonda (Andhra Pradesh) in south In- country, who first talks about the Siddhas in an informed way. This is the dia. It is precisely in this region that the ca. third-century A.D. Buddhist seventh-century Tirumūlar who, while he is generally held to be the first Gandavyūha locates the seat of the bodhisattva Mañjuśrī, whom it says of the eighteen Sittars (i.e., the Siddhas of Tamil tradition) and the founder converted many indigenous Serpents to Buddhism.# This region is also of the Sittar school, speaks of the Siddhas-although it is unclear whether one that produces a great number of figures with names containing Nāga he means the semidivine or the human variety-as if they were already an (serpent): the most illustrious of these is Nāgārjuna, whom a number of established institution. In his Tirumantiram, Tirumūlar defines the Siddhas later traditions identify as a Siddha alchemist hailing from Srīparvata. as those "who have experienced divine light and divine energy [śakti] from The ca. first-century A.D. south Indian play, the Mrcchakatikā (The within and through yogic integration [samadhi]."$3 Sittar tradition main- Little Clay Cart), the first literary source to make mention of what is ap- tains that Tirumular was the disciple of the alchemist Nandi(kesvara); parently a human Siddha, is the work of an author from Andhra Pradesh, while a certain alchemist by the same name is cited in a number of post- who very likely would have hailed from Sriparvata (also mentioned in the twelfth-century sources, these never attribute any alchemical texts to him, text).+5 This Siddha reminds us once more of Banabhatta's hapless south nor are there any extant works of his to prove he was ever anything more Indian Saiva ascetic who had lost his sight in one eye due to an invisibility than a legendary human or a god of alchemists.54 salve called siddhāñjana ("perfected, super ointment"), and who knew myri- The discipline of Siddha medicine, butressed by a Siddha alchemy that ads of stories concerning Srīparvata. Slightly prior to Banabhatta, the is in fact a synthesis of hathayogic and alchemical practice, is the present- south Indian author Subandhu provides an early mention of mercury (pār- day legacy of the Sittars in Tamil Nadu. While the extant alchemical works ada), as well as of Srīparvata, in his comic play, the Vasavadattā.16 Now, of the eighteen Sittars,55 the storied founders of these traditions, turn out śrīparvata (or śriśailam)-generally identified with a holy peak located in to be quite recent,56 the body of legend attached to these figures carries us the central Deccan plateau (Kurnool district, Andhra Pradesh) of south back, once again, to the lore of magical alchemy. India+7-is a highly generic term for a mountain or hilltop in the Sanskrit A limited number of documents indicate that a historical figure named and Hindi languages: both simply mean "splendid, auspicious, or excellent Nandi may have lived in the same century as Tirumūlar. This Nandi, a peak." Therefore, while the greatest wealth of Siddha lore, both Hindu and Buddhist monk from central India, left India in the middle of the seventh Buddhist, is attached to this toponym, we cannot be certain that it is always century, traveling by sea to Sri Lanka and southeast Asia before arriving the Srisailam or Sriparvata of the east central Deccan that is being evoked. in China in A.D. 655. A year later, the emperor of China shipped him off The earliest Hindu mention of Srisailam is found in the Mahabharata to sea again to collect medicinal herbs. He returned to China in 663.57 which, precisely, identifies it as a place at which to attain siddhis, the stuff Prior even to Nandi, according to Sittar tradition, was the alchemist Bogar that Siddhas are made of.48 It is the Buddhist religious literature, how- (or Pokar or Bhoga), whose links with China are also remarkable. The Sit- ever, that contains the greatest wealth of early references to Srīparvata- tar Bogar, who is said to have lived in the third to fifth centuries A.D., is Srisailam. Located in the heart of the Andhra country, a south Indian today worshipped at that site in the Palani Hills of Tamil Nadu where he "Buddhist fief" in the first centuries of the common era, this site was is said to have practiced and taught alchemy.58 Traditions concerning Bogar a launching point for missionaries to Kashmir, China, Bengal, and Sri cast him either as a Chinese philosopher who came to India for the study Lanka.49 There are, however, no mentions of siddhis (rddbis in Pali) or Sid- of medicine, traveling first to Patna and Bodhgaya before taking up resi- dhas in Buddhist evocations of Srīparvata,50 unless one chooses to see in dence in Madras; or as a south Indian Sittar, who traveled to China and the name of the third-century A.D. heretical "Siddhathaka" school of that taught alchemy to a ruler named Kong (his disciple Konganar,59 according region a forerunner of the term Siddha.51 Buddhism was fast losing ground to Sittar tradition) before returning to south India.60 to Saivism in the region by the seventh century, with temples to Siva Sid- The greatest historical value of these accounts of Bogar and Nandi lies dheśvara ("Lord of the Siddhas") beginning to appear at Srisailam at that in their references to early maritime exchanges between India and China. time.52 The sea trade between India's southeast coast and Han China (via the en- It is a figure from Tamil Nadu, some distance to the south of the Andhra trepôt of Hanoi), already well under way by the beginning of the common

62 63 Chapter Three The Prebistory of Tantric Alchemy

era, especially thrived during the golden age of Sino-Indian exchange-of Indian king (of Udyana, in the Swat region of present-day Pakistan) religious ideas as well as goods-of the third to eighth centuries A.D. This quizzed two Chinese Buddhist visitors to his court on the "Taoist matters" was, of course, the period that first saw the export of Buddhism from India of medicine and science and the silver and golden palaces of the Im- to China and, slightly later, the introduction of certain elements of Taoism mortals.70 into Indian religion and culture. (Overland trade along the Silk Road- The picture that emerges from this period is one of an ongoing ex- which brushed India's northwest frontier at Taksasila and Peshawar, and change between India and China regarding matters alchemical, matters in slightly above the Jhelum River in the Punjab-and its spurs, which ran which China, even if it appears always to have been ahead of India in inno- down into the subcontinent's great trading hubs at the port of Bhrgukaccha vations throughout this period, nonetheless looked to India for inspira- (Broach), Ujjain, and Patna (via Kathmandu), had already been flourishing tion.71 Such is perhaps understandable, given the incredible impact Indian since about 11O B.C.).61 Buddhism was having on China. We have already seen that Taoist alchemical speculation and experi- If, however, Indian religion and science were in vogue in China, so too mentation had already reached an appreciable level of sophistication by the were Chinese traditions in India. More than this, whatever India exported time of these exchanges.62 Here let us also note the important fact that over the mountains to China seemed to come back, in altered form, to be nearly all the mercury the Indian alchemists would later use came from reappropriated by the Indians a few centuries later. A prime example is the China: given the fact that there exist no mercurial deposits on the subcon- Indian yogic tradition, which China received, in a fairly unsophisticated tinent, China was India's nearest and obvious supplier.63 It is therefore state, along with Buddhism, in the third century A.D. But the Taoist China rather curious that the majority of historical notices from this period in- of the time was already experimenting with what it termed "embryonic dicate that it was China that was looking to India for alchemical drugs and respiration" and the "feeding of the vital principle"-practices that were data. When the Han emperors first asked for instruction in the Buddhist more or less equivalent to Indian yogic techniques, but which, in the Chi- faith, they also requested a "drug of immortality" they supposed India to nese context, already bore a veneer of alchemical symbolism. A yogic tech- possess.64 nique of apparent Chinese origin, called huan-ching or "making the Yellow While the Chinese have no record of having been visited by an Indian River flow backwards," identified with either the practice of urethral suc- alchemist named Nandi, they do mention an Indian scholar named Naray- tion or that of internally raising semen along the spinal column, would first anasvāmin, who was captured and held in the Chinese court in about A.D. make its appearance in Indian Mahāyana sources72 and later reappear as an 649 because he knew of an elixir of life; later, in 664, an Indian physician erotico-yogic technique of the Nath Siddhas.73 Indeed, Indian tantra first named Lokāditya would be summoned there to serve as an alchemist- appeared at points of contact with Taoist China and, in the words of Joseph in-residence.65 Slightly earlier, in about A.D. 646, the Chinese offered a Needham, upon whose masterful Science and Civilisation in Ancient China I Sanskrit translation of the Tao te ching to the king of Kāmarūpa (Assam), have been relying throughout this section, "the Taoist department of Bud- purportedly in exchange for information concerning transmutational and dhism was Tantrism." So it was that when Indian tantrism was first intro- elixir alchemy.66 duced into China in the eighth century by the Buddhist monks Subhākara- In charge of that translation was the great seventh-century Chinese simha, Vajrabodhi, and Amoghavajra, a certain number of its techniques pilgrim Hsuan-tsang who would, in his Su yu chi (Buddhist Records of the were merely "returning" to their country of origin, from which they had Western World), record accounts he had heard in the court of King Harsa been exported but a few centuries earlier.74 concerning the famed Indian alchemist Nagarjuna.67 Yet this was the same The principal Sino-Indian points of contact in this period of exchange royal patron for whom Banabhatta wrote his description of the hydrargyr- were (1) the mountain passes located in the northwestern region of the iasic south Indian ascetic, in his Kadambari;68 and the same seventh-century subcontinent (Himalayan Pakistan and Afghanistan) and (2) Assam to the India in which the Chinese pilgrim I-Ching reported, in the account he east-these being the two regions through which the Indian spurs of gave of his travels there between A.D. 671 and 695, cases of grave misuse the Silk Road passed-as well as (3) the major seaports on the coastlines of mercuric ores as elixirs.69 It was also an India in which a sixth-century of Tamil Nadu and Gujarat, which linked India to China and the west via

64 65 Chapter Three The Prebistory of Tantric Alchemy

the maritime routes of the day. According to the evidence at our disposal, two illustrious travelers along these trade routes in this period would have (Gujarat), Murshidabad (Bengal), Calcutta, and Madras have long been

been the "Chinese goddess" Tāra and the element mercury (pārada),75 centers for the fabrication of synthetic cinnabar and calomel (mercurous

whose entrances into the subcontinent appear to have been curiously chloride), using native Indian minerals and imported mercury, since at

linked. Both are said to be sixth-century imports from "greater China" least the sixteenth century A.D.86 Also supporting a direct Sino-south In-

(mabācīna, a blanket Sanskrit term applied to both China and Tibet) into a dian maritime link in matters alchemical is the fact that Sittar alchemy and

nascent Indian Buddhist tantric tradition, and both are purported to have medicine have always been more mineral based and "Taoist" than the more

been brought to India by none other than Nāgārjuna.76 herb-based northern traditions.87

The question then arises: Did "Nāgarjuna" bring Tārā and mercury to Against a maritime link, and in favor of an overland connection in the

India by land or by sea? Both mercury and Tara are mentioned in the mercury and Tara trade, we can marshal the following evidence. The Ti-

sixth-to seventh-century south Indian Vasavadatta,77 which would indicate betans themselves have historically purchased their mercury from Chinese

a maritime link with eastern China; but both are also mentioned in the traders who transported it to Tibet; here, the obvious link would have been

sixth-century Astanga Samgraba,78 which was probably composed in Ujjain, an overland one.88 So too, the mercury that reached Varahamihira (author

in western Madhya Pradesh. The earliest recorded case of Tara worship in of the Brhat Samhita) and Vagbhatta the Elder in northwest India quite

India comes to us from seventh-century Nalanda, the great Buddhist com- certainly traveled overland, via the Silk Road or one of its spurs. What of

plex located in modern-day Bihar (whose main links to China were over- Tara? On the strength of two passages from the third- to eleventh-century

land, via Patna to the north);7º a century later, we read of a temple built to A.D. Sādbanamālā (The Garland of Practice), Tāra is generally considered

the "Chinese goddess" in Kāñci, on the coast of Tamil Nadu.80 to have been brought to Nalanda, directly from Bhota (Tibet) by Nāgār-

This coastal location reminds us of what may have been Tara's original juna. Yet, in fact, the passages in question concern two different forms of

role: she was a goddess of navigation, of sea crossings-tāra is generated Tārā, named Vajra ("Diamond") Tārā and Ekajațā ("One Lock of Hair").89

from the verb tr, to cross over the sea.81 A fanciful eighteenth-century Ti- Now, Ekajatā is also known as Mahācīnatārā, as if to emphasize her Ti-

betan source, the dPag-bsam-ljon-bzan, brings these elements together in betan roots. In the Tibetan Buddhism of the Dragon Kaju sect, Ekajatā

a unique way. Saying that Nāgārjuna was born in Kāñci, it goes on to de- is in fact the consort of an "ancient" Nyingma divinity named Quicksilver

scribe how he sailed across the sea on a craft made of two fig leaves, until (i.e., mercury), identified with Bhairava Yamantaka ("Terrible Slayer of

he came to an intermediate continent, from which he brought a gold- Death"). In the Padmaist literature of the same Nyingma school, she is the

making elixir back to his monastery at Nalanda.82 consort of the god Amrtakundali, who is also known by the epithet "King of Elixirs."90 Now, Tibet's source of mercury has long been Yunnan, in southeastern China,83 which shares a border with Annam (Vietnam), of which Hanoi With this, we perhaps arrive at the source of the tradition that Nagar- juna brought "mercury," together with Tāra, from Tibet in this period. Yet was the major port for maritime commerce with India and southeast Asia. if Nāgārjuna did indeed bring Tāra and mercury to India from Tibet, it is That the back country of Annam was rich in mercurial ores is demon- strated by a historical anecdote from China: the great mercurial alchemist curious that the sixth- to seventh-century Astanga Samgraba, which men- tions both Tara and mercury, should make no mention of Nāgārjuna. Ko Hung (ca. A.D. 300) requested and received a transfer to that province, More than this, Tibetan tradition itself maintains that the worship of Tārā into an administrative position unworthy of his rank in the imperial bu- was introduced into that country from Nepal-in the seventh century, i.e. reaucracy, in order that he might have ready access to the cinnabar (red later than the earliest mentions of her in Indian sources!91 This would mercuric sulfide) he needed to conduct his experiments!84 We further know make the claim that Nagarjuna introduced Tāra to India "from Tibet" an that mercury was carried in the cargo holds of the ships that plied these Sino-Indian trade routes in later centuries. On the one hand, we find a untenable one, unless it is the two specifically Tibetan forms (Vajratārā and Ekajatā) of the goddess that are intended. As we will shortly see, the alche- report of over a ton of the stuff being warehoused in the Malaysian port of Malacca;85 and on the other, we know that the Indian port cities of Surat mist Nāgārjuna manages to elude the historian at nearly every turn. I have already noted that there are no mercurial ores native to the Indian

66 67

Chapter Three The Prebistory of Tantric Alchemy

subcontinent. The same, in fact, holds true for Tibet, which means that if juna the Hindu tantric sorcerer, Nāgārjuna the Nāth Siddha, Nāgārjuna Nāgārjuna brought mercury to India from Tibet at the end of the Gupta the north Indian medical author, Nagarjuna the south Indian medical au- age, then it was very likely Chinese mercury that he brought from there; thor,"7 Nagarjuna the Buddhist alchemist, Nagarjuna the Jain alchemist, for it is from Yunnan in southeastern China that the Tibetan supply Nāgārjuna the northern Hindu alchemist, Nagarjuna the southern Hindu came.92 This Chinese source likely lies at the origin of one of the many alchemist,98 Nāgārjuna the eye doctor, Nāgārjuna the sexologist, Nāgār- Sanskrit terms employed for cinnabar, the most commonly occurring mer- juna the parfumeur-so many Nagarjunas, so little time! Whereas Albert curial ore in the world. This is the word cinapista, "Chinese powder." Many Grünwedel called him the "Faust of Buddhism,"99 Max Walleser, the west- of the other terms for mercury or its ores also indicate regions to the ern scholar who probably toiled the longest at attempting to identify the northwest of India (and west of Tibet) as its region of provenance. Pārada, historical Nāgārjuna, concluded that no Nāgārjuna ever existed!100 Tibetan the most common alchemical term for mercury, refers to Pārada-desa, the historians have had it both ways, simply allowing that a single figure named land of the Parthians or Paradas of Transoxiania or the Baluchistan region; Nāgarjuna alchemically prolonged his life for 529 to 1000 years, during darada, red cinnabar, to Darada-desa, the modern Dardistan, in northern which he found the time to dabble in or master all of the fields men- Kashmir; bingula, cinnabar, to Hinglāj (-Devi) in Baluchistan or to a coun- tioned above. try called Hingula. The term mleccha, "(proper to) the (central Asian) bar- Nagarjuna bursts upon the stage of intellectual and religious history as barian (races)" would have been a more blanket term.93 Cinnabar and other the south Indian founder (ca. second century A.D.) of the Mādhyamika mercurial ores occur naturally throughout the trans-Himalayan regions of school of philosophy within the then-nascent current of Buddhism called central Asia, as well as in Chitral (Pakistan), Garmshir (Afghanistan), and Mahayana.101 An extremely prolific writer, he was the author of the seminal Yunnan, and it was from these regions that India's supply of mercury would Mahāyāna text, the Prajnāparamitāsāstra, a commentary on the revealed likely have come.94 "Teachings of the Perfection of Wisdom" (Prajñāparamitā Sūtras). Ac- cording to legend, these teachings had been preserved for millennia in the netherworld of the Nägas, before being restored to the world by Nāgār- 4. Desperately Seeking Nagarjuna: The Buddhist Evidence juna. Chinese sources relate that a great serpent (mabānāga) opened a seven-jeweled chest for Nagarjuna, from which he recovered the Vaipulya The origins of Indian alchemy are closely linked to the figure of Nāgār- (Mahāyāna) teachings;102 according to later Tibetan legend, he was initi- juna, who, whenever one attempts to pin him down as a historical person- ated by the Nägas (whence his name) and received the Perfection of Wis- age, proves to be quite as protean and mercurial as quicksilver itself. One dom texts directly from them.103 has the feeling that if it were only possible to extract Nāgarjuna the al- A number of sources, whose historical veracity is equally suspect, in- chemist out of the welter of the Nagarjunas who dot the mythology and dicate that the Mādhyamika philosopher Nāgārjuna was the protégé and history of Indian religions and medicine, one would be well on the way to companion of a certain ruler of the Satavahana dynasty which ruled over generating a coherent history of Indian alchemy.95 Yet, there are so many the Andhra country between the first century B.c. and the second century Nāgārjunas to choose from. In addition to the Nāgarjunas I have already A.D.10 Banabhatta speaks in his Harsacarita of "a monk named Nāgārjuna" mentioned to this point (the Nagarjuna who purportedly brought mercury who had presented Satavahana with an elixir in the form of a jewel called and Tāra to India, and the Nagarjuna who sailed to the intermediate conti- "Pearl-Wreath Mandakinī," which had been given to him by the king of the nent on a fig-leaf boat to recover a "gold-making elixir") there are perhaps Nagas.105 According to Hsuan-tsang, the Chinese pilgrim who was Bana's a dozen other figures with the same name-who may or may not have been contemporary at the court of Harsa, the "Bodhisattva Nagarjuna" was so the same person, and who, all told, would have authored over a hundred skilled in the art of compounding medicines that he had produced a pill works on a wide array of subjects96-connected in some way with the In- with which he had extended his own life as well as that of his royal compan- dian alchemical tradition. These include Nagārjuna the Mādhyamika phi- ion for many hundreds of years. Later, as Hsuan-tsang relates, it happened losopher, Nagarjuna the tantric Buddhist author and commentator, Nāgār- that the king ran out of funds while building a monastery for Nāgārjuna at

68 69 Chapter Three The Prebistory of Tantric Alchemy

Bhramaragiri, "Black Bee Hill." At this point, "the Bodhisattva Nāgārjuna (Lung-Mêng Phu-Sa) scattered some drops of a numinous and wonderful Tibetan sources make, mainly in the form of retrospective prophecies

pharmakon over certain large stones, whereupon they all turned to gold," concerning the Buddha. These prophecies, which begin to appear in the

thus resolving Satavahana's problems and his own.106 While there is good fourth- through fifth-century Mahamegha Sutra, are the reef against which

evidence that such a monastery was built in the Andhra country, we cannot every attempt to locate the historical Nagarjuna has foundered. The pivot

be sure that it was built prior to the fifth century A.D., at which time it is of these so-called prophecies-a figure whom tradition identifies as "Sid-

mentioned by the Chinese pilgrim Fa-hsien (who identifies it with Srīpar- dha Nāgārjuna"-would have been an important seventh- or eighth-

vata).107 Hsuan-tsang further describes a meeting he had, somewhere in century teacher in the fledgling tantric Buddhist tradition. It is this Nāgār-

the northeastern Punjab, with a brahmin disciple of bodhisattva Nāgārjuna juna who is said to have been Sarāha's disciple at Nalanda and to have

who, although he was seven hundred years old, looked to be only thirty. 108 brought mercury and the goddess Tāra from Tibet to India.

Even in those Chinese, Tibetan, and Indian legends in which the phi- Before we proceed any further, two points need to be made clear con-

losopher Nagārjuna is made out to be an alchemist of sorts-the possessor cerning "Siddha Nägārjuna." The first of these is that he is nowhere re-

of an elixir of long life and some rudiments of transmutational alchemy- ferred to as "Siddha" in the period (ca. seventh through eighth centuries) in which he purportedly lived. He is either simply called Nāgārjuna or has there is never any specific mention of the raw materials (such as mercury) the title Arya prefixed or the ending -pa (or -pada) suffixed to his name.112 or techniques he employs in his craft. This omission is of a piece with the The Siddha appellation is, as we will see, an ex post facto categorization magical alchemy of the period on the one hand and, on the other, with a handful of vague references to aurifaction contained in a number of Ma- on the part of later tantric genealogists who, in the eleventh century, began

hãyana texts, of which at least two are attributed to the philosopher Nāgār- to refer to their founding gurus as the Siddhācāryas ("perfected teachers"),

juna. These are the Mabāprajnāparamitāšāstra, which contains four refer- Mahāsiddhas, or simply the eighty-four Siddhas. However, just as persons living in Palestine prior to the birth of Jesus Christ did not refer to their ences to transmutational alchemy; and the Avatamsaka Sūtra (attributed to Nāgārjuna by Fa-hsien and translated into Chinese in A.D. 699).109 In addi- age as "B.c.," so the Indian innovators of tantric Buddhism did not call themselves "Siddhas." tion to these sources, references to magical alchemy are found in the fourth- through sixth-century Mabayāna Samgraba Bhasya (a commentary Second, the confusion of this seventh- to eighth-century figure (whom

on Asanga translated into Chinese in about A.D. 650 by Hsuan-tsang) and later convention identifies as "Siddha Nāgarjuna") with the second-

the Abhidharma Mabavibbasa (translated and considerably lengthened over century philosopher of the same name likely derives from the fact that each

a fifth-century recension by Hsuan-tsang himself, in A.D. 659).110 was associated with a figure named Aryadeva. Nāgarjuna the philosopher

Given the paucity of alchemical references in writings attributed to the was the teacher of Aryadeva, while "Siddha" Nāgārjuna's disciple Karņar- ipā was also known as Aryadeva."13 Grafted to this composite Nāgārjuna second-century philosopher Nagārjuna, one wonders why this figure are associations-which I have already noted in the writings of Hsuan- should have been perceived, in the seventh-century sources I have been tsang and others-with a king named Sātavāhana and the site of Srīparvata. citing, to have been such a great alchemist. The most plausible explanation would be that Banabhatta and Hsuan-tsang had heard of another Nagār- The most baroque concatenations of these data are the work of the Ti- betans, whose "prophecies" concerning Nagarjuna maintain that he was juna, a near contemporary of themselves, and had conflated this later figure born in Saurashtra (Gujarat) 400 years after the Buddha's parinirvāna114 with the renowned Mädhyamika philosopher-a not-untenable proposi- and lived for over 500 years, passing the first 200 years of his life at Nal- tion, given the tradition that he had produced an elixir of long life. A num- ber of lineage lists identify a Nagarjuna who would have lived in the first anda, the second 200 "in south India,"115 and the final 129 or 171 years at

half of the seventh century as the disciple of Saraha at the great Buddhist Srīparvata, where he voluntarily gave up his life at the request of Sātavāha- na's son. In nearly none of these prophecies-neither those of the Tibetans monastic university of Nalanda, where alchemy was reputedly part of the nor the Indian or Chinese sources upon which they drew-is Nāgārjuna curriculum .! 11 ever referred to as a Siddha; on the other hand, the name of this figure It is just such an assumption that a number of Sanskrit, Chinese, and alternates between Nāgārjuna, Nāgabodhi, Nāgāhvaya, and Nāgarāja.116

70 71 Chapter Three The Prebistory of Tantric Alchemy

His amazing longevity is a tenable proposition only if one is prepared to sattva-Ādibuddha, whom they would identify with Ādinatha; and Bhairava, accept the notion, as the Tibetans did, that Nagarjuna's alchemy served to Bhairavī, and Mahakala, whom Buddhism borrowed from Hinduism. In dilate his life span. concrete terms, Indian Buddhism was losing its specificity, and thereby its It is only in the late Tibetan sources, moreover, that we find explicit royal patronage, a situation which, in the face of the Muslim conquests, references to Siddha Nāgārjuna as an alchemist. The fourteenth-century proved fatal. historian Bu-ston states that Nāgarjuna procured a gold-making elixir Of course, export Buddhism of a tantric stamp continues to flourish from the intermediate continent and so saved Nalanda from the ravages of in Tibet (into which it was introduced, from eastern India, perhaps from famine. In the early seventeenth century, Tāranātha expatiates on this Nalanda itself, in the eighth century) down to the present day; and it is theme, stating that "with the help of the art of alchemy, he maintained for precisely in Tibet that we may glimpse what became of the Buddhist al- many years five hundred teachers of the Mahayana doctrine at Śri Nal- chemical tradition. What is intriguing about the alchemy of Tibetan Bud- endra [sic]." Later, he adds that "from the time he became a rasāyana siddha, dhism is that it more closely resembles earlier Taoist practices and contem- his face shone like a gemstone." In a similar vein, another late Tibetan porary Hindu erotico-mystical techniques than it does the procedures and source states that after Nagarjuna had received his tantric initiation from goals of transmutational and elixir alchemy. For all intents and purposes, his guru, the Siddhācārya Sarāha-pā, he "attained success ... especially in external alchemy disappears from Buddhism at about the same time as rasāyana" such that his body became as hard as a diamond (vajrakāya- Buddhism disappears from the Indian subcontinent. So it is that we find siddhi). This, however, runs against the teachings of his guru, given the very few Tibetan references, apart from those cited a moment ago, to the fact that Saraha-pa criticizes those who practice rasa-rasayana in one of his practice of any sort of alchemy by the Siddhācaryas-i.e., by the teachers vernacular songs.1!7 Of the fifty-nine works attributed to Nagarjuna and who founded the tantric Buddhist path, between the seventh and twelfth translated, in the twelfth through thirteenth centuries A.D. into Tibetan in centuries A.D. Karņari-pā obtains an elixir from urine; Carpațā-pā gains the Tanjur, none contains any alchemical material.118 One would therefore the power to transmute as well as an elixir of immortality from a boy's have to assume that if a seventh- through eight-century Nāgārjuna did penis and anus, respectively; and Vyādi-pā produces an alchemical elixir practice alchemy at Nalanda, he left no written records of his work there. with the aid of a prostitute; this last figure is also the author of two of only Elsewhere, the Tibetan mythology of the eighty-four Siddhas casts four Indian alchemical works to be translated into Tibetan and included in a number of "debates" between Buddhist and Hindu figures as so many the Tanjur.121 "siddhi-contests." So we read that Lalitavajra had a contest of magic power The Kalacakra Tantra, with its early eleventh-century Vimalaprabbā with the Hindus in the kingdom of Naravarma in the west. "The teacher commentary, offers us the most penetrating view we have of any specifi- swallowed a quantity of poison and two wine-jars full of mercury. Still he cally Buddhist alchemical system.122 In contradistinction to the "external" remained unaffected. The king was full of reverence."119 If, however, the Hindu tantric alchemy that was emerging in the same century, Buddhist Buddhists won this particular battle, in the end they lost the Indian war: a alchemy was headed in an "internal" direction. This is precisely what we victim of the Muslim invasions that were overtaking the continent, as well find in the Kalacakra Tantra tradition: the external manipulation of metals as of its own growing resemblance to Hindu Tantra, Buddhist Tantra and and mercurials, which it calls gold making (gser 'gyur), has been declared Buddhism effectively disappeared from the Indian subcontinent by the "mundane" and inferior to the "transmundane" inner alchemy (rasāyana) twelfth century. In a sense, the early alchemical tradition epitomized a syn- of "channels and winds," which can lead to direct enlightenment.123 Those cretistic trend that had been ongoing since the origins of Tantra, with man- passages of the Vimalaprabha which do, however, speak of external al- tras, clan lineages, and divinities converging with one another.120 The al- chemy, are remarkable in their resemblance to portions of the eleventh- chemists' gods were precisely those divinities who had come to occupy the century Hindu tantric alchemical classic, the RA.124 But from a closer per- no-man's-land of the Hindu-Buddhist tantric divide: Tārā and Aksobhya, spective-or at least from the perspective adopted by both the Padmaists of whom Hindu forms would emerge in about the tenth century; Avalok- of the Nying-ma school 125 and the encyclopedist Bo dong-inner alchemy iteśvara, whom the Hindus would identify with Siva Lokanātha; Vajra- is tantamount to an "extraction of the essence" (bcud len) that is quite iden-

72 73 Chapter Three The Prebistory of Tantric Alchemy

tical to the Taoist, Mahayana, and Hindu hathayogic techniques of urethral shall see, a significant number of Buddhist teachers, in whose number Nã- suction and the "hydraulic" raising of the semen along the spinal column. gārjuna figures prominently. Elsewhere, we may attribute uncanny paral- In Vajrayana parlance, "alchemy" consists of the fixing of the bodhicitta- lels between the Buddhist and Hindu tantrism of this pivotal period to the by combining the ambrosial essences (bdud rtsi) of Prajña (the goddess simple fact that the two traditions were, in spite of their professed mutual "Wisdom") and Upaya (the male "Skill in Means")-and the cultivation animousity, so close to one another.132 Most striking in this regard is the of a "rainbow body" (ja' lus) or a "body of light" ('od lus).126 In concrete resemblance between Tibetan alchemy's "extraction of the [innate, divine] terms, the goal of tantric Buddhist alchemy is to produce the nectar of essence [from the gross corporeal body]" and a Hindu tantric practice de- immortality and wisdom through a combination of semen and uterine scribed by Bhavabhūti in his eighth-century play, Mālati-Mādhava. Here, blood that is at once yogic and sexual. Hindu alchemy employs the same a female Kāpālika tāntrika named Kundakapālā attains the power of flight techniques, but complements these with the mineral-based elixirs of exter- by extracting the quintessence (pañcamrta) of her five bodily elements. nal alchemy. Whereas the erotico-mystical practices of tantra (which This same passage is also a Hindu watershed inasmuch as it constitutes the Hindu yogins would later term vajroli mudra) become identified as "al- earliest mention in Indian literature, sacred or secular, of the six energy chemy" in Tibetan Buddhism,127 erotico-mystical practice and external al- centers (cakras) of the subtle body in the practice of hatha yoga. It may be chemy remain two complementary yet distinct disciplines in Hindu tan- herein that the original sense of the term rasayana lies: it is the "coming tric alchemy. forth" (ayana) of the fluid essence (rasa).133 It is apposite to note that the tenth-century twilight of magical alchemy Now, the Hindu134 tantric schools in whose works the six cakras of ha- is also the dawning of no fewer than three distinct forms of tantric practice thayogic practice make their earliest appearance are the paścimāmnāya or from within and beyond the borders of India. On the one hand, we observe Western Transmission and the Yogini Kaula founded by Matsyendranath, the emergence of tantric alchemy, which combines an "external" transmu- which predate the Trika Kaula reformer Abhinavagupta, who makes veiled tational and elixir alchemy with the "internal," but nonetheless concrete allusions to both in his ca. A.D. 1000 Tantraloka.135 As we will show in some (and explicitly hydraulic) practices of hatha yoga and tantric sexual tech- detail, these two Kaula sects were closely related, mainly through the per- niques. Its goal is the production of an immortal yet concrete diamond body son of Matsyendra, whom the Nath Siddhas would later claim as their that transcends the laws of nature. Second, there is the Tibetan Buddhist founding guru. Matsyendra's classic work, the Kaulajnānanirnaya (KJñN) internalization of alchemy into a meditative and ritualized form of yoga, (Discussion of the Kaula Gnosis), revealed in Kamarupa (Assam) in the whose goal is the acquisition of a spiritualized body of light.128 Third, there ninth or tenth century, is important inasmuch as it constitutes the earliest is the Hindu Trika Kaula, which sublimates the same concrete hathayogic explicit account we have of Hindu India's earliest Siddha clans or sects, the and sexual techniques into a meditational and ritual system whose goal is Siddha and Yogini Kaulas.136 The revelations of the KJnN were "brought the acquisition of a divinized body of sound.129 All three of these emergent down" (avatārita) at a site called Candradvīpa ("Moon Island"), a toponym systems were in fact refinements upon an original tantric synthesis, which which further links Matsyendra's Siddha and Yogini Kaulas to the Western was based, to a great extent, on the worship of terrible, ravenous deities, Transmission, whose textual canon was "brought down"-by a figure vari- with the fluids produced in sexual intercourse. All three were, like the ma- ously called Srīnatha, Srīkantha, Ādinātha, or Siddhanātha-at a lunar trix from which they arose, soteriologies, individual paths to salvation from location called Candragiri, Candradvīpa, Candrapurī, or Candrapītha.137 the trammels of human existence,130 and all three were unique and inno- In one such canonical text, the Kubjikānityāhnikatilaka, we find an account vative inasmuch as they professed self-divinization-of three different that appears to dramatize the changes that were occuring on the Indian types-as the means to such transcendence. tantric landscape in this period. Here, we are told that Srinatha-with the The India that Buddhism left behind was in many ways tantric, and a aid of three Siddhas named Sun, Moon, and Fire-founded the tantric kula number of Buddhist elements find their way into Hindu tantrism in this tradition, "at the beginning of the Kali Yuga," at a site called Candrapurī period. These include a number of Buddhist deities, the four Buddhist ("Moon City"), located in Konkana, in coastal western India. The original pithas,131 various elements of Buddhist discourse and practice and, as we core of this new kula, the text continues, was composed of nine Näths who,

74 75 Chapter Three The Prebistory of Tantric Alchemy

originally Buddhist monks, had converted when, through a miracle pro- duced by Srinatha, the roof of their monastery had collapsed!138 We can also argue that by the tenth century, human individuals or or- ders known as Siddhas already existed in everything but name in India. In This "conversion," from Buddhism to Hinduism, carries much further the centuries that followed, these individuals and their real or alleged writ- in this context. First of all, a lineage found in the Grub thob, a ca. ings would come to be collated, anthologized, canonized, and institution- fourteenth-century Tibetan translation139 of Abhayadatta's eleventh- to alized into (1) Siddha lists, (2) a massive body of Siddha literature, and (3) twelfth-century Caturasitisiddha Pravrtti (Acts of the Eighty-four Siddhas), identifies two figures, Vyādi and Kubjika, as disciples of "Siddha Nāgār- a loosely structured confederation of Siddha religious orders. All of these

juna."140 As already noted, Vyādi is an alchemist (he is the last of the eighty- post-tenth-century enumerations, canonizations, and institutionalizations would look to the age of magical alchemy for their origins, and invariably four Siddhäcäryas); as we will see, he is an alchemist who figures promi- find them, in a name, a text, or a real or imagined lineage. The self- nently in Hindu lists of the Rasa Siddhas (the traditional founders of the proclaimed Siddhas of later history invented their past as a means of justi- alchemical tradition), lists that appear in a number of the major alchemical fying their own lives, works, and institutions: the "charismatic" super- tantras. Such also appears to be the case with Kubjika: the same name is heroes of the past-fast on their way to becoming "divinized" through an applied to a male Siddhācārya in Buddhist tantrism and to a female deity identification of them with the Siddhas and Vidyadharas of earlier cults- (Kubjikā) in Hindu tantrism. Whether it was the Hindus or the Buddhists were also becoming "institutionalized" into the sectarian founders they who had the first claim to these figures is a moot question; what is im- never had been.142 portant here is that both, like the Nagarjuna with whom they are associated As I demonstrate in the next chapter, historiographers of the medieval in this particular lineage, are pivotal figures in the Buddhist-Hindu tantric Siddha traditions have failed to come to terms with this fact, and have syncretism. With the departure of Buddhism from the subcontinent, the therefore always found themselves struggling to bridge unaccountable gaps alchemist Vyādi, the goddess Kubjikā, and the polymath Nāgārjuna be- of hundreds of years between the legendary "biographies" of many of these come transformed, in medieval India, into exclusively Hindu figures- Siddhas and their literary productions or institutional roles. Their efforts figures who are closely linked, moreover, to the history and mythology of have been in vain: just as new individuals named Nāgārjuna or Matsyendra tantric alchemy. They also become linked with figures-divine, symbolic, appeared on the Indian religious landscape once every one to two hundred or (super)human-identified as "Naths" or "Siddhas." years, so too entire founding lineages were created or reappropriated by self- But, just as was the case with tantric Buddhism and its Mahasiddhas or proclaimed Siddha traditions at several centuries' remove from the time in Siddhäcāryas, the Hindu application of the terms Nāth, Siddha, and Nāth which their founders may have lived.143 Siddha to semihistorical human figures is an ex post facto one. Once again, Before turning to the Hindu side of the Siddha phenomenon, I must it is not until the eleventh or twelfth century that we find these terms being attempt to account for the alchemical reputation of the Buddhist "Siddha applied, retrospectively, to such historical and legendary figures as the Nāgārjuna" for the period just discussed, the seventh to tenth centuries. Hindu Nāgārjuna, Vyādi, Goraksa, or Matsyendra. (While this last figure As with the case of the philosopher Nāgārjuna, there is little hard evidence defined the Siddha Kaula, and was retrospectively identified as the guru of to go on, beyond the body of legend bequeathed to us by Hsuan-tsang and Gorakhnäth, founder of the Näth Siddhas, he is nowhere referred to as others. In this light, it appears significant that the seventh-century Chinese "Matsyendra Siddha" or "Siddha Matsyendranāth" prior to the fourteenth scholar-pilgrim claimed to have met an alchemical disciple of Nāgārjuna century). I by no means wish to imply that such figures did not live during who looked to be only thirty, in spite of his seven hundred years. If he was the period I have been chronicling. There can be no doubt that historical indeed thirty, or slightly older, then we have grounds to believe that a re- figures named Nāgārjuna, Matsyendra, Goraksa, etc., defined the religious puted alchemist named Nagarjuna did in fact live in the first half of the landscape of a certain Gupta and post-Gupta India. It must, however, be seventh century allowed that these figures' lives, acts, and words became fused, in the popu- The Yogasataka, datable to either the seventh or ninth century and the lar imagination, with those of gods and demigods, in whose number the work of an "Ācārya Śrī Nāgārjuna," is mainly devoted to herbal, rather than divine Siddhas and Vidyādharas must be counted.141 mercurial, remedies, although some of its preparations are called "elixirs"

76 77 Chopter Three The Prebistory of Tantric Alchemy

(rasayanas) or "nectars" (amrtas).# This same Nagarjuna may have been of southern provenance, this in no way implies that it was of Tamil origin. (1) the editor of the Susruta Sambita responsible for the appendix entitled In fact, nearly all signs from this period point either to the Andhra country "Uttara Sthana" (although this section contains no innovative data on mer- (Srīparvata-Srisailam) or to western India as the cradle of the alchemical art curial medicines);1s (2) the figure singled out by the ninth-century Vrnda in India. It is, moreover, in these two regions that the so-called Māheśvara as having carved instructions for the ophthamological use of black mer- Siddhas begin to appear in the tenth century and that magical alchemy curic sulfide-a preparation he calls nāgārjuna-varta-into a pillar at Pa- became transmuted, as it were, into tantric and Siddha alchemy. țaliputra (Patna, the capital of modern-day Bihar);1+6 (3) an alchemist whom the eleventh-century Alberuni mentions as having lived some one hundred years prior to himself;147 and (4) the author of a work on metal- lurgy referred to as the Lobašāstra of Nāgārjuna by Cakrapāņidatta, in the ca. A.D. 1075 Cikitsasamgraba, a commentary on the Caraka Sambita and itself an Ayurvedic watershed in the use of mercurial medicines.148 This composite ninth-century Nāgārjuna would therefore have had to have been a figure distinct from the seventh-century Nalanda alchemist-monk, if one ever existed.149 To conclude, then, we can very tentatively identify three Nāgarjunas from what has been called the period of "magical" alchemy. These are (1) the ca. second-century A.D. Mädhyamika philosopher whose only alchemi- cal feats were mythical; (2) the early seventh-century disciple of Sarāha, and resident of Nalanda and Sriparvata, whose alchemical reputation reached the ears of Hsuan-tsang (who was himself the translator of two Mahayana texts containing references to magical alchemy), but who left no alchemical works for posterity; and (3) the ninth-century author of the Yogasataka and the appendix to the Susruta Sambita, about whom Alberuni heard reports in the early eleventh century. It is only some two to three centuries after Alberuni that we encounter an alchemical author named either Siddha or Sriman Nagarjuna in chapter colophons. These are three extant Hindu works, which shade in their content from tantric sorcery into alchemy-the Kaksaputa Tantra, the Yogaratnamala (also called the Ascaryaratnamala) and the Rasendra Mangala. Finally, I close this chapter on the age of magical alchemy with a note of caution to anyone who would attempt to interpret the general history of this period. This I do mainly as a corrective to south Indian historians of Siddha medicine who maintain that the Sittar alchemists had already perfected their craft long before the time of the sixth- to seventh-century Tirumūlar and Nandi.150 First, there is no evidence to show that any al- chemy-even as practiced by persons who may already have been calling themselves Siddhas-prior to the tenth century A.D. was anything but "magical." Second, while most of our data concerning magical alchemy is

79 Sources for the History of Tantric Alchemy

FOUR In the absence of anything comparable to a systematic theology (for which god would one choose?) or a religious anthropology, Puranic Hindu- ism lent much attention to what one scholar has termed "systematic geogra- Sources for the History of Tantric Alchemy in India phy," the elaboration of a spatial grid of the great chain of being, from the highest gods to the most abject demons, via the realms of humans, demigods, sages, and all manner of other creatures.+ In the tantric context, however, it is perhaps more proper to speak of "systematic genealogy" rather than sys- tematic geography or cosmology, given the fact that the cardinal directions to which the major clans of the tantric mandalas were assigned appear to bear no correlation whatsoever to the geographic locations of these sects or their I. Medieval Lists of Rasa Siddhas and Nath Siddhas founders. Rather, these were, in a quite literal sense, organizational "flow charts," in which the flow in question was of sexual fluids emanating first Among the most valuable historical documents we have at our disposal from the godhead and radiating outwards, through the biological transmis- for charting the period in which (1) "outer alchemy" disappeared from sion, via goddesses and yoginīs, of the clan nectar (kuladravya), into the vari- Buddhism, (2) Buddhism disappeared from India, and (3) Hindu Siddha ous sectarian clans, orders, and sects. Through their systematic genealogies, alchemy emerged out of a fusion of tantric alchemy with the discipline of the tantrikas at once located themselves within the cosmic chain of being and batba yoga are retrospectively compiled lists of Siddhas (both Buddhist and within a network of socioreligious institutions. Hindu) and Naths (exclusively Hindu), that proliferated throughout the The major tantric classificatory systems that will concern us here-in- Indo-Tibetan world between the eleventh and fifteenth centuries. It is in asmuch as certain names found in them correspond to names found in the these sources that the appellation Siddba comes to be used extensively, as a Siddha lists detailed below-are those of the pīthas and amnayas. While blanket term covering most of the figures-some historical, others legend- the former term is nearly always employed to denote a "footstool" or pil- ary, others divine, and still others clearly undecided-whose lives and acts grimage site of the Goddess in Sakta devotionalism, there also emerged, I review in these pages. around the time of the Brabmayāmala Tantra (ca. ninth century A.D.) a tra- While we cannot be certain that such was the case, the ordering of the dition of four "literary" pithas. Here, the major texts (together with the names in the Siddha lists at times appears to correspond to guruparam- practices and methods they espoused) of earlier tantrism were classified, paras, i.e., to the clan lineages1 of teachers and disciples through which the purportedly on the basis of the principal subjects they treated-but in fact mystic Siddha doctrines were transmitted. Here, the Siddha lists would rather artificially-under the headings of vidyā, mantra, mudra, and man- constitute simplifications of an earlier set of models. These were the ca. dala pithas. Most of the important texts of this classificatory system fell ninth-century A.D. idealized "post-scriptural systematizations"2 of the ma- under the rubrics of the vidyāpītha5 and mantrapītha. jor tantric sects and texts. These records often took the form of mystic Somewhat later,6 the classificatory system of the āmnāyas ("lines of diagrams (mandalas), in which the clans (kulas) of the divinities and legend- transmission") was devised to classify Kaula sects and texts. While sectar- ary preceptors of the major sectarian divisions of Hindu tantrism were set ian cleavages based on the name of the major cult goddesses figure most forth schematically. Such mandalas or yantras were at once divine and hu- prominently here, the theoretical classifications are geographical. The four man genealogies, ritual and meditational supports, and models of and for amnayas are the eastern (Trika, whose cult centers on the three Par god- microcosmic, mesocosmic, and macrocosmic reality, in which color, num- desses), northern (Krama, which worships Kali), western (which worships ber, direction, divine name, vital breath, activity of consciousness, sensory the goddess Kubjika, the "Contracted One," an alloform of the hathayogic organ, etc., were so many simultaneous proofs for the coherence of the kundalini), and southern (Srīvidya, which worships the goddess Trīpura- world system they charted: structure and function were congruent.3 sundarī) transmissions.7

78

80 81 Chapter Four Sources for the History of Tantric Alchemy While it is likely that the historical value of the Siddha lists is as limited 12. Tanti-pā as that of the mandalas of tantric pitha and amnāya classifications,8 a com- 16. Nāgārjuna parative overview will offer insights into what the medieval mystics of In- 17. Karāha-pā/Kānha-pā dia-Buddhists, Hindus, Jains (and later, Muslims), yogins and alchemists 18. Karņari-pā/Aryadeva alike, from Tibet to south India-considered to be their heritage. In the 30. Kambala-pā/Kamari-pā words of Giuseppe Tucci, the "Siddhas are the most eminent personalities of medieval India's esoterism and represent the ideal link between Sivaism 46. Jālandhara-pā 64. Carpațā-pā and Vajrayana, indeed the expression of the same religious and mystical 71. Kanthali-pā endeavor, translated through analogous symbols."9 These figures were al- ways first and foremost Siddhas, and it would be erroneous to maintain 72. Kapāla-pā

that the inclusion of a figure's name in a Buddhist Siddha list made him a 76. Nāgabodhi-pā 77. Dārika-pā Buddhist, or that a name figuring in a Rasa Siddha list necessarily made 84. Bhali-pā/Vyāli-pā that person an alchemist. The Siddhas, a pool of wizards and demigods, supermen and wonder-workers that all south Asians (and Tibetans) could In fact, the names of many of the figures in these lists are identical to draw on to slake the thirst of their religious imagination, were the most those of the authors of the earliest mystic poetry of "Buddhist" tantrism, syncretistic landmarks on the religious landscape of medieval India.10 the so-called Caryapadas13 (composed in Old Bengali in eastern India be- Our presentation of the medieval Siddha lists will, with a single excep- fore the twelfth century), as well as of a number of authors whose writings tion, be a chronological one: we will insert the list from the fifteenth- are found in the Sanskrit Sadhanamala, which dates from the same period.14 century Hathayogapradīpikā (HYP) of Svātmarāma before two fourteenth- Exceptions to this rule are (9) Goraksa, (10) Caurangi, (16) Nāgārjuna, (64) century lists (from the Varnaratnākara and the Ānandakanda [ĀK]) because Carpațā, (71) Kanthali, (72) Kapāla, (76) Nāgabodhi, and (84) Vyāli, whose its content corresponds more closely to that of earlier eleventh- to "signatures" are found in neither the Caryāpadas nor the Sādhanamālā (Nā- thirteenth-century lists than do the latter. We will begin our survey with gārjuna is the author of two sadbanas in this latter work). A later Tibetan the Indo-Tibetan lists of the eighty-four Mahasiddhas or Siddhacaryas. recension of the carya songs, incorporated into the Tanjur; includes works Two nearly identical lists are found in the Grub thob,11 a mid-fourteenth- by Ghoraksa (sic), Caurangi, and Carpațā/Carpati not found in the origi- century Tibetan text based on a ca. eleventh- to twelfth-century Sanskrit nal anthology.15 work attributed to the Indian Abhayadatta; and the A.D. 1275 Sa-skya-bka'- Six of these names-Caurańgi, Nāgārjuna, Carpați, Kapāla, Nāga- 'bum,12 which purports to be a list of the gurus who taught at the Saskya bodhi, and Vyāli-figure in another enumeration of Siddhas. These are the monastery between A.D. 1091 and 1275. In these lists, the -pā endings to lists of the twenty-seven Rasa Siddhas, as such are found in three alchemical many names are shortenings of -päda: these endings, rather than the term works.16 Two of these-a list found in the thirteenth- to fourteenth- Siddha, are the authentic mark of the eastern Indian figures and lineages century Rasendra Mangala [RM] of Śrīman Nāgārjuna17 and a list in the that founded tantric Buddhism. Following what has become a scholarly fourteenth-century Rasaratnasamucchaya [RRS] (1.2-4) [a] of Vāgbhatța convention, we will assign numbers to these figures, in accordance with II18-are nearly identical; whereas a second list, found in both the their ordinal placement in these various lists: thirteenth-century Rasaratnākara [RRĀ] (3.1.66-69) of Nityanātha Sid- dha19 and the RRS (6.51-53) [b], differs slightly from the first. Further- I. Luī-pā/Matsyendra more, certain recensions of the RRS supply as many as forty additional 8. Mīna-pā names (following 1.5) to the original list of twenty-seven: we will call this 9. Goraksa-pā supplementary list RRS [c].20 Figures whose names have already appeared 10. Caurangi-pā above, in the Tibetan lists, are given here in italics.

83

Chapter Four Sources for the History of Tantric Alcbemy

RM/RRS[a] RRA/RRS[b] ken Time's rod" found in the HYP (1.5-9). This list is nearly identical to a

r. Ādima/Āduma2 list drawn up by Caturbhujamisra, a 15th- to 16th-century commentator

2. Candrasena 2, Candrasena on the RHT, with the difference that the latter identifies the twenty-four

3. Lańkeśa 3. Lańkeśa names on his list as Rasa Siddhas.24 As before, names occurring in Tibetan

4. Visārada 22. Sārada lists are italicized; those already noted in RRS [b] have their numbering

5. Kapāla/Kapālī from that list shown in parentheses.

6. Mata/Matta RRS 1.5-10 [c] RRS 1.5-7 [a] HYP 1.5-9 7. Māndavya 10. Māndavya 8. Bhāskara 28. Rasānkuśa 28. Rasānkuśa"

7. Surasena/Sūrasenaka 12. Sūrasena 29. Bhairava z9. Bhairava 4. Bhairava

ro. Ratnaghoša/Ratnakoša 6. Ratnaghośa 30. Nandi 30. Nandi

11. Sambhu 20. Sambhu[rloka] 31. Svacchandabhairava 31. Svacchandabhairava

12. Ekaratna/Sāttvika 19. Tattvika 32. Manthānabhairava 32. Manthānabhairava 10. Manthānabhairava

13. Vāhana 4. Naravāhana 33. Kākacaņdīšvara 33. Kākacaņdīśvara 23. Kākacandīśvara

14. Indra[da] 34. Mahadeva 40. Mahadeva

15. Gomukha 35. Narendra 41. Narendra 28. Naradeva

16. Kambali22 36. Ratnākara

17. Vyādi 1. Vyālācārya 37. Harīśvara 43. Harīśvara

18. Nāgărjuna 5. Nāgarjuna 38. Korandaka 13. Korantaka

19. Surāvida/Sūrānanda 7. Surānanda 39. Siddhabuddha 39. Maithila[hvaya] 11. Siddhirbuddha

20. Nagabodbi 14. Nagabuddhi 40. Siddhapāda 15. Siddhipāda

21. Yaśodhara/Yasodhana 8. Yasodhara 41. Kanthadi 12. Kanthadi

22. Khanda 15. Khanda 42. Rsyaśrńga 43. Vāsudeva 35. Rsi Srnga

23. Kāpālika 16. Kāpālika 42. Vāsudeva

24. Brahma/Brahmā 43. Kriyātantrasamucchayī 36. Kriyātantrasamucchayī

25. Govinda 25. Govinda 44. Rasendratilaka 37. Rasendratilaka Yogi

26. Lampaka 21. Lampața 45. Bhanukarmerita 38. Bhāluki 27. Bhāluka

27. Hari 17. Hara 46. Pūjyapāda 18. Pūjyapāda 47. Kaneri 48. Nityanātha 17. Kanerī Nine names included in the RRA/RRS [b] list do not accord with those 49. Nirañjana 19. Nityanātha found in the RM/RRS [a] list. They are: (3) Subuddhi; (9) Indradyumna; 20. Niranjana (11) Carpati; (13) Vādava/Agama;23 (18) Kāmāri/Kāmalī; (23) Bānāsura; 50. Carpati (11) 16. Carpati (24) Muniśrestha; (26) Kapila; and (27) Bali. All of these names, however, 5t. Bindunatha 12. Bindunatha figure among the forty additional Rasa Siddhas listed in RRS [c]. We now 52. Prabhudeva 24.[Allama] Prabhu present the lists of these additional Rasa Siddhas as they occur in the 53. Vallabha continuation of RRS [a] (1.5-7) and RRS [c], names which, as Vagbhatta II 54. Vālākhilya-jnā [Bālākiryanjā] explains, are "names of authors of other alchemical texts." To these, we 55. Ghodacoli 25. Ghodācoli juxtapose data from another list, that of thirty mabsiddhas "who had bro- 56. Tintani 26. Tintini

84 85 Chapter Four Sources for the History of Tantric Alchemy

RRS 1.5-10 [c] RRS 1.5-7 [a] HYP 1.5-9 with Matsyendranäth. The relationship here between Goraksanatha and his guru Minanātha in the two Hindu works listed here is symmetrical to 57. Kālācārya 58. Subuddhi (3) that obtaining between the figures Goraksa-pā (9) and Mīna-pā (8) in the

59. Ratnaghośa (6) Tibetan Grub thob. This squares, moreover, with the statement made by

60. Susenaka Goraksa in the second verse of his earliest work, the Goraksasataka, that his

61. Indradyumna (9) guru was Minanatha. It is a later conflation (as in the 15th-century "Sidh

62. Āgama (13) Vandanam," below) that identifies Matsyendra (the second Siddha, listed

63. Kāmāri (18) after Adinatha, in the HYP list) or Lui-pa (the first Siddha of the Grub thob

64. Văņāsura (23) list) with Mina. The identification is a simple one: all three of these names

65. Muniśrestha (24) include the term fish. As I will demonstrate, the historical Goraksa or Gor-

66. Kapila (26) akhnäth was a 12th- to 13th-century figure, whereas Matsyendra, his pur-

67. Vali (27) ported guru, could not have lived later than the 1oth century; it is another figure, named Mīna, who was Gorakhnāth's guru. Apart from its parallels with the RRS lists, the HYP list also appears to The final two lists to be reproduced here also begin-both logically and borrow a number of its Siddhas from the Tibetan Grub thob. To these paral- chronologically-with Ädinätha, the Lord of Beginnings. The first of lel lists we also juxtapose data from yet another source, the ca. A.D. 1300 these is the 14th-century Anandakanda [AK] (1.3.47-51),27 certainly the Varņaratnākara (VRĀ)26 list of seventy-six Siddhas: longest extant alchemical text, and also the most convoluted in many ways. Behind its complexity, however, we may glimpse a number of Siddha lin- HYP 1.5-9 Grub thob VRÅ 7 eages which it was attempting to decant into a single authoritative list. Of

68. Matsyendra I. Luï-pā/Matsyendra 41. Mīna exceptional interest is the fact that it divides its list of twenty-five names

69. Sābarānanda 5. Sābara-pā 43. Savara into a set of "Nine Naths" and "Sixteen Siddhas." In this, as well as in a

70. Caurangi 10. Caurangi-pā 3. Cauranginātha number of other unique hathayogic elements found in this, but no other

71. Mīnanātha 8. Mīna-pā 1. Mīnanātha alchemical source, we find ourselves in the presence of a truly syncretic

72. Gorakhnāth 9. Goraksa-pā 2. Goraksanātha Siddha alchemical tradition. The data it contains will also lead us, in a mo-

73. Carpați 64. Carpață-pă 31. Carpati ment, into a second corpus, the myriad lists of the Nine Näths. The list we

74. Kānha-pā 13. Kānha will juxtapose here with the AK is quite unusual, inasmuch as it purports

75. Kapālī II. Kapāli to constitute a list of the "eighty-four Siddhas"28 and "nine yogesvaras

76. Tintini 13. Tanti-pā 5. Tanti-pā (Lords of Yoga)"-but which in fact only lists fifty-six names. This vernac-

77. Nāgārjuna 22. Nāgārjuna ular work, entitled "Sidh Vandanām" [SV]29 (Praise of the Siddhas) is rela-

78. Kamari-pā 12. Kamārī tively late, given that it mentions the 15th-century Kabir as its fiftieth Sid-

79. Jālandhara-pā 19. Jālandhara dha. For purposes of comparison, we juxtapose elements from the HYP list

80. Dārika 9. Dāri-pā with those of the AK and the SV. We will list the Nine Naths and sixteen Siddhas of the AK list continuously, numbering from 1 to 25. Italics refer The Varnaratnakara further accords with various Rasa Siddha lists in its to names that also appear in Rasa Siddha lists from earlier sources: mention of Carpati, Kapāli, Nāgārjuna, Kamārī-listed above-as well as of (71) Govinda (no. 25 in three lists) and (73) Bhairava (no. 29 in two ĀK 1.3.90-93 SV 1-27 HYP 1.5-9 lists). The (1) Adima of the RM/RRS [a] list is to be identified with the (1) 1. Ādi-nātha 1. Ādināth I. Ādinātha Ādinātha of HYP 1.5: ādi denotes origins; Ādinātha is therefore the original 2. Mīnanātha/Mūlanātha 2. Machindra 6. Mīnanātha and divine Rasa Siddha or Mahasiddha: as such he is sometimes identified 3. Goraksa 3. Goraksa 7. Goraksa

86 87 Chapter Four Sources for the History of Tantric Alchemy

AK 1.3.90-93 SV 1-27 HYP 1.5-9 located on the northern outskirts of Mangalore (Karnataka)-would have

4. Końkaneśvara lived in the early part of that century.31 According to Maharashtran tradi-

5. Jālāndhreśa 1O. Jalandharī tions dating from at least the final quarter of the thirteenth century, Revana

6. Kanthanīśa 28. Kanthad-pāy 11. Kanthadī (21) is listed as one of the Nine Nths. Elsewhere, the Virasaivas, who

7. Uddīśa have always been concentrated in southwestern India, revere Revana (or

8. Ciñc[hļinīśvara Renuka) Siddha as one of their five great founders.32 Uddiśa (7) is the name

9. Caurańgi 27. Caurangi 5. Caurangi of a work on tantric sorcery33; and Kukkura-păda (22) is a Mahāsiddha (no.

I0. Caurangi30 34) from the Grub thob list.

11. Carpata 4. Carpața-rāy 16. Carpați Other names on this list are found to correspond to those of certain of

12. Ghodacūli 26. Ghodācolī the eighteen Sittars, the alchemist-physicians of Tamil Nadu.34 The Sittar

13. Rāmadvaya lists-of which there exist at least eighteen variants-are infuriatingly het-

14. Bhola/Bāla 14. Bālanāth erogeneous. No two are alike, and any one is likely to include names of

15. Govinda-siddha Vedic sages (Agastya, Loma Rsi), demons (Pulastya), and nymphs (Urvasī),

16. Vyādi as well as classical Hindu gods (Rāma, Daksinamūrti, Subrahmanian), the

17. Nāgārjuna divine or semidivine founders of Āyurveda (Dhanvantari), yoga (Patañjali),

18. Koranda 16. Kodaņda Advaita Vedanta (Sankaracarya), and south Indian orthodox Saivism

19. Śūrpakarņa/Šūrpadhuņdhī (Idaikkādar, Pambatti, Auvai, Manika Vachagar, etc.),35 and the odd Euro-

20. Muktayī/Duktayī pean (Yacoppu, i.e., Jacob). In addition to these, however, we also en- 21. Revana/Renava counter names found in the Anandakanda list: Tirumūlar (whom the Nath

21. Kukkurapāda Siddhas call Mūlanatha [2a]), Macchendrar (= Matsyendranatha [2b]);

22. Šūrpapāda Korakkar Nādhar (= Gorakhnāth [3]), and Konganar (Konkaneśvara [4]).

23. Kaņairika 23. Kānerī 17. Kanerī One exemplary list of the eighteen Tamil Siddhas36 enumerates the follow-

24. Kinkinika/Tintini 26. Tantanī 27. Tintini ing (names from earlier lists are italicized): Nandi (1), Mūlanātha (2), Kā- 30. Kapālī 21. Kapālī lānganātha (3), Bhoga (4), Konkana (5), Agastya (6), Pulastya (7), Bhuśunda 32. Kāg Cand 23. Kākacaņdeśvarī (8), Romamuni (9), Dhanvantari (10), Sattaimuni (11), Matsyamuni (12), 33. Kāg Bhuśand Kańva (13), Pidinākkīśa (14), Goraksa (15), Terayar (16), Yūhimuni (17), 34. Sūranand 14. Sūrānanda and Idaikkādar (18).

25. Bhairū Nand 4. Bhairava The last chapter discussed alchemical lore on the subject of two other 36. Sāvarā Nand 3. Sāvarānanda recurring names from the Tamil lists, Bogar and Nandi; as was noted, Kon- ganar was the disciple of one of these two alchemical gurus.37 Also worth noting here are the Sittars named Kambali Siddhar (number 15 in the RM/

26. The Anandakanda List and Medieval Western Indian Saivism RRS [a] list); Navanadhar (navanatha, the "Nine Naths"), and Sattainathar (Satyanātha). This last figure, whom Tamil traditions credit with having authored twenty-one alchemical works, is identified by the Nath Siddhas A number of apparently incongruous names in the AK list may be shown with the historical founder of one of its twelve (modern) subdivisions, as to correspond to names of Nath Siddhas found in southwestern Indian sources. The names Sūrpakarna ("Shovel Ears") (19) and Sūrpapāda well as with a yogic incarnation of the god Brahmā.38 Ādināth, Mūlanāth, Konkaņeśvara (4), and Ciñcinīśvara (8) belong to ("Shovel Feet") (23) resemble that of Surpanäth who-according to a another sectarian tradition that interacted with the Nath Siddhas and Rasa thirteenth-century inscription from the Nath Siddha monastery of Kadri, Siddhas throughout this period. This is the paścimamnaya, the Western

88 89 Chapter Four Sources for the History of Tantric Alchemy Transmission. To begin, we find lists in this tradition's fourteenth-century rates Matsyendra's Yogini Kaula cult into the secret ritual of the Trika Kubjikānityahnikatilaka and twelfth-century Manthanabhairava Tantra that Kaula virtuosi.+5 Elsewhere, Abhinavagupta calls Macchanda the great are structurally identical to that of the AK: that is, they are subdivided into Kaula master of his age, the kali yuga; and, following earlier scriptural a set of nine and a set of sixteen of teachers. Two consecutive chapters in tradition, places Macchanda, together with his consort Konkana, in the the SSS, itself an expansion on the Kubjikāmata (KM), treat of the Kubjikā northern quadrant of the Siddha Cakra.16 In this way as well, Matsyendra revelations that were brought down to earth by Ādināth (1) and Mūlanāth may be identified with Konkaneśvara (4): he is the lord (isvara) of his con- (2). That these sources are in some way related is further supported by sort Końkaņā. the fact that the Konkana in Konkan-esvara-the "Lord of Konkana"-is Now, it happens that both the scriptures of the Western Transmission47 precisely the (western) coastal region in which the Western Transmission and the early works of Matsyendranäth have, for centuries, been extant in claims to have been founded, through the agency of a group called the Nepal alone. How is it that revelations transmitted at a "Moon Island" "Nine Naths."39 In a list of the Nine Naths that probably originates from (Candradvīpa) or "Moon Hill" (Candragiri) either on the west (the West- the Deccan, one of the group is named Konkananath.10 ern Transmission's Konkana) or east (the Yoginī Kaula's Kāmarūpa) coast The name Ciñcinīśvara (8) evokes the title of another important West- of India should now find themselves in a landlocked Himalayan valley? An ern Transmission work: this is the Nepali Cincinimatasārasamucchaya, abundance of local legends explain Matsyendra's "transfer" from Kāma- which states in its first verse that the cult of Cincini was founded by a figure rūpa to Nepal.48 As for the latter toponym, a geographical explanation is named Siddhanātha. Cincinī ("tamarind") is moreover identified with possible: Candragiri is the name of a hill located at the western end of the Kubjikā, the goddess whose cult lies at the heart of the Western Transmis- Kathmandu Valley which was, according to myth, originally flooded, prior sion.41 Siddhanātha is, according to a number of texts, an alternative name to being drained by Mañjuśrī: this western Moon Mountain would origi- for Matsyendranath, whose name is also cited in the Kubjikānityāhnikati- nally have been a Moon Island .* Another probable explanation is of a his- laka as one of the appellations of its fifth teacher.42 torical order. There exist significant medieval links between southwestern Inasmuch as he stands as the prototypical Siddha, I return to the pivotal and eastern India on the one hand, and Nepal on the other, by which the figure of Matsyendra throughout this and the next chapter. In addition to transplantation of these cults can be explained. From the early twelfth cen- having a role within the Western Transmission, Matsyendra is also consid- tury onwards, the Kathmandu Valley was penetrated first by religious ered to have founded a number of other tantric clans. More than this, he proponents of south Indian Saivism and subsequently by kings from the is at once revered as the superhuman intermediary who transmitted-or Deccan region. Following the fourteenth-century Malla restoration, "brought down" (avatrita) from the divine to the human plane-the orig- priests from Bengal and the Deccan officiated in temples in the city of inal tantric revelation and as the human reformer responsible for having Bhaktapur, at the shrine of Pasupatinath, and elsewhere in the Kathmandu transformed the old Kula tradition into "reformed" Kaulism." As such, his Valley.50 A number of Nepali religious institutions and traditions reflect principal teaching, the KJnN, is a watershed for all manner of tantric and this historical connection, down to the present day.51 A third, esoteric, ex- Siddha traditions. In this work, Matsyendra describes himself as an adher- planation for these toponyms will be discussed in chapter eight. ent of the sectarian Siddha Kaula, the founder of the Yogini Kaula, and Nepal's relative isolation has also fostered the preservation of forms of revealer of the (hathayogic) doctrine of Matsyodara ("Fish Belly"). The Hindu tantrism that have long since disappeared from India itself. The KJnN is, in fact, one of the earliest Indian texts to make wide and general Kathmandu Valley remains a "living museum" of both Hindu and Buddhist use of the term "Siddha"; moreover, it anticipates the batha yoga of later tantrism, in which carya songs belonging to the same tradition as that of Näth Siddhas (who also claim Matsyendra as their founder) and contains a the seventh- through twelfth-century Siddhācaryas continue to be sung at wide array of original data on tantric theory and practice. religious performances, where the old Western Transmission cult of Kub- Abhinavagupta singles out "Macchanda Vibhu" (Matsyendra) for praise jikā continues to be observed,52 and where hordes of tantric Bhairavas and in the opening lines of his monumental A.D. 1000 Tantrāloka* and incorpo- Kālīs continue to receive blood sacrifices on a massive scale.

90 QI

Chapter Four Sources for the History of Tantric Alchemy

3. Excursus: The Nine Naths, the Twelve Panths, ras?], lord of the Kaulas and leader of the Kula" (sadpuradbipatirnathah kau- and the Historical Gorakhnäth lisah kulanāyakab).53 We have already mentioned a later text belonging to the same tradition: this is the Kubjikanityahnikatilaka, which relates that At this point, it becomes necessary, for historiographical reasons, that the clan was founded, in Konkana, by Srīnatha, who was accompanied by we trace the origins of the Nath sampradaya or Näth panth, the institu- "twice nine Naths." He emanated into three Siddhas, who had nine sons tionalized religious order or sect of the Näth Siddhas. The reason for whom they made their disciples; from these originated the sixteen dis- this is simple. It is entirely impossible to make historical sense of the Nath ciples.54 Two later works place an equally allegorical gloss on the Nine Siddhas if we fail to separate pre-thirteenth-century legend from post- Naths: the Tantraraja Tantra correlates the Nine Naths with the nine thirteenth-century historical fact. Armed with this interpretive tool, we bodily orifices,55 while the Tantra Mabarnava identifies eight Nath Siddhas will be able to neatly bypass such burning nonissues as whether Gorakh- with the eight cardinal directions. This latter list places Adinath at the näth was originally a Buddhist living in eastern India and at the same time "center" of the Indian universe, in the land of Kuruksetra; a location that dovetail historical and geographical evidence for a western Indian origin would correspond to a widespread Kaula doctrine, which calls the central of the Nath Siddhas with data concerning the Rasa Siddhas, such as that pītha of its mystic universe the Adipitha.56 According to the Western Trans- found in the AK list. mission, the supreme goddess (Parā, Kubjikā) dwells in a pītha of the same The point I wish to make here is that the Nāth sampradāya, a great medi- name: radiating out from her are eight other such centers, at each of which eval changing house for western Indian sectarian Saivism, could not have is found a Nath through whom she generates her empowering energy.57 emerged prior to the late twelfth to early thirteenth century A.D. I will Here, the Nine Naths appear to be nothing other than a nonuplication show that this was not a monolithic order, but rather a confederation of of Matsyendranäth: they are the superhuman intermediaries who brought groups claiming a similar body of Saiva and Siddha tradition, the basis for the Western Transmission teachings down to earth.58 In the SSS and the whose unity was and remains (1) the identification of the twelfth- through Samvartārthaprakāśa, two other works of the Western Transmission, this thirteenth-century historical Gorakhnäth with earlier historical, legend- group of nine (abstract) Näths is doubled into a set of eighteen Näths or ary, or divine figures named Gorakh or Goraksa; (2) the retrospective asso- Siddhas. In this light, it is noteworthy that Tamil tradition holds that there ciation of this or some earlier Gorakh-often in the role of guru or dis- were originally nine Sittars, before their number was doubled to eigh- ciple-with founders of other Siddha or Saiva sects and clans, many of teen.59 Finally, the abstract or symbolic becomes merged with the historical which came to be absorbed into the Nāth sampradāya, (3) the transforma- in the vernacular Gorakb Upanisad, whose list of the Nine Naths begins tion of the abstract concept of the Nine Näths into a number of quasi- with seven stages in an emanatory schema, followed by a set of eight "his- historical lineages; and (4) the continued appropriation, in later centuries, torical" Nath Siddhas, of which Matsyendra is the first and Gorakh the by groups outside the Nath sampradaya, of the names Gorakh and other of last.60 the Nine Näths as a means to integrating themselves into that order. When, however, the aggregate of the Nine Näths came to be identified, We begin our discussion by undertaking a conceptual archaeology of some time in the twelfth to thirteenth century, with the legendary or his- the Nine Naths (navanatha). To date, no scholar has ever doubted that torical founders of the Nath sampradaya, the "historical" name that headed these were ever anything but a group of legendary, if not historical figures. most lists was that of Matsyendra. If, as many have done,61 we conflate But what we find when we look at the Nine Naths historically is that they Matsyendra[-nāth] with Mīna[-nath] and Luī[-pā] (all three names can be were, at least a century prior to their transformation into historical lin- loosely taken to mean "Lord of Fishes"), we find that it is he whose name eages, an abstract or symbolic category belonging to the Western Trans- figures the most frequently (together with Goraksa, Carpati, and Nāgār- mission. The earliest veiled reference that we find to this group is found juna) in the Siddha lists we have compared to this point (six times). Next in the Kubjikamata which calls a (presumably divine) Nath the "lord of the in frequency are Caurangi, Vyādi, Ādināth, Tanți / Tiņțini, Karņarī/ nine cakras" (nātho navacakresvarab), "founder of all the Siddha [lineages]" Kanerī, and Bhairava (five times), followed by Jālandhara, Kanthadī, and (sarvasiddhānām . .. samvyavasthitab), and "governer of the six cities [cak- Govinda (four times). Of these, the following names appear in all three

92 93 Chapter Four Sources for the History of Tantric Alchemy

types of Siddha lists-i.e., of the Indo-Tibetan Siddhācāryas, Indian Ma- founder of the Jogi panth. His descendants were two demons, Rudragan hāsiddha yogins, and Indian Rasa Siddhas: Matsyendra-Mīna, Goraksa, and Jalandhar. Descended from Jalandhar, the demon convert and initiate, Nāgārjuna, Carpați, Caurangi, Tanti / Tiņtini, Karņarī / Kaņerī, and Kan- were Matsyendra (3) and Jālandharī-pā (4). Matsyendra was the guru of thadī. In fact, many of these names correspond to most frequently ap- Gorakh (5), Arjan Naga (6), and the father and guru of the two Jain pearing names in the myriad lists of the nine "historical" Nath Siddhas. tīrthamkaras Nīmnāth and Pārasnāth (7). Jālandharī-pā's disciples were The most commonly occurring names in such lists are (1) Matsyendra, Bārtrināth (Bhartrhari) (8) and Kāņipā (Kānha-pā) (9).67 (2) Gorakh, (3) Carpati, (4) Jālandhara, (5) Kanerī, and (6) Caurangi. Also figuring in many lists are (7) Nāgārjuna, (8) Bhartrhari, and (9) Gopīcand Bengal Maharashtra Andhra Punjab (Govinda-candra, who is never identified with Govinda the alchemist). I. Ādi Matsyendra Śivanāth Śiva There exist a seemingly infinite number of permutations on this set of 2. Mīna Jālandhara Mīna Ude (Udaya) nine, with historical and subsectarian variants abounding. Ādināth is often 3. Jālandhari Gorakh Sārangdhara68 Matsyendra added to head the list, in the role of creator god or demiurge: this is a 4. Gorakh Carpața Goraksa Jālandharī-pā tradition that dates back at least to the Kularnava Tantra (6.63), which 5. Mayanāmatī Revana Meghnād Gorakh places him at the head of the divine stream (divyaugha) in its enumeration 6. Kānha-pā Kariņa Nāgārjuna Arjan Nāga of the founding Kaula gurus.62 Furthermore, there are regional variations 7. Gopīcand Bhartrhari Siddhabuddhū Nīm-/Pāras-nāth in the name employed for any given Näth Siddha: just as Matsyendra can 8. Bāil Bhādāi Gopīcand Virūpāksa Bārtrināth become Mīna; so Nāgārjuna can become Nāgnāth or Arjun Nāga; Kaņerī 9. Gahani Kaņika Kāņipā either Karnarī, Krsnapāda, or Kānha; Gorakh Goraksa; Carpati either Kar- pați or Carpața; Caurangi Pūran Bhagat; and Jālandhara Hādi-pā. By way Elsewhere, multiple lists of names of the founders of the twelve "offi- of illustrating such variants, I offer four lists of the nine founding Naths, cial" subdivisions of the Näth order (barah panth)-generally disciples of all of which date from the thirteenth through sixteenth centuries. In Ben- either Matsyendra or Gorakh-often include Bhartrhari, Gopīcand, Caur- gal, two sources dating from the seventeenth century but harking back to angi, Nāgnāth, and Satyanāth (or Satnāth).69 Once again, the lineages of earlier traditions63 provide the following data (in which the ninth Nath is masters and disciples (guruparamparās) of these subdivisions are useful lacking): (1) Adinath (identified with Siva) is the founder of the order; his in dating the origins of the Nath Siddhas as a structured religious order two disciples are (2) Mīnanāth (Matsyendranāth) and (3) Jālandhari-pā (sampradaya), as opposed to an idealized list of names, a body of legend, or (Hādi Siddha); (4) Gorakhnāth is the disciple of Minanāth; (5) Queen a corpus of literature. All of these data point to a twelfth- to thirteenth- Mayanāmatī the disciple of Gorakhnāth; (6) Kānha-pā and (7) Gopīcand century western Indian watershed, which witnessed a major realignment the disciples of Jalandharī-pa; and Bāil Bhādāi (8) the disciple of Kānha- of a number of preexisting Saiva religious orders, the creation of a number pā. According to a thirteenth- to fourteenth-century Maharashtran of new orders, and the appearance of a "canon" of literature on the tech- source,64 the divine founder of the clan is (Visnu-)Dattatreya, who is the nical and experiential aspects of hatha yoga. This historical convergence guru of (1) Matsyendranāth and (2) Jvalendra (Jālandharanāth). Matsyen- lends credence to the notion that a historical figure, named Gorakhnāth, dra is the guru of (3) Gorakhnāth, (4) Carpati, and (5) Revana; while Jva- actually founded a new religious order-the Nath sampradāya-in this lendra is the guru of (6) Kariņa-pā (Karņarī-pā); (7) Bhartrhari and (8) period. Gopīcand. Gorakhnāth is the guru of (9) Gahaninath. The third list is By way of demonstration, let us look at the suborder of the Nath sam- found in a ca. A.D. 1400 Telugu text from Andhra Pradesh, the Navanātha- pradāya known as the Vairāg panth, which was, according to tradition, caritra of Gaurana, which was written at the behest of the abbot of a Saiva founded by Bhartrhari, a thirteenth-century disciple of Gorakhnäth.70 The monastery at Srisailam.65 The fourth list is adapted from a ca. sixteenth- current head of this suborder is a householder Nāth named Nārāyan Nāth, century parampara recorded in the Punjab,66 which gives the following suc- whose family compound, in a village named Padu Kala (Meerta City tehsil, cession: Sakti and Siva (1) initiated Ude (2), second of the Nine Näths and Nagaur district, Rajasthan) is attached to a temple containing a dozen

94 95 Chapter Four Sources for the History of Tantric Alchemy

extremely worn burial tumuli (samādbis). These are, according to Nārāyan who in 1624 visited "the famous hermitage of Cadiri" to see the "Batinate Näth, the tumuli of Bhartrhari's immediate disciples, as well as of certain [panthi nath?], called King of the Gioghi [Yogis]."75 It was also on this coast of the twenty-six other Nath Siddhas-most of whom he could name from that Marco Polo encountered his "Ciugi" alchemists, in or around the memory-who had intervened between Bhartrhari and himself as titular year 1295. leaders of that subsect. If these samadhis do indeed contain the bodily re- While we can be certain that a historical Matsyendranath lived and mains of seven centuries of Näth Siddhas (and they certainly appear to be wrote at least two centuries before the symbolic Nine Näths begin to ap- very ancient), they constitute stronger evidence for the actual institutional pear in twelfth-century Western Transmission sources, the notion of nine existence of this sect than do the lists, legends, and even writings of its historical figures who, taken as an aggregate, constituted the Nine Naths purported founders that have come down to us.71 probably did not solidify until the foundation, by Gorakhnath, of the Nath Another piece of archaeological evidence, also from the thirteenth cen- sampradāya as an institutionalized religious order, in the twelfth to thir- tury, is an A.D. 1279 inscription from Karnataka (Jagalur tāluka), which teenth century. Additional datable evidence from this period includes the records the donation of a village by a general of the Yadava king Rāmacan- Maharashtran Jñāneśvara's ca. A.D. 1290 reference to Gorakhnāth as his dra to a "yogi world-conquerer" (yogi cakravarti) named Prasāda Deva. In guru's guru's guru;76 and a reference in Visobakhecara's thirteenth-century this inscription, we find what appears to be an enumeration of Prasāda Ghatasthala to the founders of his lineage as Ādināth, Mīnanāth, and Deva's lineage: "Ādigadedunātha Caturaraginātha Gorakhanātha Vistāra- Gorakhnäth. This proof is also supported, apophatically, by further evi- devī ... nātha Kāhaļinavi [Kanthadī?] Šurppāņanātha Lonanātha Naranā- dence from Maharashtra. Less than a century prior to Jñāneśvara, Mu- tha pantha." Although no one of the twelve modern subsects of the Nath kuņdarāja, the first poet to write in the Marathi medium, gives an account sampradāya appears to ever have been based in Karnataka, the ancient and of his lineage in his Vivekasindhu. Adinatha, who founded the lineage, had prestigious monastery of Kadri, located on "Jogi Hill" on the northern Harinātha (A.D. 1159-86) for his disciple. Harinātha's disciples were outskirts of Mangalore, is a permanent fixture of Nathdom, and it is the Surendranātha, Bhāratīnātha, Janārdana, Nāgnātha, Viśvambharnātha, and founding lineage of the abbots of this monastery that is recorded in this Raghunatha-with the last of these being Mukundarāja's guru.77 Here, we inscription.72 Located in the monastery precincts at Kadri is a certain Mañ- are presented with a lineage founded by Adinatha, and in which a large junātha temple, whose principal image, of Lokanātha, was dedicated in number of individuals whose names end in -natba are figured, but from A.D. 968. In the inscription that records that dedication, it is stated that whose number Gorakhnath is conspicuously absent. Another early "the image of the god Lokeśvara [was placed] in the beautiful vibāra of Marathi-language work, the twelfth- to thirteenth-century Lila Caritra of Kadirikā." Now, vibāra is a term that can refer only to a Buddbist mona- Cakradhara Svāmī, founder of the Mahānubhava sampradāya, also lists stery; this, in conjunction with a number of other indices (Lokanātha- number of familiar Näth Siddhas, with the significant omission of Gorakh- Lokeśvara's three-faced image conforms with northern descriptions of the nāth.78 Gorakhnath's appearance, in Jnanesvara's A.D. 1290 Jnāneśvarī, as Buddhist Avalokiteśvara, etc.), indicates that coastal Karnataka and Kadri the disciple of a semidivine Matsyendra and as Jnanesvara's guru's guru's were Buddhist until shortly before the end of the tenth century. All of this guru, indicates the appearance of the historical reformer Gorakhnath on is most interesting, in the context of the ca. twelfth-century Western the western Indian scene in the late twelfth to early thirteenth century. Transmission legend, related above, which stated that the "Nine Naths" of Additional evidence of the same order may be elicited from an A.D. 1030 Konkana converted from Buddhism to Hinduism after the collapse of their inscription from Karnataka (Nelamangala tāluka) which commemorates monastery,73 and in light of the fact that Nath Siddha traditions at Kadri the founding of a Siddhesvara ("Lord of the Siddhas") temple there. While maintain that the Mañjunatha appeared, in the form of the three-faced im- this inscription names the Candrapuri ("Moon City") of Western Trans- age of Lokanātha, to the three Jogi gurus Gorakhnāth, Matsyendranāth, mission tradition, a founding guru named Adinatha, and a number of other and Sarnganath. Shrines to these three founders enclose the vibara-temple figures with names ending in -natha, as well as the orthodox Saiva textual of Mañjunātha.74 The Kadri monastery is also the venue for a number of corpus (saivagama), it makes no reference to Gorakh or any one of the Nath legends, and is mentioned by the Italian traveler Pietro della Valle, "authorized" nine founders the Nāth Siddhas:79

96 97 Chapter Four Sources for the History of Tantric Alchemy

At the foot of a wonderful tree in Candrapuri, [which is] situated in the Western Ocean, Adinatha is installed. By merely recalling his A significant volume of evidence, then-archaeological, literary, and in-

excellent lotus feet, the residual effects of acts commited in past lives scriptional-points to the institutionalization of the Nāth sampradāya, via

are destroyed. His disciple ... was Chayādhinātha ["Shadow Ādinā- the recognition of Gorakhnath and other figures as its historical founders,

tha." His disciple was Stambhanatha] ... His son, versed in the in the thirteenth century. What appears to have occurred was the "Gorakh-

meaning of the Kalagama, was the yati Dvipanatha ... His disciple ization" of disparate monasteries (Kadri), lineages (the Vairäg panth), and

was born Mauninätha munipa. The bearer of the latter's commands even entire religious orders (the Pāsupatas and Kāpālikas), which became

was Rūpasiva [the priest given charge of the temple] ... devoted to absorbed into this new institutional entity. In the case of the latter two old

the śaivāgama. Saiva orders, an A.D. 1287 inscription from Somnath (Junagadh district, Gujarat), the original homeland of the Pāsupata sect, is significant, inas-

As is common practice, the founder of the Siddheśvara temple traces much as we find in it the name of Gorakh appearing together, for the first

the lineage of its priest back to the divine founder of his sectarian tradition. time, with that of the "Lākulīśas" (i.e., of Pāsupatas).84 In many ways, the

As such the greatest historical interest of this guruparampara lies in the fact Nāth Siddhas appear to be the direct heirs to the Pāsupatas and Kāpālikas.

that it does not mention the traditional "historical" founders of the Nath Saiva sectarianism in fact begins, some eleven hundred years before

Siddhas (who figure in the later A.D. 1279 Kadri inscription, from the same the advent of Gorakhnäth, with this former order, whose teachings were revealed by Srīkantha or Nilakantha Siva himself to the sect's founder region, mentioned above), even as it commemorates the founding of a "Siddha" temple. In this, it supports the proof elicited from the Mahar- Lakulīsa-the "Lord with the Stave"-who hailed from Kayavarohana (the modern Karvan, Baruch district, Gujarat) in the first half of the second ashtran data a moment ago. There were, prior to the thirteenth century, a number of groups, mainly Saiva, who called themselves Siddhas, as well century. The Pasupatas were the prototypes of the Saiva ascetic, whose daily "bathing" in ashes, antisocial behavior, sectarian markings, goals of as groups whose founding gurus had the suffix -natha appended to their siddbis and bodily immortality, and even an emphasis on "yoga"85 have names. There were, however, no groups calling themselves Siddhas with names ending in -nätha who traced their lineage back to a founder named been passed down to present day Saivites via such sects as the Kāpālikas.

Gorakhnath prior to this time. It is not until the thirteenth century that This latter group, which emerged in the beginning of the first millennium

such specifically "Gorakhnäthi" lineages suddenly appear, in at least half a of the common era, constituted the more heterodox, left wing of Saiva

dozen places in western India. sectarianism. Situated at the orthodox end of the Saiva spectrum in this

It is, moreover, to the same period that the major Sanskrit-language period were the Kalamukhas, mainly based in Karnataka, and the Saiva

hathayogic works of Gorakhnāth, as well as a "canonical" body of vernacu- Siddhäntins, who were based in both south (Tamil Nadu in particular) and north (Madhya Pradesh, Kashmir) India. lar mystic poetry-attributed to Gorakhnath and others and written in a mixture of Old Rajasthani (Dingal), Old Punjabi, Khadī Bolī, and Apa- While the Pasupatas and Kapālikas were clearly the forerunners of the

bhramsa-are dated.80 We also find external references to Gorakhnāth in Näth Siddhas, and whereas the disappearance of the Kālamukhas more or

(1) the hatha yoga section of the Sarngadhara Paddhati, a work dating from less coincided with the appearance of the Virasaivas in the Deccan, it is the

the thirteenth to fourteenth century;81 (2) the A.D. 1372 foundation of the Saiva Siddhāntins who have defined mainstream Saiva ritual devotionalism for the past twelve hundred years or more. Saiva Siddhanta was, by the ninth important Nath monastery at Dhinodara (Kacch district, Gujarat) by Dharamnath;82 and (3) two Kathmandu Valley inscriptions, dating from century, the Saiva orthodoxy of both north and south India.86 Following this ninth-century watershed, however, a north-south rift developed, with much A.D. 1382 and 1391, left by "a follower of the sect of Gorakha," and by a of Kashmir Saiva metaphysics being propelled in the direction of a nondual- Yogi named Acintanäth.83 Prior to this surge of solid historical data, rein- forced by the literary evidence of the thirteenth-century "canon" attrib- ist idealist gnoseology by such great theoreticians as Abhinavagupta and the

uted to Gorakhnäth, we have little but legend to go on. south remaining dualist and more concerned with external practice than in- ner realization. Because, however, the Näth Siddhas have historically sub-

98 99 Chapter Four Sources for the History of Tantric Alchemy

scribed to none of the central tenets or canons of the Saiva Siddhanta ortho- a significant one here; one who belongs to a panth is a panthi, a "path-taker," doxy-devotion to Siva, external worship, conventional behavior-this and an itinerant lifestyle has long been the hallmark of the Näth order. important tradition need not concern us further.87 For our purposes, it is most important to note that the twelfth- to A number of Näth Siddha traditions maintain that there originally ex- thirteenth-century Gorakhnäth, while he was a reformer, innovator, and sys- isted twelve Śaiva subgroups-the "Yogis" of the post-Gupta age; i.e. Pāśu- tematizer, did not innovate either the techniques of hatha yoga or the Näth patas, Kāpālikas, Kālamukhas, etc .- to which Gorakhnāth added another sampradaya out of whole cloth. There was already something there: an em- twelve of his own creation.88 These figures are indirectly supported by a bryonic hathayogic tradition, a body of legend in which there already figured list from the relatively recent Sabara Tantra (quoted in the ca. seventeenth- a Siddha or Vidyādhara named Goraksa or Gorakh, and several religious or- century Goraksa Siddhanta Samgraba) of the "twelve Kāpālika teachers" and ders and sects, some of whose adherents were already calling themselves their twelve "disciples." The former list, which is headed by Ādinatha, con- "Nāth" or "Siddha" before Gorakhnāth appeared on the scene.95 tains names that hark back to the older Saiva orders: Kāla, Karāla, Maha- While such may, at first blush, appear to be overly facile, we may quite kāla, Kālabhairava-nātha, Vātuka, Bhūtanātha, Vīranātha, Śrīkaņtha, etc. safely characterize the emergence of the Näth Siddhas as a marriage be- In the latter list, of the twelve "disciples" of these teachers, we recognize tween Nāths (i.e., Saiva groups-Pāsupatas, Kāpālikas, and Sāktas-for the names of the traditional founders of a number of the twelve panths of whom Siva had long been called Nath, "Lord") and Siddhas (Mahesvara institutionalized Nāthdom: Nāgnāth (or Nāgārjuna), Satyanāth (or Sat- and Rasa Siddhas and Sittars, as well as the Buddhist Siddhacaryas)96 which nāth), Goraksa, Carpata, Vairāgya (Bhartrhari), Kanthādhārī (Kanthadī), took the institutionalized form of the Nath sampradaya. Starting in this Jālandhara, etc.89 period, all manner of preexisting Saivite and Siddha clans, lineages, or sects Nath Siddha tradition further maintains that the twelve Saivite orders would have funneled themselves into the Nath suborders or have been and the twelve Gorakhnathi orders battled one another, with all but six of identified as Nath Siddhas by outside sources. In addition to groups al- each of the two groups being destroyed. These twelve surviving orders ready mentioned, we also find the Vaisnava Avadhūta sampradāya (founded were confederated by Gorakhnath to form the twelve original panths of by Dattātreya),97 Dasnāmī Nāgas, Jains, Sufi Muslims, and a group of snake the Nath sampradaya.90 All of these twelve original subsects were based in charmers claiming allegiance, in the fourteenth and later centuries, to one Saivism's ancient heartland, i.e., the northwestern part of the Indian sub- or another of the twelve subsects.98 By claiming descent from Gorakh or continent (Gujarat, Punjab, Sind, Rajasthan), with one group, the Rawals Matsyendra, or some other founding Nath Siddha, these various groups or Nagnathis, having its base in Afghanistan."1 This configuration or con- grafted themselves onto what they considered to be a past with a future.90 federation has changed over the centuries: twentieth-century enumera- It appears that, from the outset, there have always been more candidates tions of the twelve panths base two of these in Bengal, one in Orissa, one in for the twelve "slots" than there have been panths, which has made for a Nepal, and one (the important Gorakhpur monastery) in Uttar Pradesh.92 protean aggregate.100 Indeed, even the correct name for members of the The emergence of the Nath Siddhas or Nāth sampradāya was probably aggregate I have been blithely referring to as the Nth Siddhas is a matter of a less dramatic order than that described in this legend. Its coalescence, of disagreement. Other terms include: Yogi or Jogi, which is both too broad both prior and subsequent to the institutional organization founded by a term, covering as it does all manner of itinerant sādbus, including certain Gorakhnäth, was more likely a gradual process, catalyzed by the itinerant of the historical forerunners of the present order,101 and too pejorative, lifestyles of India's sadbus and the oral communications network they have since 7ogi has long been the name of a scheduled caste in India;102 Kānphata always maintained at the many pan-Indian pilgrimage sites they have fre- ("split-eared") Yogi, which applies to the sectarian marking of some, but quented over the past seven centuries. The end result of this process was not all, cenobitic members of this order, who wear great earrings (kundalas, the loose confederation of lineages that is today known as the barab panth, mudras, darsanas) through the thick of their ears, which are bored open a hydra-headed group with no single leader,93 which holds general assem- upon full initiation; Gorakhnäthi, which evokes the name of the founder or blies of sorts on the occasions of major religious festivals (for example, at reformer of the institutional order (however, many suborders trace their the kumbha melas at Hardwar and Allahabad).94 The term panth, "path," is lineage back to Matsyendra or Jalandhara, and not Gorakh, and those who

100 I0I Chapter Four Sources for the History of Tantric Alchemy

claim descent from Jalandhara generally eschew the kundala earrings, even Saiva and Sakta traditions, 108 as well as with the seventh- to twelfth-century upon full initiation, and call themselves "Aughars," rather than "Kānpha- tas" or "Darsanīs");103 and Nāth Panthi or Nath Sampradayi, which appear founders of tantric Buddhism, known as the Mahāsiddhas or Siddhācāryas.

to be overly institutional in their emphasis. If we are to judge from recent publications coming out of the important Gorakhpur monastery, Näth Siddha, the term I employ, is the term most 4. Sorting Out the Siddhas

favored by the leadership of the order itself.104 This is also a term that ac- knowledges the historical Saiva and Siddha roots of the sect, roots evoked While the Nath Siddhas have left us burial tumuli and a living (though

in the Yogabija of Gorakhnäth, which evokes both the siddha-mata (Siddha perhaps not thriving) institutional order-in which disciples are initiated

doctrine), and the nätha märga (Nath path).105 It also acknowledges the fact by gurus, often in a monastic setting, within the barah panth structure-all that the Rasa Siddhas have bequeathed us are lists of names and a corpus that most adherents to this sect suffix their names with -nath.106 Having said this, we must allow that the reality of the situation cannot be decanted of alchemical literature. Just as there are no gurus initiating disciples into

into the neat typologies I have been attempting to generate here. So it is alchemical orders or lineages today,109 we can be quite certain-in spite of the Rasa Siddha lists and scattered references one finds to the Rasesvara that we find, under the umbrella heading of Nath Siddha, Nath sampra- daya-or any of the other alternatives enumerated-notions of both sect doctrine (darsana, mata), or to alchemical initiations of disciples by

and subcaste (especially among the large communities of householder gurus-that there never were institutionalized orders of alchemists in me-

Naths found scattered throughout north India) and a wide array of lineage dieval India.

claims, as well as variations in dress, vows, devotional worship, and forms This is supported in the classic RA, which, like a number of other al-

of behavior that appear to be both regional and sectarian in origin.107 chemical sources, discusses alchemical gurus, disciples, and initiation, but

At this point, let us return to the AK list, which was the point of depar- which simply states, on the subject of lineages, that the guru should be "devoted to the kula-märga" and "the member of a true order (satsampra- ture for this long excursus on the sectarian origins of the Siddhas. Our hypothesis-that the Nath sampradaya emerged in the early thirteenth dāyin)."110 More significant is its contention (1.27) that "to those who say this [alchemical] order (sampradāya) is not a 'womb,' it is maintained that century as a great medieval changing house of Saiva and Siddha sectarian- ism-is supported by this source. Here, our thirteenth-century terminus mercury is a 'womb.'" !! 1 In the context of tantric doctrines according to which initiation into an order, i.e., a clan, consists in partaking of the "clan- ante quem is clearly demonstrated by the fact that no alchemical source nectar" that streams from the womb of the Goddess, this statement is a prior to the fourteenth-century AK lists the "Nine Naths," including Gor- defense of the fact that the medieval alchemists were "biologically" linked akh (whose name it sanskritizes to Goraksa), in its Rasa Siddha lists. In the final analysis, the myriad Siddha lists of eleventh- to fifteenth-century In- through initiation, even if theirs was not an institutionalized sect. Quick-

dia provide us with the pool of names, mystic disciplines, and sectarian silver entrepreneurs, they took their initiation directly from Siva. This squares with the rare autobiographical data the alchemical authors orientations that would come to funnel themselves into the Nath order, offer in their own writings, which indicate that these figures were attached confederated in the late twelfth or early thirteenth century by the western to royal courts, rather than to any monastic or religious order. The Indian reformer Gorakhnäth. This order, which inherited the techniques of tantric alchemy from the Rasa Siddhas, would transform that discipline mercury-eating regular and secular orders about which Marco Polo and François Bernier wrote belonged to "Yogi" (i.e., Siddha alchemical), and into what we have termed Siddha alchemy, the melding of external alchemy not Rasa (i.e., tantric alchemical) lineages. with internal hathayogic practice, as such was taught by the tenth-century At this point, we must clarify the position of yet another Siddha tradi- Matsyendra and systematized by the twelfth- to thirteenth-century Go- tion, an aggregate that overlaps many of the other Siddha groups I have rakh. It is in this sense that all manner of medieval Indian traditions have been discussing. This is a group called the Mahesvara Siddhas. As their been quite correct in their retrospective identifications of the Nath Siddhas name indicates, this group defines itself by its devotion to Siva, the Great with the tantric alchemists, the semidivine revealers or transmitters of Lord (mabesvara); and in fact nearly every Saiva sect, school, or institu-

I02 103 Chapter Four Sources for the History of Tantric Alchemy

tion-the Pāśupatas, Kāpālikas, Kālamukhas, Saiva Siddhāntins, and Vīra- a siddbi contest. When Allama Prabhu takes a sword to Gorakh, its blade aivas (or Lingāyatas)-has called itself or been called Māheśvara.112 Ety- shatters on his adamantine body; when Gorakh does the same to Allama mology and logic would then lead one to conclude that the term mabeśvara Prabhu, it passes through his body, which is wholly ethereal. Allama siddha ought to apply to any mystic devotee of Siva Maheśvara who has Prabhu then chides Gorakh, saying that such bodily density is merely the perfected (siddha) himself through his practice. The specific use of the term mark of a density of illusion.115 Another Virasaiva source maintains that it is more restricted, however, and relates directly to our subject. The was Allama Prabhu who initiated Gorakh and Nagarjuna into the alchemi- Māheśvara Siddhas were, generally, alchemists who did not seek bodily im- cal arts at Srisailam. Yet the same Allama Prabhu (and he was not alone in mortality (jivanmukti) as the final goal of their practice, but rather aspired this) scorns Gorakh and the Nath Siddhas for the "this-bodily" emphasis to another sort of liberation (parāmukti). Rather than a perfected (siddha- of their alchemy.116 We may therefore conclude that while all or some of deba) or adamantine (vajradeha) physical body, the Māheśvara Siddhas' goal the Mahesvara Siddhas may have been alchemists, and while certain of was a divine body (divyadeba) of a more ethereal, even incorporeal nature. the Virasaivas may have even been itinerant alchemists, they were not al- At times, one has the impression that the Māhesvara alchemists (the au- chemists of the same stripe as the Nath Siddhas, the Siddha alchemists of thors of the Rasabrdaya Tantra [RHT] and Rasopanisat [RU] are examples) medieval India. were seeking the same sort of liberation (moksa) that nearly all mainstream The Mähesvara Siddhas were concentrated in the Deccan and western Hindus aspire to-i.e., the absorption of the individual soul (atman) into coastal region of India.117 As noted, a Siddheśvara temple was dedicated, the absolute (brabman)-and that their alchemy was merely a means in A.D. 1030, in coastal Karnataka. Several centuries earlier, a number of to prolonging the lives of bodies within which their hearts and intel- temples to Siva Siddheśvara were constructed in the Srisailam region of lects could discover their innate unity with the universal essence. Such is Andhra Pradesh, in the Deccan.118 Elsewhere, architectural or archaeologi- also the purport of the "Raseśvara Darsana" chapter of Mādhavācārya's cal data provide us with very few clues to the history of tantric alchemy. fourteenth-century Sarvadarsana Samgraba, which opens with the state- The Visnudharmottara Purāna, a ca. eighth-century Kashmiri text that is ment: "Other Mahesvaras there are who, while they hold to the identity of most attentive to the iconography and plastic reproduction of divine im- self with Paramesvara, insist upon the principle that the liberation in this ages, contains descriptions of mercurial preparations used to coat stone life taught in all the systems depends upon the stability of the body, and sculptures and thereby increase their resistance to the elements.119 The al- therefore celebrate the virtues of rasa ... as a means for stability of the leged presence of mercury and other elements in a number of metal images body." As minister to King Bukku I (A.D. 1356-77) of Vijayanagara (mod- from the Buddhist monastic university complex at Nalanda (Bihar) would ern Hampi, Bellary district, Karnataka), Mādhavācārya identified what he indicate that alchemy was part of its curriculum prior to its destruction in termed the "revealed system of the Lord of Mercury" with the doctrines of the twelfth century.120 A ca. twelfth-century sculpted image, dominating the Mahesvara Siddhas, who were active there in his time.113 This emphasis the southern facade of the sikhara of the Menāl (Mahānādakāla) temple differs radically from that of both the Nath Siddhas and the RA, whose in the Bhilwara district of southeastern Rajasthan is said to be a representa- powerful rhetoric regarding the this-worldly nature of bodily liberation tion of Siva Raseśvara.121 In Malaysia, the foundations of a twelfth- to will be discussed in chapter six.114 thirteenth-century Siva temple containing a mercurial linga and other al- In practical terms, the alchemy of the Mahesvara Siddhas was more chemically prepared elements are evidence for the spread of medieval therapeutic than transmutational, for which reason it funneled directly Hindu alchemy to greater India.122 into south Indian Siddha medicine. In certain sources, the distinction be- This paucity of hard evidence and a plethora of mythic references have tween this group and the Rasa or Nath Siddhas is cast in more subtle terms, spawned a number of wild hypotheses concerning the origins and chronol- with the distinction between visuddha-māyā and mabā-māyā categories be- ogy of Indian alchemy. A persistent theory has it that the Muslim conquests ing emphasized. A legend from a Virasaiva source illustrates the difference drove the Buddhist alchemists in residence of such intellectual centers between the two sectarian and philosophical perspectives. Gorakhnāth and as Nalanda (Nalanda district, Bihar) south into the Deccan or up into the Allama Prabhu (also known as Prabhudeva, no. 52 in the RRS [c] list) have northern perimeter of the subcontinent, where they converted to Hindu-

104 105

Chapter Four Sources for the History of Tantric Alchemy

ism or took on Hindu disciples. According to this theory, the great works maśīla, Udandapura, etc., were the great seats of learning ... and alchemy of Hindu tantric alchemy would have issued from this diaspora, in such was included in the curricula of studies." Therefore, Ray concludes, most far-flung places as Bhutan, Srisailam, Devagiri (the capital of the Yadavas, of the early alchemical tantras were Buddhist.128 Later, in a very ambigu- in present-day Maharashtra), and Junagadh (Gujarat). A parallel theory, ously worded passage, Ray notes that the (then) recently discovered Kubji- dear to historians of medieval Hindu sectarianism, maintains that the Nath kāmata "presupposes the existence of other schools and we have the Siddhas were originally a group of Buddhists who apostasized in the face distinct mention of the Mahayana"; then, vaguely citing Shastri's Cata- of the Muslim invasions in order to continue their tantric practices. Both logue, he quotes a passage from this work in which "Siva himself speaks theories, attractive as they may appear as explanations for the medieval of pārada (mercury) as his generative principle and eulogises its efficacy emergence of the Rasa Siddhas and Näth Siddhas respectively, are seri- when it has been killed six times."129 The Sanskrit, which he gives in a ously flawed. footnote, is nowhere to be found in Shastri's description of the Kubjikā- Through the compounding of a scholarly error, first promulgated by mata, nor have I been able to locate it in the manuscript which Ray claims Prafulla Candra Ray123 in 1903, and later amplified by Sylvain Lévi, Jean was Shastri's source (NNA MSS no. I-285 ka). Yet, Ray's confabulations Filliozat, and Mircea Eliade,124 it has been standard practice, in the histori- have been duly reiterated by scholars, including me, throughout the twen- ography of Indian alchemy, to identify a work, attributed to Nāgārjuna and tieth century.130 entitled Rasaratnākara (RRĀ), as India's primal and primordial alchemical In addition to the western scholars cited above-who accepted the Bud- classic. However, in an article written in 1983, Dominik Wujastyk125 dhist cachet of Nāgārjuna's "Rasaratnākara" as authentic, and who thereby proved that the so-called RRA of Nāgārjuna was a conflation of three sepa- subscribed to the theory that the foundations of Hindu tantric alchemy rate texts: the RRA of Nityanatha, and the RM and Kaksaputa Tantra (KPT), were Buddhist-P. V. Sharma, India's most respected historian of y- both authored by Nāgärjuna. The youngest of these three works, the RM, a urveda, is clearly following Ray when he states that derivative source which borrows extensively from other Hindu alchemical when, due to the 12th century advance of [Muhammad] Bakhtyar tantras, is a thirteenth- to fourteenth-century work. It is unique, however, Khilji, the universities of Nalanda and Vikramaśīla were deserted, in its dramatic presentation of a portion of its data (most of its fourth chap- ter, which it shamelessly cribs from the RA)126 in the form of a dialogue, their scholars and scientists fled, some to Nepal, Bhutan, and Tibet,

atop Srisailam, between the Mādhyamika philosopher Nāgārjuna, the Bud- and others to south India-most especially to Devagiri, where they

dhist goddess Prajnāparamitā, King Sālivāhana,127 a yaksiņī haunting a sa- took refuge in the courts of the Yadava kings ... In south India, rasa sastra continued to develop under the custodianship of the Siddhas cred fig tree, and a host of mainly Buddhist luminaries, including Ratna- [Sittars]. The Siddhas are said to have been eighteen in number ... ghośa, Sūrasena, and Nāgabodhi. A smattering of Buddhist terminology They have been dated to the 1oth century and after.131 completes the tableau, leaving the superficial impression that the Nāgār- juna who authored the RM was the same figure as the Nagarjuna who had In spite of the fact that the Muslim scholar Alberuni himself maintained received the Perfection of Wisdom teachings a millennium earlier. that the invasions of his own patron, Mahmud of Ghazni, was the cause In spite of the fact that this text is in every other respect a straightfor- for the flight from western India of "the Hindu sciences,"132 a number of ward Hindu alchemical Tantra, which terms mercury harabija (the seed of data militate against this hypothesis. The first of these is the absence of any Hara, i.e., Śiva), and invokes Saiva gods, goddesses, and tantric practices, extant medieval Buddhist alchemical works outside of the Kalacakra and Ray finds in its fourth chapter proof positive that "the Rasaratnakara [sic] Tibetan Nying-ma traditions.133 Once one allows that the RM and RHT seems to us to be a typical production representing the Mahāyanist period were not Buddhist works, and when one takes into account the fact that of intellectual activity and we may not be wide of the mark if we put down the Tibetan Tanjur-which seemingly incorporated every extant Buddhist 7th or 8th century A.D. as its latest date." From here, Ray passes from the text into its voluminous mass-contains only four alchemical works, of specific to the universal: "From the 5th to the 11th century A.D. the col- which none bears a clearly Buddhistic stamp, one begins to wonder who leges in connection with the monasteries of Pātaliputra, Nālandā, Vikra- the armies of university-trained Buddhist alchemists were who flocked

106 107 Chapter Four Sources for the History of Tantric Alchemy

to the Himalayan kingdoms, the Deccan, and Tamil Nadu. Elsewhere, in the poetry and technical works of the Nath Siddhas to indicate that the Sharma musters ample evidence to prove that the Yadavas of Devagiri were "origins" of the Näth Siddhas were something other than Saiva.141 indeed great patrons of the alchemical arts.134 Yet, as he himself demon- Armed with this data, writers since at least the time of Taranatha, the strates, on the basis of solid historical evidence, the Yadavas' twelfth- and Tibetan author of the A.D. 1608 History of Buddhism in India,1+2 have ad- thirteenth-century alchemist protégés were a highly mobile group. Bhās- vanced all manner of argument to demonstrate the following thesis: that kara circulated between Kashmir, south India, and Devagiri; his son Sod- due to the similarity between names of Siddhacaryas and many of the nine hala between Devagiri and Gujarat (where Muslim penetration had begun founding Näth Siddhas beginning with Lui-pa (who was one of the found- in the closing years of the tenth century); and so on.135 ers of the Siddhācārya lineage) and Matsyendra[nāth]),143 and because Luî- One does not have the impression that the Hindu scholars of this period pā and other early Siddhācārya figures (including Kānha-pā and Jālandarī- were huddling in Hindu refuges from Islam; why would the Buddhists have pā) authored a number of the pre-twelfth-century Caryapadas, which were done so? Moreover, there is good evidence that Hindu physicians and al- written in Old Bengali, some or all of the Nath Siddhas, whose legendary chemists were welcomed into the courts of Muslim princes whose thirst founding guru was named Matsyendra, were originally Buddhists based in for immortality, increased virility, and the philosopher's stone would have eastern India, who abandoned the faith as a means to escaping Muslim been stronger than their religious fervor.136 Finally, we know that Muslim persecution in the first half of the thirteenth century. Taranatha does not physicians, alchemists, and mystics were avid for the wisdom of their In- mince his words: "At that time, most of the Yogi followers of Gaurksa dian counterparts, as evidenced by the translation, in the sixteenth century, [Goraksa] were fools; driven by the greed for money and the honour of- of a treatise on batha yoga, attributed to Goraksa, entitled the Amrtakunda fered by the tīrthaka [Hindu] kings, they became the followers of Isvara. (The Pool of Nectar).137 Some time prior to the seventeenth century, it They used to say 'we are not opposed even to the Turuskas [Turks, i.e., may have been a Bengali Muslim, Sheikh Fayzulla, who glorified the ex- the Muslims]'"1+ ploits of Gorakh, Matsyendra, and other Näth Siddhas, in his Goraksa (or To be sure, certain of the Siddhācārya guruparamparās, as preserved in Gorakba) Vijay.138 Elsewhere, the venerable and powerful Nath monastery both the Caryapadas and Tibetan sources, parallel data on the original at Goraksanāth Țilla (Jhelum district, Punjab, Pakistan) has been a center Nath Siddhas found in pre-seventeenth-century Bengali sources.145 So, for of Hindu-Muslim syncretism since a very early date.139 Finally, as we indi- example, in caryā song no. 11, Kānha-pā calls himself a "Kāpālika" disciple cated on the opening page of this book, at least one highly placed Mus- of Jālandharī-pa, and mentions the unstruck sound (anahata) and the wear- lim-the fanatically anti-Hindu Mogul emperor Aurangzeb-was still ing of earrings (kundala) that are hallmarks of Nath Siddha theory and petitioning the Nath Siddhas for their alchemical expertise well into the practice.146 It is also significant that the Nath Siddha lineage (pantha) seventeenth century. founded by Jalandhara is the sole lineage in which initiates' names are suf- Theories concerning relations between the Näth Siddhas and Buddhists fixed with -pa or -pada, rather than -nath; and that the -pa suffix is em- in the period of the Muslim conquest bear a certain resemblance to those ployed only by Näths or Siddhas in the eastern part of the subcontinent: I have reviewed on the subject of India's alchemists, with a number of Uttar Pradesh, Bihar, Bengal, Nepal, and Tibet.147 While there is no ex- scholars maintaining that certain of the founders of the Nāth sampradāya plicit legend on this subject in the Grub thob (which, it must be recalled, is were originally Buddhists hailing from eastern India.140 Such theories gen- a mid-fourteenth-century translation of an eleventh- to twelfth-century erally take the Siddha lists as their starting point. Seven of the figures Sanskrit source), a legend concerning the Siddhācārya Caurangi (Siddhā- whose names most frequently occur in enumerations of the nine "histori- cārya no. 10)-who is dismembered after being falsely accused by an evil cal" Nāths-Matsyendra (Luī-pā; Mīna), Gorakșa, Caurangi, Nāgārjuna, stepmother-does bring a cowherd (go-raksa) into play, who is instrumen- Kaņerī (Kānha-pā; Kāņipā; Karņarī), Jālandhara (Hādi-pā), and Carpați tal in restoring Caurangi's limbs. The sole figure to be identified as a yogin (Carpata)-figure in the Buddhist Siddhacarya lists. In addition, a number in this account is, however, Mīna-pā (no. 8), and not Goraksa (no. 9). Later, of scholars evoke the "Buddhist" terminology (sūnya, sahaja, etc.) employed in Taranätha's seventeenth-century work, it is Goraksa rather than Cau-

108 109

Chapter Four Sources for the History of Tantric Alchemy

rangi who is the dismembered and restored yogin.148 Finally, the existence torical author and reformer Gorakh. Indeed, he is portrayed as a cowherd in eastern India of large "Jogi" and "Kapālika" subcastes is marshaled as in his Grub thob legend.154 This figure may well have been "Buddhicized" evidence for a mass apostasy, from Buddhism to Hinduism (and thence to into a Vajrayāna deity in the northeastern Indian subcontinent: three juxta- Islam!), that would have been led by "Gorakhnath." 149 posed images in a monastery in Sikkim are identified as the divine Buddha However, just as Taranatha gets Goraksa's legend wrong, so too he and Amitabha, the human founder Gautama Buddha, and Goraksanātha!155 scholars who follow him are wrong when they attempt to make Näth Sid- Another striking piece of evidence comes from Gorakhpur, where dhas out of those seventh- to twelfth-century A.D. Buddhist Siddhācāryas who were based in eastern India. First, while there were indeed Vajrayāna Gorakhnäth ... discovered at the site of the present [Gorakhnāth]

figures named Goraksa, Luī-pā, Kānha-pā, Jālandhara-pā, and so on, their temple a shrine sacred to the god Gurakh or Gorakh, who appears

writings, as preserved in the Tanjur, are either clearly Buddhist or heavily to have been a deity of great fame in the Nepal country; and having

indebted to Hindu sources.150 Second, it makes no sense that Goraksa-if devoted himself to the service of this deity, practiced the greatest aus-

he lived, as the Grub thob says he did, in the time of the ninth-century king terities. He ... took the name of Gorakhnäth or servant of Gorakh.

Gopala151-would have had to have led his mass conversion of Buddhist Shortly after his death ... members of the [ruling Satāsi] family es-

Siddhas to Hinduism some four hundred years prior to the Muslim con- tablish[ed] themselves near the shrine, from which the town they

quest (1256-60) of Bengal! In fact, there are no extant Bengali works by founded took the name of Gorakhpur.156

anyone named Goraksa or Gorakh and no tradition whatsoever of there I have already noted that images of Goraksa were being worshipped ever having been a Bengali Siddha author named Goraksa. A very short from an early date at both Kadri and Somnath. Elsewhere, the sixteenth- work on the physiology of the subtle body, attributed to Ghoraksa (or century alchemical Rasakaumudī (3.50) prescribes the worship of the Nine Ghorakha, but not Goraksa, i.e., Siddhācārya no. 9), is included in volume Näths. The worship of a semidivine Gorakh or Goraksa, as well as of many 21 of the Tanjur:152 This work, included in the "Tantra" section of the other Näth Siddhas, remains widespread in India and Nepal down to the Tanjur, may date from as late as the fourteenth century, as might a number present day. of carya songs attributed to the same author, found in the same source. To conclude: since no extant tantric or Siddha alchemical works, either The Amrtakunda, attributed to a certain Goraksa from Kāmarūpa (Assam), Hindu or Buddhist, emerged out of Bengal prior to the thirteenth century, cannot be dated prior to the late sixteenth century, 153 and Bengali literary we need not concern ourselves any further with the imagined east Indian versions of the Goraksa Vijay postdate fifteenth-century Nepali and Mai- Buddhist origins of Gorakhnath or the Nath Siddhas.157 Elsewhere, the thili versions of the same by at least two centuries. If these data constitute Nāth Siddhas were, more than any other medieval Hindu sect, most ame- proofs for the existence of a pre-thirteenth-century "historical" Goraksa nable to syncretism with Islam. Many are the Nath Siddhas who are known in eastern India, it nonetheless remains the case that the contents of these as "Guru" or "Nath" by their Hindu devotees and "Pir" by Muslims. The two works in no way resemble the hathayogic summa of Gorakhnāth (even Bäuls of Bengal are a prime example of such a phenomenon: they are often if this figure's name is sanskritized to "Goraksanatha" in these works). The Muslims who revere Gorakh and other Nath Siddhas, and whose sung po- existence of a ninth- to eleventh-century eastern Indian Goraksa is un- etry curiously resembles that of the Buddhist Mahäsiddhas.158 The promi- proven; a more likely scenario would have the works of the twelfth- to nent place occupied by Jogīs in Indian Sūfi hagiography is a further indica- thirteenth-century western Indian Gorakhnath being transmitted back to tion of this syncretism, a subject that is only now beginning to receive the Bengal and attributed to a Bengali Goraksa in the thirteenth through four- scholarly attention it deserves.159 teenth centuries. The volume of legend surrounding Gorakh, Matsyendra, Kānha, Gopī- Prior to this thirteenth-century watershed, the sole Goraksa in Bengal cand, Bhartrhari, Carpati, Caurangi, and the other major Nath Siddhas is so or eastern India would have been a Siddha demigod worshipped by cow- massive as to constitute an entire field of study in itself. The names of these herds for the protection of their livestock, with no connection to the his- traditional founders, common enough to begin with, have over the centuries

I1O Chapter Four Sources for the History of Tantric Alchemy

been constantly reappropriated in such a way as to generate a plethora of tury, at which time they were supplanted or absorbed there by the Virasaivas figures with the same names and often the same gurus and disciples, but liv- who, under the leadership of Basava, were emerging as an important south ing in different centuries, under different kings, and in different regions of India sect, in much the same fashion as were the Nath Siddhas further to the India. Our discussions of the historical and legendary data pertaining to the north.163 The jyotirlinga named Mallikarjuna, the heart of the Saiva cultus at major Nath Siddhas will therefore be limited to those cases which serve to Srisailam, was reconsecrated there by the Vīrasaivas.164 trace the history and doctrines of tantric and Siddha alchemy.160 Following Banabhatta's seventh-century description of the hydrar- gyriasic south Indian Saiva ascetic who, in his mercury-provoked delirium, recounted "thousands of wonderful stories about Srīparvata," we also en- 5. Western India: The Heartland of Tantric and Siddha Alchemy counter references to Kāpālikas from Srisailam in Bhavabhūti's eighth- century Mālatī Mādhava and Ksemīśvara's tenth-century Candakaušika. As soon as we turn away from Buddhism and eastern India and towards The former of these is also important inasmuch as it contains the earliest Saivism and the west of the subcontinent, the alchemical trail suddenly extant Indian literary reference to the yogic physiology of the six cakras becomes hotter. I have already noted that all of the original twelve panths and the ten nādīs; in the latter drama, the Kāpālika hero is cast as a divi- of the Näth order were based in western, and especially northwestern, In- nized alchemist.165 Following the twelfth-century advent of Vīrasaivism, dia.161 As I will show, nearly all the historical data we have at our disposal alchemical references to Srisailam multiply. The Viramāheśvarāgama main- indicate that the medieval alchemists, too, were centered in western India, tains that Gorakhnāth was schooled in alchemy, by a Māheśvara Siddha, although further to the south than the Nath Siddhas' original haunts. on the shore of the Tungabhadra River. He, in turn, taught what he had However, as already noted, the Näth Siddhas were also an important pres- learned to Raseśvara Siddhas in the "Antarvedi" region of Maharashtra.166 ence at Kadri in Karnataka; and we know too that their well-traveled net- According to the sixteenth-century Telugu Prabhulingalila of Pidapatti So- work of sacred pilgrimage sites also drew them north into the Himalayas manātha Kavi, Gorakhnāth and Nāgārjuna were both initiated into the al- and as far south as Srisailam. In broad terms, the geographical area of con- chemical art by the Virasaiva teacher Allama Prabhu at Srisailam itself.167 vergence between the medieval Näth Siddhas and Rasa Siddhas covers a The Tamil Sittars echo the same tradition, tamilizing Gorakh's name to region roughly corresponding to the modern Indian states of Rajasthan, "Korakkar"; and adding that Nāgārjuna established a transmuting (spar- Gujarat, Maharashtra, and Karnataka. savedhi) linga of Siva there.168 According to a work by the Sittar Konganar, Korakkar had an animated mercurial pill called "bogi" [bhogi?] which, a. Srisailam when he held it in his mouth, afforded him the power of flight.169 He was the author of a work entitled Korakkar Malai Vagatam (Korakkar's Moun- Our survey begins, however, with Srisailam, perhaps the same site as the tain Medicines), the lore of which he collected during his life in the Dec- Śrīparvata of Buddhist fame around which an early body of alchemical lore can region.170 concerning Nagarjuna coalesced. As already mentioned, from the seventh Perhaps influenced by earlier Buddhist traditions, Nityanatha Siddha- century onwards, a number of temples were dedicated, on or around this who was very likely a Nath Siddha-states in his RRĀ that Nāgārjuna set peak, to Siva Siddheśvara. Already mentioned in the sixth-century Vāsava- up an alchemical laboratory on Srisailam.171 The A.D. 1400 Navanāthacari- dattā (together with the goddess Tara and the element mercury) as a site at tra of Gaurana indicates that Gorakh attained yogic bliss (yogānanda) in a which liberation could be realized, Srisailam was portrayed, in descriptions cave near a subterranean stream of the Ganges River, somewhere below from two thirteenth- to fourteenth-century Hindu alchemical sources, as an the sacred peak.172 The same source has Nagarjunanath teaching his son, alchemical wonderland.162 It was also a center for the Pāsupatas, Kāpālikas, Siddha Nagarjuna, the "gold-making" siddhi at Srisailam. When the young and Kälamukhas, three sects that rode the wave of Saivism that swept Bud- alchemist sets about to transmute the entire mountain into gold (in an ob- dhism out of western and southern India, from the seventh century onwards. vious retelling of accounts of the alchemist Nagarjuna of Buddhist legend) Indeed, it was these sects that controlled Srisailam down to the twelfth cen- his experiments are halted by Visnu.173 Here, the alchemical trail from Sri-

I12 113 Chapter Four Sources for the History of Tantric Alchemy sailam ends. What is most troubling is the fact that there exists no hard great patron of vaidyas and alchemists.182 Thirty years after his demise, it evidence-geological, chemical, archaeological, or epigraphical-to indi- was during the reign of the Yadava king Ramacandra (1271-1311) that the cate that mercurial alchemy was ever practiced at Srisailam!174 Inasmuch, Nāth Siddha Jnāneśvara wrote his Jñānesvarī and Yogisampradāyāviskrti however, as it was a hub of Saiva activity, prior to and following the advent (YSA), and that a grant of land was made by that king to the same Nath of the Virasaivas (whose links with the Mahesvara Siddhas were strong), order in the Jagalur tāluka of Karnataka in 1279.183 According to Nāth Sid- Srisailam was in fact linked to those regions, to the west and north, in dha tradition, Jñaneśvara's guru Nivrttināth fully realized his yogic practice which both alchemical raw materials and expertise were abundant.175 at Tryambaka Ksetra ("Plain of the Three-Eyed [Siva]"), the source of the Godavari.184 Gahanināth, the disciple of Gorakhnäth and guru of Nivrt- tināth in Jñāneśvara's lineage, is mentioned together with "Goraksanāth" b. Maharashtra. by Nityanätha in his thirteenth-century alchemical Rasaratnākara, an indi- The highly generic Śrīsailam, "excellent peak," has been identified, cation that this was a Maharashtran work. The Tantra Mabarnava, possibly throughout history and across several regional traditions, with a number a Western Transmission text, maintains that Nāgarjuna hailed from "a for- of mountains of the Indian subcontinent. In addition to the two or even est near the Godavari"; still another states that Gorakh was born on (yet three candidates for this toponym within Andhra Pradesh itself, there have another) Candragiri, on the banks of the Godavari.185 The Avadhūta sub- also been peaks called Srisailam in western Uttar Pradesh (at Devalgadh, a sect of the Nath Siddhas, said to have been founded by a twelfth- to hill eighteen kilometers northeast of Srinagar, in Pauri Garhwal district), thirteenth-century figure named Dattatreya, was also originally based in Kerala,176 and Maharashtra. The Garhwal toponym is closely connected to this region; the Dharamnāthi subsect is presently based there.186 Satyanath who, according to the Sankaravijaya of Anandagiri, conversed Another pan-Indian toponym found this region is Kadalī-vana ("Plan- with Sankaracarya atop this peak in the Himalayan foothills. The Nath tain Forest"), which rivals Moon Mountain and Moon Island as the most Siddhas identify this Satyanath with the Sittar author of twenty-one al- frequently recurring venue of Näth Siddha legend. A Plantain Forest, lo- chemical works.177 Elsewhere, the KJnN clearly identifies Srisailam with cated in the vicinity of Tryambaka Ksetra187 corresponds to a toponym the pītha of the Goddess at Kāmākhyā; the KM locates a śrīśaila-vana above found in the RA: this is Kadali-nagara ("Plantain City"), which this text the brabmarandhra and the four pīthas; and Siddha Nāgarjuna, in his KPT, too locates on the Godavari. In fact, in all of the rare cases in which the refers to Srīparvata as a kula-parvata or "clan peak." These three identifi- RA gives specific geographical data-most of it on the subject of "magical cations ought, however, like the "Nine Naths" of the Western Transmis- waters"188 (hot springs, corrosive mineral waters, poisonous waters, etc.)- sion, to be consigned to the realm of the subtle physiology of the bodily the locations it details are clustered around the headwaters of this river; microcosm, rather than to the geography of the Indian subcontinent.178 when the discussion is more general, it is locations in southwestern India Nath Siddha traditions locate a Maharashtran Srisailam in the vicinity that predominate. So, for example, this work's Plantain City description of the upper Godavari River,179 a region that was, for at least three centu- states that "in the south ... is a pure and auspicious river, as renowned as ries, an important center of activity for Nath and Rasa Siddhas alike. It the Ganges, called the Godavari. On its southern shore is a city called was near the headwaters of the Godavari that Devagiri (modern Daultabad, Plantain City; to its south is the world-famous mountain called Krsnagiri Aurangabad district, Maharashtra), the capital city of the Yādava kings, was ('Black Hill'); nearby is a town called Ambika, where sañjīvinī jalam ('resus- situated.180 In the centuries prior to the fall of their kingdom to the sultans citating water') is found." 189 of Delhi in 1318, the Yadavas appear to have directly patronized all of the Yet another RA (12.260-62) reference to this region designates a loca- groups I have been discussing in this chapter. Bhillama (1175-91), the tion in the Sahyadri (Vindhya) range to the north of the Godavari as the founder of the dynasty, was the royal patron of Bhāskara, the guru of site of a hot spring (usnodaka); a Nāth Siddha source, the Kadalīmañjunātha the Nagarjuna who authored the Yogaratnamala.181 Bhillama's successor, Mahatmya, locates the Plantain Forest on the southern flank of the same Jaitugi, was instrumental in introducing Saivism into the Kathmandu Val- range.190 ley; Jaitugi's successor Singhana was himself an Ayurvedic physician and a This data, together with what appears to be a reference, in the RRS, to

114 II5 Chapter Four Sources for the History of Tantric Alchemy the author of the RA having indirectly passed down the formula for a min- tion. He manages to secure both, and prepares the elixir on the shore eral preparation to Singhana-the Yadava king of Devagiri from A.D. 1210 of the Sedi River, with the aid of his royal patron's queen Candra- to 1247-make it likely that this, India's most important work on tantric lekhā. Her two sons kill him before he can benefit from the elixir alchemy, was compiled along or near the upper Godavari, albeit prior to (but not before he buries some of the elixir on Mt. Satruñjaya); the the A.D. 1175 founding of the Yādava dynasty.191 site on the Sedi where he made it becomes the Jain pilgrimage site called Stambhana Tirtha, the "holy site of immobilization," for this is where Nāgārjuna solidified his [mercurial] elixir.1% c. Gujarat One of the physicians who graced the Yadava court was Bhaskara, who A number of elements found in this legend link it to other sites on the came to Devagiri from Kashmir in the latter half of the twelfth century. Kathiawar peninsula of Gujarat, as well as to a site in Assam. Mount Sa- Two of Bhaskara's disciples, however, were based in Gujarat. The first of truñjaya/Dhānka was a peak renowned for wells and springs said to contain these was his son Sodhala, who was the author of the an Ayurvedic lexicon the elixir of immortality. A mid-fifteenth-century work, the Satruñjaya- entitled Sodhalanighantu, and the second was Nagarjuna, who eulogizes kalpa, relates a story in which a young man learns from a Jain monk of a this Bhaskara as his teacher in the opening verse of his Yogaratnamala. This well of rasa at the site, which is only visible to persons who perform reli- same Nagarjuna may have been the author of the Kaksaputa Tantra: both gious austerities in front of an image of the Jina Santinatha and make a are works on tantric healing and sorcery, and a number of verses, formulae, nocturnal offering to a snake goddess named Vairothya. Satisfied by the and colophons in both works are identical.192 We can be more or less cer- young man's acts, Vairothya opens the well for him, and he gains the elixir. tain of the date of the Yogaratnamala because its important "Laghu Vivrtti" This account closely resembles a myth from the ca. fourteenth-century commentary, written by Śvetāmbara Bhiksu Guņākara, is dated to A.D. Simbāsanadvatrimšika (Thirty-two Tales of the Throne), in which King Vi- 1239. While Gunakara was, as his title indicates, a Jain, he was also very kramāditya, by offering his own head to the tantric goddess Kāmāksī, af- likely a tantrika (a Jain tantric tradition did indeed exist alongside the fords access to a Siddha who has been attempting, without success, to enter Hindu variety) with a knowledge of alchemy. He was also from Gujarat, as into a sealed cave, next to Kāmākhyā pītba, in which a vessel containing evinced in the use he makes of a number of vernacular terms.193 the elixir is hidden.197 Now, Gujarat has long been a stronghold of Jainism in India, and it is The second connection concerns the toponym Dhanka. According to here as well that Jain alchemical lore is the most frequently encountered. another Jain source, the A.D. 1277 Prabhavakacarita, Nāgārjuna was born So, for example, two Svetāmbara Jain sources, the A.D. 1304 Prabandha Cin- not on a peak named Dhānka-Satruñjaya-Palittanakapura, but rather in the tāmani of Merutunga and the A.D. 1349 Prabandhakośa, give accounts of a town of Dhänka. On the urging of Dominik Wujastyk, who had been there Nāgārjuna who is a Jain alchemist, accounts which appear to borrow freely two years previously, I visited this site in the winter of 1993. A series of from Buddhist and Hindu lore on his subject. According to the latter work caves behind the Siva Dungaresvara temple to the west of this village con- tain old bas-reliefs and a sculpted image which the local residents say is a Nagarjuna is born the son of the snake king Vasuki and the human representation of Nagarjuna. It was in these caves that Nagarjuna would princess Bhopalā, from Mt. Dhänka, [a peak identified with the sa- have practiced his alchemy. 198 cred Jain site of Mt. Satrunjaya, also in Gujarat], an alchemical won- A third connection lies in the many mountains that become conflated derland.19 He grows up to become the preceptor of a Sātavāhana in the Jain sources. In addition to the three mentioned a moment ago, a king, but leaves him to go to Palittanakapura, the city of the Jain fourth, Valabhi-saila-either another name for the same peak (identified wizard Padalipta.195 After a series of incidents, Padalipta accepts him with the modern Palitana, in Bhavnagar district) or another peak in the as his student, and instructs him in the preparation of an elixir of area-is also reputed to have been a center for Jain alchemy.199 Valabhipur immortality (rasasiddhi), with the aid of a magical image of the Jain is the name of a town (twenty-five miles north of Palitana) which, more- tīrthamkara Pārśvanātha, and a chaste woman to stir [grind] the po- over, becomes the focal point for what has to be the most oft-repeated

I16 117 Chapter Four Sources for the History of Tantric Alchemy legend in Gujarat. This is the legend of the sudden destruction of a city by ment ago in the context of Nagarjuna legends. Now, if a tenth-century fire, through the agency of an angry holy man (often a Nath Siddha) or in Gujarat-based Nagarjuna indeed wrote a treatise on alchemy, that work some connection with alchemy. In Kathiawar, the name of the destroyed has been lost: none of his extant Hindu alchemical works predate the thir- town is Valabhī; in Kacch, it is called Pattan. teenth century. It is, however, worth noting that the sole text attributed to The historical source of the Valabhi version of the story was the sack of Nāgārjuna in the Jain canon is a rasa tantra.208 Elsewhere, the Jain equiva- the city by Amru ibn Jamal of the Sind in A.D. 788.200 Apparently, physical lent of the Buddhist Caryapadas, the Pāduha-doha written by Muni Rāma- traces of this destruction remained visible for several centuries after the simha in the Apabhramsa vernacular in about the year A.D. 1000, contains event and gave rise to a body of legend. An early account is that related an alchemical poem.209 An untitled anonymous Jain work on Āyurveda, by Alberuni in the first half of the eleventh century, according to which a written in Gujarati script, is perhaps the earliest Indian medical work writ- merchant who had become fabulously wealthy through the possession of ten in a medieval vernacular. It dates from the twelfth century.210 These an alchemical touchstone had managed to buy up an entire town, which a vernacular works aside, a number of major Hindu alchemical works, writ- king named Vallabha was also eager to own. The merchant entered into an ten in the Sanskrit medium and straddling the line between tantric alchemy alliance with a "Lord of Almansura," who made a night attack on King and therapeutic alchemy, are Gujarati productions. The most important of Vallabha and destroyed his town.201 Alberuni concludes: "People say that these are the twelfth- to thirteenth-century Rasendracūdāmani of Soma- still in our time there are such traces left in that country as are found in deva, a resident of (the lost Gujarati city of) Bhairavapura; and the places which were destroyed by an unexpected night attack."202 At the town thirteenth-century Rasaprakāśa Sudhākara of Yasodhara Bhatta, who hailed of Dhank some one hundred miles to the west of Valabhipur, blackening from Junagadh.211 on certain of the outer facades of Nagarjuna's caves is identified by the Surat, on the Saurashtra coast of Gujarat, has long been a center for the local people as traces of the sack of Valabhī.203 manufacture of synthetic cinnabar, through the sublimation of imported Some one hundred miles to the northwest of Dhank, on the southern mercury with native sulfur. Gujarat moreover continues to be a leader in coast of Kacch, is a town named Mandavi. According to Näth Siddha leg- alchemy and Ayurvedic rasa śāstra, down to the present century. The Gon- end, a city named Pattan was once located two miles to the north of Man- dal Rasāyansāla (Gondal, Rajkot district) was, throughout the first half of davi. Unfortunately for Pattan, the Nath Siddha Dharamnath, who was on this century, a center for the edition of the major alchemical texts by the his way east from Peshawar in the fourteenth century, chose to stop near resident vaidya, Yādavjī Trikamjī Ācārya. It was at the Gujarat Ayurved this town and sent two of his fellow Yogis to beg alms there. With a single University in Jamnagar that the first critical edition and translation of the exception, the hard-hearted people of Pattan gave nothing; and for this, Caraka Sambitā was compiled; and at the same site that alchemical gold Dharamnath upset his alms bowl and cursed the people saying, "be fallen was most recently synthesized, in 1968. Many of India's greatest present- [pattan] all the Pattan cities." So it was that all the cities in Kacch bearing day specialists in rasa sastra have received their training there from Profes- this name were swallowed up, together with their inhabitants.204 (The sor Hari Shankar Sharma.212 source of all of these legends may in fact be a Linga Purana myth of a Some twenty minutes by scooter rickshaw to the south of Junagadh demon named Jalandhara, in which it is stated that with the fire from his is the magnificent peak of Girnar which, according to the A.D. 1333 Vi- third eye, Siva destroyed two cities of the Sind, which fell into the vidhatīrthakalpa of Jīnaprabhāsuri and the Skanda Purāna, is also a site ocean.)205 In his Kaksaputa Tantra,206 Nāgārjuna refers to what appears to abounding in mercurial pools (rasa-kūpika) and alchemical miracles.213 be the fall of a city named Pattan; this would further support the theory Girnar is also a peak abounding in Jain temples, as well as caves and crags that he was a Gujarati author.207 sacred to the Näth Siddhas, and, to all appearances, the two groups have In about the year 1030, Alberuni indicates that a Nagārjuna practiced been enjoying a symbiotic relationship there for centuries. Girnar is dotted his alchemy some one hundred years before him-i.e., in the early tenth with shrines to a number of important Nath Siddhas (Gorakh, Dattātreya, century-at a site north of Somnath that he calls "Fort Daihak"; this would Bhartrhari, Gopicand), and has been an important pilgrimage site for the appear to correspond to the toponym Dhank, which was mentioned a mo- order since at least the thirteenth century; the site is already mentioned by

118 119 Chapter Four Sources for the History of Tantric Alchemy

Hsuan-tsang, in connection with "supernatural rsis," by which Pāsupatas dharas identified with such heights, to Pasupata spells to gain power over of the sort described by his contemporary Banabhatta are likely in- the same Vidyädharas, to the yogic cultivation of the power of flight (khec- tended.214 The beautiful Jain temples located on its southern flank date ara), to the claim they have laid to the high places of the Indian subconti- from the thirteenth century, but Jains may well have been there from an nent, to the widespread practice of using cannabis as a means to "getting earlier date as well. Gujarat is also a region with a relatively high concen- high,"220 the Nath Siddhas have always been upward bound. tration of Nath Siddhas, who have monasteries and temples throughout The Jain temples at Abu and Girnar alike feature massive images of the state.215 Nemīnātha (in black stone) and Pārśvanātha, who are considered to have been the twenty-second and twenty-third tirthamkaras of that faith. It will be recalled that it was a miraculous image of the former that was employed d. Rajasthan by Nagarjuna to produce an alchemical elixir in the Jain legend recounted Another western Indian site at which one finds a similar sort of symbiosis in the Prabandhakosa. Parśvanāth's name can in fact be simply read as "Lord between Näth Siddhas and Jains is the equally ancient sacred site of Arbuda of the Touchstone" (parsva, paras), making the alchemical connection an Devī (Mount Abu, Sirohi district, Rajasthan), located some 250 miles to obvious one. There is another dimension, however to this figure and his the northeast of Girnar as the Siddha flies. Like Girnar, Abu has an al- successor, which moreover aids us in understanding the apparent Jain- chemical reputation: in the early stages of my fieldwork, I was assured by Nath Siddha symbiosis encountered at Abu and Girnar. This is the body Yogi Narharinath, abbot of the Caughera monastery (in the Dang Valley of Näth legend that transforms Parsva and Nemi into Nath Siddha sons of of the inner Terai region of Nepal) and the most illustrious Nath Siddha Matsyendra. In a variant on the classic myth of Matsyendra's brush with scholar of our time, that I would find living, practicing alchemists at that death in the form of sexual dissipation in the Kingdom of Women in As- site.216 In fact there is a curious symmetry between Girnar and Abu, begin- sam, Matsyendra becomes a king in Simhala (generally identified with Sri ning with the fact that both are located on east-west ridges comprising a Lanka), and fathers two sons, Pārśvanāth (or Pārasnāth) and Nemīnāth (or series of three peaks. At the base of both pilgrimage routes, one encounters Nīmnāth). In a classic ruse, Gorakh seeks to "disenchant" Matsyendra by Nath establishments called "[Gorakh] Tileti."217 From these, the pilgrim's killing the two boys, skinning them, and hanging their hides out to dry. west-to-east progress first passes through the Jain temple complexes, fol- When Matsyendra grieves for his sons, Gorakh decries his attachment to lowed by shrines devoted to the Hindu goddess Amba (on the first peak) the gross bodily husk and revives the two boys. Matsyendra returns to the and other mainstream Hindu deities; on the second peak is a Nath shrine yogic path and initiates his sons as Näth Siddhas. They become the found- (Gorakh at Girnar; Gopīcand at Abu);218 and both pilgrimage routes termi- ers of the Nimnathi and Parasnathi pantbs, the two Jain suborders of the nate or culminate, at their easternmost points, at sites identified with Dat- Näth sampradaya.221 In such a syncretistic milieu, we should not be sur- tātreya. This last peak is called Guru Sikhara, the "Guru's Pinnacle," at prised to find Hindu and Jain alchemists exchanging expertise with one Abu, while it is simply called Dattātreya at Girnar. So too, the central peak another or Hindu and Jain sectarians sharing sacred mountains. at Girnar is referred to not as Gorakb kī Dbūnī (Gorakhnāth's Fireplace), There also exist historical connections between the two sites, connec- which is technically more exact, but simply as Gorakhnāth. tions which account in part for the significant numbers of Marwari and The reason for these Girnar toponyms goes back, I believe, to the ar- Mewari Hindus who travel to Girnar and of Gujarati Hindus who come to chaic Buddhist and Hindu cults of divine Siddhas and Vidyadharas, who Abu (of course, Jains from both regions circulate between the two sites as were denizens not only of the atmospheric regions, but also, in the Hindu well). Southeastern Rajasthan and Gujarat are linked, both culturally and popular imagination, of mountaintops. The Girnar peaks of Gorakhnāth linguistically, in a number of ways that point to a common historical legacy. and Dattätreya are so called not because they feature shrines to these Näth This legacy dates back to the time of the Maitraka kings of Valabhi. Ac- Siddhas, but because they are these Siddhas.219 In fact, one might quite cording to legend,222 when Valabhī was sacked and Silāditya VI, its last aptly summarize the legacy of the Nath Siddhas by calling theirs a con- ruler, slain, his queen, Puspavati, was returning to the capital from a pil- quest of high places. From identifications with the Siddhas and Vidya- grimage to Ambā Bhavanī, in her native land of Candravatī, a short dis-

120 I21 Chapter Four Sources for the History of Tantric Alchemy

tance to the south of Abu.223 Great with child, she hid herself in a cave in chronology, however, it is far more likely that Bappa's ear-boring initiation the mountains of Malia to give birth to a son. She then committed suttee. was performed by a Pasupata. This is further supported by the name of his Her son, raised by a foster mother, was named Guha, "Cave-Born." The initiator: Rasi was a common name ending among the Pāsupatas.229 It is child, who grew up among the aboriginal Bhils of nearby Idar (Sabarakan- certain, however, that the custodianship of the temple passed through the tha district, Gujarat)-in the Rajasthan-Gujarat border region-was later hands of the Näth Siddhas before being given over to the Rāmānandīs in elected to be their king. So it was that he became the founder of the Guhi- the sixteenth century.230 lot (from Guha) clan, the family of the future Ranas of Mewar.224 Ninth Following his initiation, Bappa assumed the title of "Rawal" (from the in line from Guha was Bāppā ("father") Rawal, who, according to the Sanskrit, rāja-kula, "royal lineage") whence the name by which he is known fifteenth-century Ekalinga Mabatmya, founded the kingdom of Mewar in to history: Bāppā Rāwal. According to Hazari Prasad Dvivedi, Rāwal was, A.D. 728 and the Eklingji temple (fourteen miles north of his capital of already in the eighth century, a clan name proper to the Pasupatas which, Udaipur). In fact, there is no extant epigraphical mention of Bāppā Rawal in the thirteenth century, became the third of the old Saivite clans absorbed prior to the A.D. 971 Eklingjī inscription of Naravāhana.225 into the Näth sampradāya. The Rawals have, in the course of the inter- The Ekalinga Mahatmya and a number of other legends further associate vening centuries, become transformed into a Muslim suborder, based for Bappā Rawal with sectarian forerunners of the Nath Siddhas. According to the most part in Pakistan and Afghanistan.231 These were "great wander- these sources, the young Bäppa, forced to live in forest exile after the slay- ers" (they give the Persian rawinda, "wanderer," as the etymological root ing of his father Nagaditya by the Bhils,226 encounters an itinerant Saivite of their name), who were to be found peddling quack medicines and other ascetic named Hārīta Rāśi, who accepts him as his disciple. Hārīta Rāsi wares of a dubious nature in nineteenth-century Europe and who continue agrees to initiate Bappa into his Saiva order and thereby imbue him with to sell their services as hail stoppers in Kumaon, where they are called Oli- immortality and supernatural powers. But when Bāppa comes to the ap- yas, "hail men." Rawals are also based in Andhra Pradesh, Himachal pointed initiation site, the ascetic has already begun an ascent to the atmo- Pradesh (Jwālamukhī), and Hariyana (Asthal Bohar). spheric realms. Before rising out of sight, however, he spits down upon his As a suborder, the Rawals are closely associated with the Dharamnāthis, disciple, commanding Bāppā to receive his expectorate in his mouth. Satnäthis, Pāgalpanthis, and Parasnäthis.232 They are also known as Nāg- "Bappā showed his disgust and aversion by blinking, and the projected näthis, by virtue of which fact they are said to bear some connection to the blessing fell on his foot, by which squeamishness he obtained only invul- Nāth Siddha known as Arjun Naga or Nagarjuna, whose suborder is based nerability by weapons instead of immortality."227 With these, he grows up at Jwalamukhī in the Kangra district of Himachal Pradesh. If only by virtue to defeat his father's slayers and become the founder of the Mewar king- of his name, this figure is identified as an alchemist. Elsewhere, Gorakh dom. As I show in chapter ten, this common theme of Saivite and Nath refers to an unnamed Rawal Yogi as an alchemist in one of his bānīs.233 Siddha legend has important philosophical and practical implications. Nāgnāth, the legendary founder of the Rawal suborder, is said to have been Legend has it that another itinerant ascetic whom Bappa met in his a disciple of Gorakh; since, however, this clan was originally Pāśupata, it wanderings in the wilds of Udaipur, and who gave him a two-edged sword predates Gorakhnäth by several centuries. Another early Nāth Siddha hail- with which to defeat his enemies, was none other than Gorakhnath. Of ing from the present-day region of Himachal Pradesh is Carpati, who is course, this is chronologically impossible. However, inscriptional and nu- said to be the disciple of Matsyendranäth. A historical Carpati is named mismatic evidence supports the Eklingjī temple's claims to antiquity and in a tenth-century vamsavali from Camba, and his samādhi is located at Bappa's connection with the Saivas of his time. A gold coin from the time Chambādevī (Chamba district), to the north of Jwālamukhī.234 Jwālamukhī of his reign is inscribed with the words Sri Voppa on the obverse, together was already a pilgrimage site in the time of the Mabābbārata;235 at present, with a trident, a Siva linga, and a bull; below is the image of a man, pros- the Nath Siddhas control a subsidiary shrine-a miraculous pool of cold trate, having large pierced ears, the holes exaggerated.228 As such, this seal boiling water-at the site.236 would appear to be a representation of Bappa's initiation, featuring the ear Today, it is the Nagnathi-Rawals who accompany pilgrims to the far- boring that has so long been identified with the Nath Siddhas. Given the flung western pītha known as Hinglāj Devī (Las Bela district, Baluchisthan,

I22 Chapter Four

Pakistan).237 A gas vent and "fireplace of Gorakhnath" are maintained there down to the present day.238 In spite of the inaccessibility of her original FIVE

worship site, Hinglāj Devī remains an important goddess for Hindus in western India, and temples consecrated to her are found throughout this Tantric and Siddha Alchemical Literature region. Tradition links Pasupata forerunners of the Nath Siddhas with the founding of yet another dynasty in western Rajasthan. These were the Rā- wal kings of Jaisalmer.239 According to the bardic chroniclers who were the informants of Colonel James Tod, the founding of this dynasty occurred in the following fashion: Deorãj [b. A.D. 836], the future founder of the Rawal dynasty, is a In an important theoretical work written in 1981, Marcel Detienne argues prince without a kingdom. One day, he is visited by a Jogi named that "raw myth," if such a thing ever existed, would have taken the form of Rita who bestows upon him the title of Sid[dha]. Rita, who possesses lists or catalogues of names, lists from which generally oral native exegesis the art of transmuting metals, one day goes away, but leaves his tat- would have subsequently generated genealogies or lineages, and which tered cloak [jarjari-kantha] behind. Inside the folds of this cloak is written interpretation would later have transformed, from the vantage point Rita's elixir vessel [rasa kumbha], from which a drop falls upon De- created by the act of writing, into mythology. The myth, fragile and eva- orāj's dagger, turning it to gold. Deorāj decamps with the elixir ves- nescent, is metamorphosed into something else as soon as it is recounted. sel, and uses it to raise an army and the walls of a fortified city. Rita Therefore, for all intents and purposes, the myth does not exist: it is rather is well aware of the theft, and later comes to visit Deoraj in order to mythology, the exegesis and interpretation of myth that scholars study, legitimate the latter's possession of his stolen property. This he does whether they be located "inside" or "outside" the tradition in question.' on the condition, however, that Deoraj become his disciple and, as a With the lists of Rasa and Nath Siddhas that grounded our discussion in token of his submission and fidelity, adopt the external signs of his the preceding chapter, we seemed to find ourselves in the gray area be- order. He gives him the ochre robes of his order, places the earrings tween the ground zero of myth (catalogues of names) and the first impulse (mudra) in his ears, the little horn [singnad] around his neck, and loin- of native exegesis, i.e., the organization of list into lineage. We also wit- cloth [langoti] about his loins; in this garb, and with gourd in hand, nessed the written interpretation, and thus the transformation of these Deoraj then perambulates around the dwellings of his kin, exclaiming original "myths" into mythology, in the legends of the eighty-four Maha- alakh! alakb! Then, having exacted that these sectarian rites of initia- siddhas; the creative appropriation, in chapter four of the RM, of certain tion should be continued to the latest posterity, Rita disappears. of the names figuring in the Rasa Siddha lists; and the proliferation of my- Thereafter, the title of Rao was abandoned for that of Rawal.240 thology generated by the multiple lists of the Nine Naths. By reading be-

If this account is historically accurate, then the ninth-century western tween the lines of these mythologies, I attempted to glean a certain number

Rajasthani forerunners of the Nath Siddhas were already Nath Siddhas in of elements to aid in reconstructing a history of the medieval Siddha tradi-

everything but name. The jarjari-kantha, mudras, singnad, and use of the tions. In the present chapter, I attempt to effect a similar hermeneutics of

expression alakh ("attributeless," an apophatic description of the absolute) retrieval, this time through a survey of the alchemical literature.

are all hallmarks of the sect. If it contains elements from a later time than it purports to describe, it nevertheless portrays the Näth Siddhas as wonder- I. General Survey of Alchemical Literature in the Sanskrit Medium working king-making alchemists.24 Given the chronology, Deoraj's itin- erant Yogi Rita would, like Bāppā Rāwal's Hārīta, have been a Pāśupata Of what value are the Rasa Siddha lists for anyone who would attempt to rather than a Nath Siddha. reconstruct the history of Hindu alchemy? Nearly half of the names on the

123

124 125 Chapter Five Tantric and Siddha Alchemical Literature early lists (RM; RRS [a]; RRA) appear to be so many voyageurs sans baggages, gods whose divine works have yet to be revealed to humanity. Whatever with either no alchemical texts, attributions, or citations connected to their the case, I will follow the lead of the tantric and Siddha alchemical authors names in later sources, or, at best, unverifiable references in this or that themselves and generally disregard those Rasa Siddhas whose names they solitary source.2 Many appear to simply be the names of gods. As we have but rarely mention, save in the lists reviewed in the last chapter. already noted, Adima (1) is none other than a variant on Ādideva or Ādinā- A number of names from the early lists appear to correspond to authors tha, the Lord of Beginnings. Sambhu (11), Brahmā (24) and Hari (27) are of various works. Candrasena (2) is said to be the author of a lost work the three gods of the so-called Hindu trinity. Ratnaghosa (10) is a Buddhist entitled Rasacandrodaya.5 Lankeśa (3) is generally identified with Rāvana- figure, while Kambali (16) and Nagabodhi (20) are names carried over Lanka's Lord (lanka-isa)-who authored the Kumara Tantra, an early and from the Buddhist Mahasiddha lists. These would appear to be names of important text on the treatment of childhood diseases, and whose daugh- figures classified as divine or semidivine Siddhas.3 ters are said, in the tenth- to eleventh-century Tamil Irāmāvatāram of Another group of names are so many black holes in an alchemical void. Kampan, to be the wives of the eighteen Tamil Sittars.6 Māndavya (7) is the Who were Visārada ("Skilled"), Matta ("Intoxicated"), Ekaratna ("One name of an Ayurvedic teacher cited by Vagbhatta the Elder in the Astānga Jewel"), Indrada ("Gift of Indra"), Surananda ("Joy of the Gods"), and Samgraba and the author of a lost alchemical work entitled Rasavāridhi. He Khanda ("Portion")? Were they teachers of known alchemists and as such is also cast, in a passage from the fourth chapter of the RM,7 as a conversa- mute links in lost alchemical lineages? If they were authors of alchemical tion partner to Nagarjuna and [co-]author of a work entitled Vasistamānda- works, were their works lost to or simply disregarded by all later alchemical vya. (As we have already indicated, however, the historicity of many of the authors? Or have their works and the works of later authors who may have figures introduced as alchemists in the RM-including Ratnaghośa [10] cited them all been lost? Indeed, certain south Indian historians of alchemy and Nägabodhi [20]8-is doubtful). Somadeva, the twelfth- to thirteenth- maintain that all of the major alchemical works were originally committed century Gujarati author of the Rasendracūdāmani [RC],° mentions Bhāskara to writing in the medieval Grantha script of Tamil Nadu and that certain (8) and Sambhu (11), but gives no details on who they were or what they of these were never transliterated into the north Indian Devanagari script. wrote.10 Since Somadeva was himself from Gujarat, however, the Bhāskara As evidence, they note the fact that many south Indian alchemical works he invokes may well have been the guru of the twelfth- to thirteenth- cite alchemical sources of which north Indian traditions ignore the exis- century Nāgārjuna who authored the Yogaratnamālā. Sūrasena (9) is pur- tence.+ Another possibility is that these were individuals whose teachings portedly the author of the lost Rasendrasuraprabhāva, Naravāhana (13) that were oral, rather than written; or that these were people who were living of the lost Rasanandakautuka, and Lampata that of the equally lost Lam- exemplars of the alchemical arts. Such persons may be living, even today, pata Tantra.11 in bodies immortalized through the use of mercurial preparations. Or, like It is with the names that appear in the latter portions of the older lists, the Zawgyi alchemists of Burma, they may have lived fast, died young, and as well as a number of names from the later lists (RRS [b] and [c]), that we left good-looking corpses (for mercury does indeed tend to "pickle" one's appear to move from the realm of legend into that of textual history: many epidermis, in life and after a premature death), prior to realizing a certain correspond to known authors of often extant alchemical texts. As was men- modicum of immortality in the Rasa Siddha lists. tioned in the last chapter, Vyādi (17) is an alchemist whose legend was The most likely explanation is that these unaccountable names, like so known to the eleventh-century Alberuni and two of whose alchemical many of the "begats" of the Old Testament, were fillers in "systematic ge- works were translated into Tibetan in the Tanjur. In addition, the Garuda nealogies" that manuscript copyists dutifully passed on to posterity for the Purāņa praises Vyādi as an authority on gemstones.12 Nāgārjuna's (18) al- simple reason that copyists copy. Or again, they were names that filled the chemical pedigree is solid: we will return to him shortly. Govinda (25) is lotus petals of so many lost alchemical worship mandalas. Elsewhere, my the name of the author of the ca. tenth- to eleventh-century RHT, very remarks concerning the interplay between divine and human Siddhas and likely the earliest extant Hindu alchemical tantra. Carpati-the figure Vidyādharas appear to be apposite: a number of the Rasa Siddhas were whose name appears the most often, after Nagarjuna, in lists of both Nāth

I26 I27 Chapter Five Tantric and Siddba Alchemical Literature

and Rasa Siddhas, is the author of two short post-fourteenth-century also cast as divine revelations transmitted by human authors who chose to vernacular alchemical works, entitled Carpat Rasāyan and Carpat Nāth jī ke remain anonymous. One of these presents us with no problems, however. slok-as well as of two lost Sanskrit works, entitled Svargavaidyakāpālika This is the Kākacandeśvarīmata (KCM), a text revealed by Siva to Kākacaņ- and Carpati-siddhānta. Conclusive textual evidence (a local vamšāvalī; and deśvarī-the "Fierce Crow[-faced] Goddess"-and clearly the text attrib- the Grub thob, which names him as Vyādi's guru)13 indicates that a Carpati uted to the alchemical author Kākacandīśvara (no. 33 in RRS [a] and [c]).19 lived in the kingdom of Camba (Himachal Pradesh) in the tenth century; The two other such works are more problematic. The Rasopanisat (RU) is, if this figure was the author of the two Sanskrit alchemical works men- according to Siddhinandan Misra, the work of a certain Kāpalika master, tioned above, he would have been one of the earliest of the Hindu alche- which evokes the name "Kapalika," the twenty-fourth Rasa Siddha in the mists." Carpati is also the subject of alchemical legend in both Tibetan RM/RRS [a] lists. In its first chapter, the Rasopanisat itself states that it Buddhist and south Indian traditions and is cast as an alchemist in a Sikh is nothing other than an abridgement, in eighteen chapters, of an earlier work, the Pran Sankali.15 work, in twenty-nine chapters, entitled the Mabodadhi.20 Now, there is a By far the biggest name-dropper among the alchemical authors was So- tradition according to which a certain Kapālī (no. 5 in the RM/RRS[a] list) madeva, whose RC was mentioned a moment ago. In addition to naming authored an alchemical work entitled Rasaraja-mabodadhi. If this is the Bhāskara, Sambhu, Govinda, and Nāgārjuna, he also mentions Nandi (no. case, then the Rasopanisat author who epitomized the Mabodadbi in his 30 in the RRS [a] and [c] lists) and Bhāluki (no. 38 in the RRS [a] list), as work could have been a "Kāpālika"-i.e., a follower of Kapālī rather than well as Manthanabhairava, Svacchandabhairava, Bhairavācārya, Srīkantha, a member of the "skull-bearer" Saiva sect-or both. A lost work, simply and others. Srikantha, whose name appears in none of the lists passed in entitled Mabodadhi, is attributed to a certain Sivanāth Yogi; if this is the review here, is simply another name for Siva: the Pasupata teachings were work to which the Rasopanisat refers, then Misra's theory is untenable (un- revealed to Lakulīśa by Śrikantha (or Nīlakantha). He is cast as codiscus- less Śivanāth Yogi was a Kāpālika).21 sant (with the goddess Uma) in an alchemical work that has recently been The last work in this group is the most tantalizing of all, given that the edited under the title of Goraksa Sambita: it is likely to this work that Soma- anonymous Rasarnava is truly the summum of tantric alchemy. According deva is referring.16 Although Manthanabhairava (no. 32 in RRS [a] and [c]) to P. V. Sharma, the history of this text is intimately connected to a Yadava does not appear in the Rasa Siddha list of the RM, this work mentions king named Singhana (1210-47), who was one in a series of great royal him elsewhere and, after the fashion of nearly every other post-eleventh- patrons of the medical and alchemical arts.22 Sharma maintains that a cer- century alchemical work, describes an elixir called manthānabbairava rasa. tain "Bhairavānanda Yogi"-whom the RRS (16.126) names as having indi- According to P. V. Sharma, Manthanabhairava would have been the author cated (vinirdistā) a preparation called the "Lokanātha packet" (potali) to a of the Anandakanda (AK) and personal physician to the king of Simhala King Singhana-was the author of the RA. Three manuscript copies of (Sri Lanka?). I have found no evidence to support this claim. On the one the RA, held by the Anup Sanskrit Library in Bikaner, carry the following hand, the AK as a text is at least a century younger than any of the refer- colophon: "thus concludes Isvara's alchemical dialogue ... the great trea- ences to Manthanabhairava found in other tantric alchemical works; on tise [entitled] Rasārņava ... [as] uttered by Yogānanda."23 By metathesizing the other, the AK's chapter colophons state that it was revealed by the two components of this latter name, one arrives at Ananda-yoga. The Mahabhairava.17 The eleventh- to twelfth-century Manthānabhairava work is, moreover, cast as a dialogue between Bhairava and Bhairavī: by Tantra is, however, one of the most important works of the Western Trans- juxtaposing the name of the divine revealer of this work to that of the figure mission, and one that is moreover linked to the Goraksa Samhita.18 As who recounted it, one could conceivably generate the name Bhairava- we will see shortly, a RM reference to a text entitled Manthanabhairava änanda-yog[i]. If, however, we accept Sharma's argument that the Bhaira- may be a clue to the presence of an eponymous figure in the Rasa Siddha vänanda Yogi who indicated an Ayurvedic preparation to Singhana was also lists. the author of the RA, this pushes the date of this work forward to the early Apart from the AK, three other important tantric alchemical works are thirteenth century, which is chronologically impossible.24 The RRS passage

128 129

Chapter Five Tantric and Siddha Alchemical Literature

cited by Sharma, which concerns a preparation called the Vaiśvānara and a number of important tantric texts, including the massive Svacchanda packet, states that "this preparation was declared by king Singhana ... Its Tantra (SuT), quite naturally bear his name.3 It is therefore uncertain application is known [to be the same as that of] Singhana's packet called whether this Rasa Siddha was a human alchemist or an alchemists' god. Lokanātha, which was handed down (vinirdistā) by Bhairavānanda Yogi." Indeed, the alchemical god Rasabhairava, whose iconography is described The operative term here is "handed down": the vi- prefix here indicates in the RA (2.63-64), is an ectype of Svacchanda Bhairava.32 that the Bhairavananda Yogi who handed down this alchemical preparation With this, we return to "signed" alchemical works. Rasendratilaka Yogi was not a contemporary of Singhana.25 It is moreover curious that the (no. 37 in RRS [c]) is the name of the author of the Rasasāratilaka. Vāsudeva preparation the RRS attributes to Bhairavananda Yogi is nowhere to be (no. 42 in RRS [c]) is the purported author of the lost Rasasarvesvara.33 The found in the RA. As I will demonstrate, the RA is an eleventh-century text. name Kriyātantrasamucchayī (no. 43 in RRS [c]), literally the "Compiler It therefore follows that its author is not to be identified with the figure to of the Kriya Tantra," is likely a reference to the ninth- to tenth-century whom Singhana attributes the invention of this preparation. The author Kriyākālagunottara Tantra, an early work on tantric sorcery and healing. of the RA would more likely have lived under the imperial predecessors of This work is, however, revealed by the ubiquitous Śrīkantha.34 Śrīnātha, the Yadavas, i.e., the Calukyas of Kalyanī, whose tenth- through twelfth- who doubles with Srikantha as the divine revealer of the Western Trans- century kingdom covered most of the same western territory of India as mission texts, is also the purported author of an alchemical work entitled did that of the Yādavas.26 the Rasaratna. This work has also been attributed to Manthanabhairava.35 It is much easier to simply identify the RA with its divine revealer, which Ghoda Colī (no. 55 in RRS [c]) is the author of a short Sanskrit alchemical is what I believe Somadeva does when he refers to "Bhairavācarya" and work entitled Ghoda Coli, or Ghodācoli Vakya.36 He is also identified with what Vagbhatta II (RRS [a] and [c]) does when he lists Bhairava as its Colināth, the founder of the Ãi panth, one of the original twelve Nth sub- twenty-ninth Rasa Siddha. The evidence of the RRS lists is all the more orders.37 We have already noted the south Indian traditions concerning compelling for the fact that the Rasa Siddha whose name immediately pre- Allama Prabhu (also known as Prabhudeva, no. 52 in RRS [c]), a founder cedes that of Bhairava is Rasankuśa: throughout the RA, Bhairava describes of the Virasaiva sect and considered, in Tamil traditions, to be an incarna- the worship of his own and the Goddess's alchemical forms. Whereas he is tion of the alchemist Nandi.38 called Rasabhairava, the "Mercurial Bhairava," she is always referred to as Ratnākara and Nityanātha (nos. 36 and 48 in RRS [c]) may be one and Rasankusī, the "Elephant Goad of Mercury." It follows that the two should the same figure: Nityanätha is the author of the thirteenth-century Rasa- appear together in the Rasa Siddha lists.27 ratnakara, the same text from which Vagbhatta II would have copied his The RRS ([a] and [c]) also lists a third Bhairava-named Svacchand- second ([b] = RRS 6.51-53) list of Rasa Siddhas verbatim. The fifth di- abhairava-as a Rasa Siddha (no. 31). This figure is cited as an alchem- vision of the RRA, entitled "Siddha Khanda" or "Mantra Khanda," is the ical author in the thirteenth-century RPS of Yasodhara Bhatta and the manifest source of the KPT of Siddha Nāgarjuna, a work on tantric sor- sixteenth-century Rasakalpa.28 In its chapter colophons, the so-called al- cery.39 Like the "Mantra Khanda," the Sabaracintamani, which treats mainly chemical Goraksa Samhitā calls itself (among other things) the Svacchanda- of tantric sorcery, is also a work of Nityanatha Siddha, "son of Pārvati"; saktyavatāra ([The Revelation] Brought Down by Svacchanda Sakti). A in it, the author states that the teachings of his work were originally ex- number of references are made in this text to the cult of Svacchanda pounded by Matsyendra.40 In addition to being listed as a Rasa Siddha in (Bhairava), and the opening chapter intimates that this is an abridgment of the RRS, Nityanatha is also listed, in the HYP (no. 19), as one of the yogic a greater alchemical work in twenty-five thousand verses originally re- Siddhas who had "broken Time's rod." Was Nityanatha, as his name and vealed by Svacchanda.29 As already noted, however, this work is a dialogue his writings would seem to indicate, a Nath Siddha? Chances are that he between Srikantha and the goddess Uma. Moreover, in the final verses of was. Certain manuscript colophons to the Siddha Siddbānta Paddhati, gen- a number of manuscripts of this work, the "Nath [named] Goraksa" is erally attributed to Gorakh, call it a work of Nityanatha Siddha.+ If this is mentioned.30 Like Manthanabhairava, and Bhairava tout court, Svacchan- the case, then Nityanatha, whose RRA is dated to the thirteenth century, dabhairava is first and foremost the name of an important tantric divinity; was a Näth Siddha polymath who, not unlike his near contemporary Gor-

130 131 Chapter Five Tantric and Siddba Alchemical Literature

akh, would have been well versed in the alchemical, yogic, and tantric dis- is of another order. So, for example, we find a sizable corpus of works on ciplines. It may be, on the other hand, that Nityanätha Siddha and Nitya- tantric sorcery-works that may be classified under the general heading nātha, son of Pärvati, were two distinct persons, in which case the double of Kriya Tantras-which include chapters on or references to alchemical listing-as Nityanatha and "Ratnakara" in the RRS [c]-would be an accu- preparations and procedures.56 The earliest of these is the Kriyākālagunot- rate one. tara Tantra, a work dating from the ninth to tenth century. This text, which The authors of approximately half of the major extant tantric alchemical takes the form of a dialogue between Srikantha (identified with Siva) and works are found to correspond to names figuring in the Rasa Siddha lists. Kärttikeya, is a work on tantric healing through the use of mantras, amu- These texts are the Rasabrdaya Tantra of Govinda, the Kākacandeśvarīmata lets, rituals and sorcery.57 Another later work, the Yogaratnavali of a Saiva and Kākacandīśvara Kalpa Tantra attributed to Kākacandīśvara, the Rasendra master named Srīkantha Siva, is a so-called "poison tantra" dating from Mangala of Nāgārjuna, the Bhūtiprakarana (or Svacchandašaktyavatāra or perhaps the fifteenth century. Its sixth chapter contains alchemical data.58 Goraksa Sambitā) attributed to Goraksa, and the Rasaratnākara of Nityā- From the same period is another work on tantric sorcery, the Kautukacintā- nātha. mani of Pratāpadeva, which contains a significant amount of data on gener- Major extant alchemical texts by authors whose names do not figure in ally "magical" alchemy-potions for invisibility, magical flight, attraction, the Rasa Siddha lists all date from the thirteenth century or later and de- projection, etc.59 crease in the originality of their content with the passing centuries. More- Two texts dating from the twelfth to thirteenth century-the KPT and over, following a paradigm shift that occurred in the thirteenth to four- the shorter Yogaratnamala (or Aścaryaratnamālā)-are both the work of a teenth century, the content of most of these later works tends to shade Siddha (or Srīman) Nāgārjuna. Their content, moreover, is so similar that from an emphasis on tantric elixir alchemy towards either therapeutic or they are likely the work of one and the same author: the former work, purely transmutational ends. These works include the ca. 13th-century Ra however, is little other than a reworking of a portion of the RRA of Nitya- sendracūdāmaņi of Somadeva, Rasaprakāša Sudbākara of Yaśodhara Bhatta,42 nath Siddha.60 Both the RRA and the KPT postdate the Dattātreya Tantra and Rasapaddhati of Bindu;13 the 13th- to 14th-century Rasaratnasamuc- and the KCM; all three cite their sources in their opening verses; apart chaya of Vāgbhatta II and Rasādbyāya of Kankāla Yogī;4 the 14th-century from a nearly identical list of four works (in which both give "Kālacandes- Rasarājalaksmi and Rasasindhu of Visnudeva or Vitthala;45 14th- to 15th- vara" for Kakacandesvara), the RRĀ and KPT both add another twenty-four century Rasasāra of Govindācārya;46 the 15th-century Rasaratnadīpika of sources not mentioned in the Dattātreya Tantra.61 A final text belongs to a Rāmarāja;+7 the 15th-century Rasacintāmani of Anantadeva Sūri;18 the 15th- category all its own. This is the ca. thirteenth-century Matrkabbeda Tantra century Rasendracintāmani of Dhundhukanatha;4 the 15th-century Rasa- (MBbT), a Sakta-Saiva work on the tantric worship of Siva and the God- sanketakālikā of Camundakāyastha;50 the 15th- to 16th-century Sivakalpa- dess, whose first, fifth, and eighth chapters contain original alchemical data druma of Sivanātha;51 the 15th- to 16th-century Rasarājasiromani of Para- mainly pertaining to the fabrication of mercurial lingas.62 This is also the surāma;52 the 16th-century Rasakalpa, Paradakalpa, Gandbakakalpa, and sole alchemical work under discussion here to betray an east Indian origin. Dhatukalpa (all of which spuriously claim to be portions of the Rudrayāmala Also containing data on tantric alchemy are a number of medieval com- Tantra, and authored by Nāgārjuna)53 and Rasapradīpa of Prānanātha;54 and pendia and encyclopedias, all of which are of western Indian origin. These the 17th-century Rasendrakalpadruma of Ramakrsnabhatta.55 The reader include the ca. A.D. 1040 Rājamārtanda of Raja Bhoja of Mālava; the A.D. will have noted that a significant number of these texts are the work of 1131 Mānasollāsa of the Cālukya king Someśvara III "Bhulokamalla";63 the authors whose names end in Yogi or Nath. There is, however, little or no fourteenth-century Sarngadbara Paddhati;64 and the fourteenth-century internal evidence to indicate that any of these-with the possible excep- Sarvadarśana Samgraba of Mādhavācārya. Those portions of the Mānasol- tions of the RRA of Nityanatha and the Rasendracintamani of Dhundhuka- läsa that are devoted to yoga closely resemble passages of the Yogasīkhopani- natha-were in fact the work of Nath Siddhas. sad, which is one of nine works from the south Indian corpus of 108 Upani- In addition to these works on Hindu alchemy, a number of other sources sads that borrow visibly from works by Gorakh and other Nāth Siddhas.65 lend an important place to alchemy, even though their main subject matter Also contained in this work are discussions of elixir (rasāyana) therapy (in-

132 I33 Chapter Five Tantric and Siddba Alchemical Literature

cluding the technique of kutīpravesa, "entering into the hut"), and the culti- none of his poems in the Nath Siddha anthologies treat of alchemy, the vation of a divine body, possessed of the siddhis of flight, etc.66 figure of Dattātreya (7) does in fact bear a tenuous connection to the al- chemical tradition: the Dattätreya Tantra, a ca. twelfth-century work attrib- uted to him, contains a chapter on "magical" alchemy. A lost alchemical 2. Matsyendra and Gorakh: The Nath Siddha Literature work perhaps attributed to him is the Rasaratnāvali of Guru Datta Siddha.72 Belonging to Dattätreya's Maharashtran lineage is Revana [Siddha], to The two greatest modern contributions to the historiography of the ver- whom two lost alchemical works have been attributed: the Rasadarpana and nacular literature of the Nath Siddhas have been a critical edition of the Rasarājaśiromani.73 mystic poems of Gorakhnath, compiled by Pitambaradatta Barthwal in Major Näth Siddhas who are conspicuous by their relative absence from 194267 and that of a corpus of poetry attributed to a number of other Nath these vernacular anthologies are Ādinātha (no poems), Matsyendra (one Siddhas, compiled by Hazariprasad Dvivedi in 1957. Both editors, who poem), and Caurangi (four poems). The "Prān Snkalī," Caurangi's contri- consulted a number of manuscript sources in the preparation of their bution to the standard anthologies is, however, exceptional in its length, works, found that "standard anthologies" of the poetry of Gorakhnath on content, and antiquity. The literary productions of Matsyendra and Nāgār- the one hand and a number of later Nath Siddhas on the other have existed juna are nearly wholly restricted to the Sanskrit medium. since at least the fourteenth century and that these standard antholo- There are a number of Sanskrit works attributed to Ādinātha. As with gies have not been greatly altered over time. There is, moreover, general Dattätreya and Srīkantha, however, it is difficult to determine whether we agreement that the language of these poems appears to be a blend of Din- are to take these as the works of a historical author, or rather the received gal (Old Rajasthani), western Punjabi, Khadī Bholī, and Apabhramśa.68 revelations of Siva, the "Lord of Beginnings." Some of these works belong Following Gorakhnath, whose literary output (at least by attribution), in to the Western Transmission;74 however, the principal work attributed to both the medieval vernaculars and Sanskrit media, dwarfs that of any other Ādinātha, as far as our interests are concerned, is the ca. fourteenth- Nath Siddha or, for that matter, nearly any other medieval poet, we find century Khecarī Vidya, the "Aviator's Science." This Sanskrit work, which the following frequently recurring "signatures" on the medieval mystic po- places itself squarely in the hathayogic tradition of the Nath Siddhas, is etry of the order. These authors, in order of the volume of their antholo- unique in that its final chapter is devoted to a description of the alchemical gized literary production, are (1) Gorakh, (2) Jatī Hanavant, (3) Carpați, complement to yogic practice. As such, it is an exemplary work of Siddha (4) Gopīcand, (5) Bhartrhari, (6) Mahādev, (7) Dattā[treya], (8) Ghodā alchemy.75 Colī, and (9) Jālandhara/Hādi-pā. None of the eight works76 attributed to Matsyendranäth-either in San- Two of the names listed here-Jati Hanavant (2) and Mahādev (6), who skrit or vernacular languages-contains any alchemical data. This not- figure in none of the lists of the nine Nath Siddhas-will not concern us withstanding, his teachings are recognized by a vast number of later tantric here. A single verse from a poem by Bhartrhari (5) is alchemical.69 There authors as fundamental to their own syntheses, as I demonstrate in this are no alchemical works attributed to either Caurangi, Kanerī, or Gopī- and later chapters. A mention of the Siddha[-amrta] Kaula in Matsyendra's cand (4), nor do alchemical references appear in any of their mystic poems. masterwork, the KJnN (16.47a, 48b; 21.4-7), is significant inasmuch as All do, however, have much to say about hatha yoga, and alchemical motifs Siva-Bhairava is praised as the leader of the Siddha Kaula in an opening recur in all of their legends. The sparse extant alchemical works of Carpati verse of the RA (1.4b); the same alchemical work (18.228a) closes with- (3) and Ghoda Coli (8) have already been surveyed; a poem contained in and perhaps takes its title from-a verse it appears to borrow from the Dvivedi's anthology, the "Srī Carpat jī kī sabadī," is a short treatise on the KJiN (15.7a). It also mentions worship of the Siddha Cakra (RA 14.24), a subtle body. Carpati also figures in a number of alchemical legends.70 The tantric "systematic genealogy" in which Matsyendra figures prominently.77 two short poems by Naga Arjan found in Dvivedi's anthology contain no In this, we may see that, in addition to his pivotal role within the broader references to alchemy,71 and it is doubtful that this is the same figure as any tantric tradition, Matsyendra also constituted a bridge between main- of the candidates for the historical alchemist named Nāgārjuna. While stream works of the Hindu tantric tradition and the alchemical tantras.

134 135 Chapter Five Tantric and Siddha Alchemical Literature

Matsyendra's K7nN is itself a synthesis of the Siddha and Yoginī Kaula brows), to which he adds the crowning brabmarandbra;86 he notes the tech- traditions, inasmuch as it portrays the latter as emerging out of the for- niques of prānāyāma and kapālabheda (20.2b-3a), kbecarī mudrā (6.18-19), mer.78 The Mrgendragama, a ninth- to tenth-century Saivasiddhānta work, "binding" and "piercing" (4.11); and one may discern in matsyodara ("fish- casts the two as separate but equal groups: "The sages know of eight [other] belly"), the name he gives for his sect in the fourth and present age, the currents, connected respectively to Siva, the Mantreśvaras, the Ganas, Kali Yuga, an occult reference to the hathoyogic technique of diaphrag- Gods, Rsis, Guhyas, the Yoginī Kaula and the Siddha Kaula ... The Yo- matic retention (kumbhaka).87 There is, moreover, no conclusive evidence ginīs received a wisdom that immediately causes yoga to shine forth. It was to prove that Matsyendra's Yogini Kaula and KJnN are later than the West- called yoginīkaula because it never went beyond the limits of their circle. ern Transmission's Kubjika cult and KM. On the one hand, the earliest The same is the case for the other [i.e., the Siddha Kaula current]." This manuscript of the KJnN is older than the oldest Kubjikā manuscripts;88 relationship between Matsyendranäth the Siddha and the yoginīs of Kāmar- Abhinavagupta mentions Macchandanatha89 and the KM;90 and both Ab- üpa (Assam) becomes the subject of a rich body of Näth Siddha legend in hinavagupta91 and the KM92 mention Matsyendra's Yogini Kaula. On the later centuries.79 other hand, Matsyendra makes no direct mention of either the goddess Matsyendra also constitutes an early and vital link between two peren- Kubjikā, the Western Transmission, or Abhinavagupta's Trika Kaula. nial heartlands of tantric practice: eastern India (Assam, Bengal, Orissa) Matsyendranath's pivotal role in the history of tantric sectarianism and Nepal. While his KJnN was revealed in Kamarupa, both the oldest became further expanded when the myth of his "recovery" of the original manuscript of the work and the oldest mythology of Matsyendra hail from tantric teachings of Siva related in the K7ñN (16.27-36) became appro- Nepal.80 Nepal has also long been the homeland of the Western Transmis- priated as the origin myth of the Eastern Transmission (pūrvāmnāya), that sion and its cult of Kubjika, and it is in the Kathmandu Valley that nearly tradition within which the Trika Kaula scriptures were classified. It is all the important manuscripts of the Western Tradition scriptures are moreover through an appropriation of this myth that Abhinavagupta was found. A great number of other data, some of which I have already noted, able to reincorporate-into the "secret worship" of the same Trika Kaula further link Matsyendra (and later Gorakh) with this latter Kaula tradition. whose reforms had removed such practices from the public sphere- Both the revelations of Matsyendra's Yogini Kaula and the Western Trans- the orgiastic, power substance-based observances that had been inherited mission were "brought down" at the "lunar" site of Candra-dvīpa (or from the Kapalika practices associated with the scriptures of the old Vidyā Candra-giri)81-even if, in Matsyendra's case, Candradvīpa is to be iden- Pītha. This he did by adapting into his synthesis the diagrammatic render- tified with coastal Assam and the Western Transmission's Candradvīpa or ing of that Eastern Transmission "systematic genealogy" known as the Sid- Candragiri with coastal Karnataka. The YSA locates Gorakh's birthplace dha Cakra, the "Circle of the Siddhas." This diagram was used as a support at Candragiri, "on the shore of the Godavari River" in western India.82 for the tantric worship of the Kaula lineage deities (Kuleśvara, Kuleśvarī, The K7nN of Matsyendra and the Western Transmission's KM contain and the eight Mother Goddesses together with their Bhairava consorts), the earliest extant references we have to the six cakras of hathayogic prac- as well as of the yuganathas, the "Naths of the Cosmic Ages," i.e., the Kaula tice. Whereas the earliest mention of the six cakras of Hindu83 hatha yoga masters who taught in the four yugas. Of these four mythic Naths, each of appears in an eighth-century A.D. literary source, the Mālatī Mādhava of whom bore an animal name, the fourth and last, associated with our own Bhavabhuti,84 a number of scholars maintain that it was in such Western Fourth Age was Macchanda (i.e., Matsyendra).93 While Macchanda was Transmission works as the Kubjikāmata (KM) that these practices were first identified with the northern direction in the Siddha Cakra, his female con- discussed systematically.85 Here, I offer evidence to the contrary. In fact, sort was named Konkana, whose name again evokes the region in which Matsyendra's KJiN gives a more extensive account of the subject than does Śrinatha and the "Nine Naths" founded the Western Transmission. The the KM. In KJnN 17.2b-4a, Matsyendra lists the six cakras by their loca- Siddhas as a group are placed in the southern quadrant of this diagram, tions i.e .: (1) gūdha (anus), (2) gubya (genitals), (3) nābhi (navel), (4) brd and the siddhaugha ("Siddha stream") in the western quadrant.9 We may (heart), (5) ghantikägrantha (throat), and (6) bhrümadhye (between the eye- therefore see in this diagram evidence both for the tantric reappropriation

136 137 Chapter Five Tantric and Siddba Alchemical Literature

of the preexisting cult of the divine Siddhas and for the incorporation of offerings and the communal consumption of blood, flesh, wine, and sexual certain human or divinized Siddhas, led by Matsyendra, into that elite fluids),99 Abhinavagupta's later synthesis toned down the grimmer side of circle. the public cult while promulgating a "sublimated" form of its erotico- The Trika Kaula considered Macchandanätha to have been the founder mystical practices for public consumption.100 (The notion of the kula nev- of the kula, the line of transmission of the Kaula tantras. This is explained ertheless remains operative in the Kaula context. The Kaulas are those per- in a myth according to which Macchanda fathered twelve sons, sending six sons who belong to a kula, i.e., to a clan lineage, a particular tantric line of them out (from Srisailam) into the Indian subcontinent to found six of transmission through a series of masters and disciples, and who make Kaula orders in the Fourth Age: these are the six ovallis ("currents of doc- use of the power substance of the Goddess's sexual emission, also called trine") of the Trika Kaula.95 Elsewhere, it is likely that Matsyendra's teach- kula).101 Abhinavagupta's reinterpretation of the kula practices turned espe- ings also influenced the development of the Buddhist Kāpālika-Yoginī cult cially on the matter of sexual orgasm and the use of its by-products: within the highest yoga tantra (anuttarayoga) current of Vajrayāna Bud- While its principal purpose in the Vidyapitha was to produce the dhism.6 power-substances needed to gratify the deities, here the ritual of cop- To what are we to attribute Matsyendra's pivotal role? The answer to ulation is aestheticised ... the emphasis has now moved to orgasm this question may be elicited from a passing reference made by Abhina- itself. It is no longer principally a means of production. It is a privi- vagupta to the distinction between the earlier Kula practices of primitive leged means of access to a blissful expansion of consciousness in tantrism, and the later Kaula traditions, for which Matsyendra's reforms which the deities of the Kula permeate and obliterate the ego of the would have constituted a watershed. In the words of Alexis Sanderson, worshipper.102 The distinction between Kula and Kaula traditions ... is best taken Even as he plays to the Kashmir Saiva gallery, Abhinavagupta does to refer to the clan-structured tradition of the cremation-grounds in fact retain the concrete use of sexual fluids in the secret ritual of those seen in the Brahmayāmala-Picumata, Jayadratha Yamala, Tantra- initiated into the Trika Kaula.103 In certain cases, he prescribes, for the sadbbāva, Siddhayogeśvarīmata Tantra, etc. (with its Kāpālika kaulikā attainment of siddhis, the consumption of sexual fluids after intercourse. vidhayab) on the one hand and on the other its reformation and do- Here, the partners are to pass these fluids from mouth to mouth prior to mestication through the banning of mortuary and all sect-identifying placing them into a collecting vessel as an offering to the gods of the tan- signs (vyaktalingatā), generally associated with Macchanda/Matsy- tric "sacrifice."104 In the Siddha Cakra worship of the yuganathas, the offi- endra.97 ciant offers to Bhairava (with whom he has identified himself), as well as In other words, Siddha Matsyendra, founder of the Yoginī Kaula, shifted to the circle of energies that surround him, by drinking a mixture of male the emphasis of early tantrism away from the "terrible" practices and clan- and female sexual fluid from a sacrificial jar.105 Noteworthy here is the gloss based (Kula) system featured in the scriptures of the Vidya Pitha, and given by Jayaratha, the thirteenth-century commentator on the TA, for towards the erotico-mystical practices that became the bedrock of later the term siddbi: he identifies this term with pindasthairya, "stability of Kaulism. In the ninth and tenth centuries, we witness the emergence and the body," which is precisely the watchword of the alchemical RA (1.14b, three-stage evolution of Trika Kaulism, which reached its peak of refine- 18a, 28b) and the hathayogic works of the Näth Siddhas. In the same com- ment, in the eleventh century, in the exegetical summa of Abhinavagupta mentary, Jayaratha refers to the fluids produced by sex as "the best of elix- and his disciple Ksemarāja. While publicly carrying Matsyendra's reforms irs (rasāyana)"106 to a still higher level of abstraction (in order to render Trika ritual accept- Elsewhere, Abhinavagupta employs alchemical imagery to describe the able to the broader Kashmiri Saiva community), the Trika reformers pre- abstract states of consciousness ideally realized by the practitioner. So, for served, as a cult of their virtuosi,98 the erotic ritual of Matsyendra's Yoginī example, he compares the attainment of absolute consciousness with the Kaula, described in the KJiN. Whereas early (pre-A.D. 800) Trika and the transmutation, by mercury, of base metals into gold; and speaks of the K7iN both emphasized the cult of yoginīs (who were to be invoked with burning away of all difference within the "stomach of consciousness" in a

138 139 Chapter Five Tantric and Siddha Alchemical Literature

way that draws an analogy between digestion in the gross body, the effects in the realm of ritual mantra), which was subsequently adopted into the of hatha yoga and alchemy on the subtle body, and the gnoseological iden- Śrīvidyā cult of the Southern Transmission (daksināmnāya).111 tification between the finite and absolute self.107 Elsewhere, he invokes These intellectualized, abstract forms of late "high" Hindu tantrism- Matsyendra's "fish-belly" technique, stating that the yogin is to "repose which have, of late, defined the field of western scholarship on Tantra- devoted to the condition of the belly of the fish." 108 fall outside the purview of the present study."12 We will rather concentrate In this, the Trika Kaula theoreticians, even as they effected their re- on the Siddha perpetuation of the tantric legacy, a tradition whose three formation of Hindu tantra, remained in touch with the concrete ground of components are neatly summarized by the great tantric practitioner- the tantric enterprise, which was and remains human sexual fluids and their scholar Gopinath Kaviraj: "Some ... were accomplished (siddba) in the al- symbolic correlates. The cosmic force that activates and actualizes every chemical path (rasa-märga), some accomplished through batha yoga, and facet of tantric practice-that originates from the womb of the Goddess still others had perfected themselves through tantric practices or through and passes through every link in the chain of transmission, from guru to the use of sexual fluids (bindu-sadhana).""13 Herein lies the "Siddha distinc- disciple and thence to his or her disciple, via the inner channels of yogic tion." Whereas the tantric synthesis generally ignores alchemy and rele- transformation and through the channels that energize the mystic dia- gates hatha yoga to a secondary role in its hierarchy of practice while plac- grams that serve as supports for worship and meditation-is ultimately ing a very high premium on worship through the use of real or sublimated nothing other than a stream (ogha) or flow (srotas) of sexual fluid. The life sexual fluids, the Siddha traditions privilege all three forms of comple- and structure of the tantric family or clan (kula) is defined by the life- and mentary practice. The guiding principle here remains one of controlling immortality-giving flow of the clan essence (kulamrta) that is transmitted, a universe that is understood to be a body, the body of the divine consort concretely and in the form of sexual fluids, in tantric initiation and worship of Siva, the body of one's own consort, and the feminine in one's own rituals. This is attested in a wide array of sources. In his KJnN (22.2-3), body. In the Siddha play of analogies between microcosm and macro- Matsyendra speaks of the transmission of the kaulamrta through the six cosm, and the corresponding hierarchies of the interpenetrating realms yoginīs; elsewhere, his entire twenty-first chapter, in which he discusses of universal being, this body is at once a divine, human, and alchemical the kaula-sadbhava ("true essence of the Kaula"), is of the same order: this body, to be perfected through yogic, alchemical, and erotico-mystical term as well is used to designate the yogini's sexual emission.109 Down to practice. the present day, Assamese tantrikas identify their "lineage nectar" (kulam- Following Matsyendra's lead is Gorakh who is, in nearly every Nath rta) with the goddess's menstrual fluid or the commingled sexual fluids of Siddha lineage reckoning, called the disciple of Matsyendra/Mīna. While Siva and the Goddess.110 Gorakh certainly embraced and greatly expanded upon the hathayogic pole If then the preservation of the universe depends upon-indeed, is noth- of Siddha practice, his relationship to those of erotico-mystical ritual and ing other than-the endless cosmic orgasm of the divine, and if the bliss alchemy is more ambiguous. In spite of the fact that the historical Gorakh of orgasm is that human experience which is closest to the very being of and Matsyendra lived at least three centuries apart, a rich body of legend godhead, then the stuff of orgasm-male semen and the female sexual describes the freeing of Matsyendra, by Gorakh, from sexual bondage in a emission and uterine blood-will, of necessity, play a vital role in the tan- Kingdom of Women in Assam. As I will demonstrate in a later chapter, tric quest for divine autonomy, immortality, and power. No matter how this is a narrativization of the reforms Gorakh effected within the (Nath) abstract and refined tantric practice has become over the centuries follow- Siddha tradition, which he purged of its erotico-mystical elements in favor ing Abhinavagupta's reformation of the Trika Kaula, sexual fluids remain of a nearly exclusively hathayogic emphasis. present, in one or another sublimated form, to every sect or order claiming It nonetheless remains the case that Gorakh is considered to have been tantric affiliation. In the idealist and nondualist syntheses of Abhinava- the author, if only by attribution, of a distinctly Western Transmission text, gupta and Ksemarāja, the sublimation of sexual fluids took the (outer) form entitled the Goraksa Sambita, that was devoted to precisely this sort of of a remarkably complex and coherent semanticization of ritual (especially practice. In fact, there exist no fewer than three major works known by the

140 141

Chapter Five Tantric and Siddba Alchemical Literature

title of Goraksa Samhita, in which we find all three of the building blocks Upanisad.118 A great number of hathayogic works, written in Sanskrit, are of the medieval Siddha traditions: alchemy, batha yoga, and tantric worship attributed to Gorakh. There is a great deal of overlap between these texts, involving the use of sexual fluids. The first of the three works, entitled most of which appear to be elaborations on the so-called Goraksa Sataka, the "Kadi Prakarana" of the Goraksa Sambita, is a thirteenth-century Sākta the "Hundred Stanzas of Goraksa" (of which the critical edition contains work belonging to the canon of the Western Transmission, yet another 201 verses!), which is datable to the thirteenth century."19 These works in- indication of the interconnectedness of this and Nath Siddha traditions.114 clude the Amanaska Yoga, Amaraugha Prabodha, Amaraughaśāsana, Goraksa (Its ninth chapter is attributed to another Nath Siddha, Jalandhara. Like Siddhānta Samgraba, Goraksa Paddbati, Mabārthamanjarī, Siddha Siddhānta the rest of the Kādi Prakarana this chapter is Sakta in its orientation.)115 Paddhati, Vivekamārtanda, Yogamārtanda, Yogabīja, and Yogacintāmani.120 This is a work in twenty-seven chapters which, devoted for the most part As for the "third," hathayogic Goraksa Sambita, this is a work that curi- to tantric worship of the Goddess in a variety of her forms, also lends space ously links Gorakh with Dattätreya, who, in western India and Maharash- to a discussion of hatha yoga, the Siddhas of the Western Transmission, and tra in particular, is more of a god (incorporating Brahmā, Visnu, and tantric cosmology. Śiva-but especially Visnu-in his iconography) than a legendary human Another thirteenth-century work, also referred to by certain authors as figure; and in this he plays a role similar to that of Ādinatha. Yet, there was the Goraksa Sambita, is one of the major texts of the tantric alchemical at least one historical Dattätreya, who authored a number of works on yoga tradition. This, by far the most voluminous alchemical work attributed to and tantra in the twelfth to thirteenth century. In fact, the fourteenth- Gorakh, is variously known as the Goraksa Sambitā, Bhūtiprakarana [BbP], century Sarngadhara Paddhati goes so far as to classify the two major forms Svacchandasaktyavatāra, and Hādibbeda. It is, nonetheless, cast as a dialogue of yogic practice, the "six-limbed" practice and the "eight-limbed" prac- between Srikantha and the goddess Uma, the former being, much like tice, as "Gorakhnathi" and "that of the son of Mrkanda" (a reference to Ādinātha, a pseudonym for Siva in works of the Western Transmission. I Dattätreya, inasmuch as it is this figure who reveals the yogic doctrine discuss this alchemical work at length later in this chapter. Apart from the of the Markandeya Purana) respectively. In the Gorakb Upanisad, Gorakh BhP, a small number of Gorakh's vernacular mystic poems and technical terms the former akula and the latter avadhuta.121 Indeed, there exists a writings in Sanskrit include references to alchemy; curiously, as many of work entitled the Dattātreyagoraksa Samvada (A Conversation between these disparage alchemy as praise it. In the latter category, the sixth Pad of Dattātreya and Goraksa), whose principal subject is yogic practice. This his bānīs presents an allegorized description of the yogic body as a gold- work is known by a number of other names, however. The most common smithy cum alchemical laboratory.116 Another vernacular work attributed of these is the title Avadhuta Gita, which indeed mentions both Dattātreya to Gorakh, entitled the Yogadipika, is written in the Old Kannada lang- and Goraksa in its opening verses.122 However, this dialogue appears uage. I have not consulted this work; however, a description of it leads me to be nothing other than the fifth and final chapter of the hathayogic to believe that it is a south Indian version of the Hathayoga Pradīpikā Goraksa Sambitā (whose third chapter moreover bears a striking resem- of Svātmarāma, a fifteenth-century work that borrows heavily from the blance to the Akulavira Tantra of Matsyendranath). The organization of Goraksa Sataka and other Sanskrit hathayogic works attributed to Gor- this third, hathayogic, Goraksa Sambita also resembles that of the Siddha akh. Unlike the HYP, however, the Yogadipika contains some alchemical Siddhanta Paddhati, another of Gorakh's seminal hathyogic works. The passages.117 Goraksa Paddhati, a manual on batha yoga, also refers to itself, at one point, By far the most important body of teaching attributed to Gorakh, in as the Goraksa Sambita.123 In fact, the alchemical Goraksa Sambitā, i.e., the both the Sanskrit and vernacular media, is that consecrated to batha yoga, BbP (9.133), states that the Goraksa Sambita is a work in five parts.124 In in which field he was India's major systematizer and innovator. Gorakh's the light of the evidence we have passed in review, we must conclude major poetic works, anthologized in the Gorakh Bāni, include the Sabadī, that these were originally independent works which became concatenated Pad, Prān Sankalī, [Machīndra-] Gorakh Bodh, Gyān Tilak, Ātma Bodh, into the Goraksa Sambitā, "Goraksa's Compendium," a fact that points, if Caubīs Siddhi, Sisyā Darsan, Naravai Bodh, and Kafir Bodh. Another im- nothing else, to Gorakhnath's imposing medieval reputation as a Siddha portant vernacular work, not included in this anthology, is the Gorakh virtuouso.

142 143 Chapter Five Tantric and Siddha Alchemical Literature

3. On the "Tantric" Element in Tantric and Siddha Alchemy tantric tradition.129 The content of these works contains references to a number of tantric sects (the Kaulas, Kāpālikas, Paścimāmnāya, etc.), while A number of scholars have maintained that neither hatha yoga nor alchemy a number of alchemical authors and Rasa Siddhas (Nāgārjuna, Nandi- is, properly speaking, a component of Hindu Tantra.125 To the extent that [-keśvara], Goraksa, Carpati, Manthānabhairava, and Svacchandabhairava) mainstream tantric literature appears to turn a blind eye toward the two bear the same names as do the divine or legendary founders of other tan- disciplines, or at least ignores the names of their founders or exponents in tric traditions. their elaborate "systematic genealogies," this contention appears to have Also in the pages of the alchemical sources, we find myriad references an element of truth to it. The matter is, however, of a rather more complex order. On the one hand, a number of tantric sources do mention alchemy to the mystic geography of tantrism, concerning the sakta pīthas of the Goddess as well as the Himalayan and other mountain shrines generally and hatha yoga, either directly or by way of illustrating this or that abstract associated with the cult of Siva. Tantric mantras, yantras,130 initiation,131 concept or process. On the other, the alchemical and hathayogic sources forms of worship and devotion to the tantric gods, and "guruism"132 also themselves often claim to be tantric. Finally, there is the matter of chronol- figure in the alchemical sources. ogy. With the important exception of Matsyendra, who, as we have seen, Lastly, Hindu alchemy is tantric in its goals and in the means it appro- is included in tantric reckonings of lineage founders, nearly all the alchemi- cal and hathayogic teachers and authors postdate the foundational tantric priates to realize those goals. Total autonomy, omniscience, superhuman powers, bodily immortality, and a virtual identification with godhead-al- compendia and their postscriptural systematizations. though not at the expense of one's autonomy-are the aims of the Hindu In his compelling study of Sākta tantrism, Douglas Brooks argues for a alchemist, just as they are of the great majority of nondualist tāntrikas.133 polythetic approach to "tantric" phenomena, in which one is to look to In a more general sense, Hindu alchemy shares the same universe as do family resemblances between the various tantras (texts with the word Tan- many of the other medieval Hindu traditions that identified themselves as tra in their title, sects which claim such texts to be their foundational scrip- tantric and which western and Indian scholars alike today identify with tures, etc.) without taking any single feature or body of features found in tantrism, particularly of the nondualist variety.134 This is a divine, world- such traditions to be normative.126 It is such an approach that I follow here affirming135 universe, the field upon which the godhead fully realizes itself, in my treatment of Hindu alchemy, the discourse of which was as much "in and offers realization to those humans who propitiate it. It is also an an- the air" in medieval India as were the antinomian cremation-ground rites of the Käpälikas and the refined metaphysics of Abhinavagupta and Ksem- thropic136 universe, seemingly created for human self-realization, with man the measure of all things and that creature who is specifically adapted to arāja Here, we will assert that Hindu alchemy is tantric because (a) it says it plumbing the depths of its mysteries. The tantric universe is a pulsating, vibratory universe,137 in which matter, souls, and sound are the stuff of the is; (b) it is tantric in its language and referents; (c) it is tantric in its meta- physics; (d) it is tantric in its practice; (e) it is tantric in its theology; (f) it outpouring of godhead into manifestation, with godhead generally identi- fied with Siva and his self-manifestation or self-reflection taking the form is tantric in its goals; and (g) alchemical terminology is employed discur- sively or metaphorically in tantric works of a more general order. of the Goddess.138 It is a bipolar, sexualized universe,139 in which all change

The alchemical classics are revealed by the same gods (Siva, Bhairava, and transformation are viewed as so many instances of an interpenetration

Śrīkantha) to the same goddesses (Devī, Bhairavī, Sakti, Caņdī[kā]) as are of male and female principles, with metaphysical categories, animals, plants, and minerals all being possessed of a gender marking. It is a ver- the majority of Hindu tantras. Their form and style are tantric, with most tically bierarchized universe, in which that which is higher, closer to the texts consisting of "semidialogues" between god and goddess. Several al- source of all manifestation, is subtler and capable of encompassing, pene- chemical preparations bear the names of tantric deities (Nīlakantha, Ānan- dabhairava, Svacchandabhairava, Lokanātha, etc.).127 A great number of trating into, and reabsorbing into itself that which is lower on the great

alchemical classics contain the term Tantra in their titles:128 others call chain of being. It is a radiating universe,140 with the source of the manifest

themselves Sambitās, or -matas, also appellations proper to the broader world being located at the center of a vast network of metaphysical catego- ries, divinities, phonemes, etc., all of which are interconnected through a

144 145 Chapter Five Tantric and Siddba Alchemical Literature

complex interplay of correspondences. And, ultimately, the tantric universe the power of flight." (5) Ritual worship, initiation, and the use of mandalas, is an emancipating universe, a universe that is primordially and virtually mantras, etc. as necessary adjuncts to the manipulation of alchemical sub- free: born of the boundless playing out of divine consciousness, its every stances are stressed in most works. (6) The alchemical laboratory and its constituent part, including the human body and spirit, as well as brute mat- apparatus (yantras), crucibles (musas), and enclosed "fire-pits" for the ter, are intrinsically free. Tantrism therefore places a high premium on ex- roasting of mercurial preparations (putas), and materiae primae (herbs, perience-bodily, practical, concrete experience-which, in conjunction minerals, gemstones, metals, animal substances, salts, etc.) are described. with knowledge, is liberating.141 Most sources list eight primary (mahärasas) and eight secondary (uparasas) minerals, nine gemstones (ratnas), anywhere from six to twelve metals (lo- bas, dhātus), and (7) eighteen principal alchemical operations (samskāras). 4. Hindu Alchemy: The "Canonical" Works Following this, most alchemical works launch into (8) an extended discus- sion of these operations, by means of which mercury and other minerals Generating a description of the Hindu alchemical canon is a perilous task, and metals are prepared and combined prior to transmutation, (9) followed given the number of Rasa Siddhas whose works are either unknown or not by a discussion of applications of said preparations to the human body of extant on the one hand and the paucity of critical editions of extant al- the alchemist, by which he is rendered immortal. (10) Later works also chemical works on the other. Also clouding the picture is the absence of contain sections on the preparations of various elixirs and aphrodisiacs, to any certitude that the texts which have come down to us are single-author therapeutic ends: this is rogavada, the appropriation of tantric alchemy into works and not compilations made over a period of several centuries. Much Ayurvedic mercury-based pharmacy (rasa šāstra). of what follows will therefore be tentative, given the numerous gray or As is the case with most Tantras, these works are written in a dialogue blank areas remaining in our textual "map" of the Hindu alchemical tradi- style, in which Siva-Bhairava reveals the alchemical gnosis to Devī- tion. To this point, the limited historical treatment I have given of certain Bhairavī. Unlike most Tantras, however, a great emphasis is placed on texts of these major works has been based, for the most part, on data external to (sāstras, granthas) as opposed to knowledge communicated orally by a mas- the texts themselves. Here, the analysis will be based on textual and inter- ter, even if initiation by a guru is described in many of these works. No textual data. The presentation of the "canon" will follow what I feel to be single alchemical Tantra contains all of these elements; all contain some of the most plausible chronological order. In fact, the evidence at our disposal them, and some contain additional elements. Not all of the texts I review indicates that most of the alchemical classics were compiled over a rela- here are, properly speaking, alchemical tantras; in fact, only five of them- tively brief period, with the great majority being generated within a two the Rasārnava, Kākacandeśvarīmata, Rasendra Mangala, Rasaratnākara, and hundred-year span covering the twelfth and thirteenth centuries. Bhūtiprakarana-can be said to belong to this category. What distinguishes Just as exegetes evoke the "five attributes" (pancalaksana) that define a these from the other major alchemical works is their overriding debavāda Purana, so we can point to a number of common features, or family resem- emphasis upon bodily immortality as the alchemist's ultimate goal. Here, blances, that unite the major works of the tantric alchemical canon. These they may be contrasted with the softer line taken by Mahesvara Siddha include (1) an opening statement concerning the divine origins and great- works, which may be qualified as muktivāda (i.e., using increased longevity ness of mercury and sulfur, and (2) a praise of the alchemical path as supe- as a means to the higher end of gnoseological release into the absolute) on rior to all other Hindu forms of religious practice. Here, bodily liberation the one hand and, on the other, the emphasis on dhātuvada and the attain- (jīvanmukti) is compared favorably to the impersonal workings of moksa, ment of siddhis found in a number of late works. The alchemical Tantras, and the complementarity of yogic and alchemical practice is noted. (3) whose watchwords are jīvanmukti and kālavañcana ("skewing Time"), are Many works include origin myths of mercury, sulfur, and other of the works of Kaula inspiration, which implicitly follow the doctrines promul- prime chemical reagents. (4) Many also contain an aphorism in their open- gated in such texts as the KJnN of Matsyendra, the KM, and the Kulārnava ing chapter that is quite identical from one text to another: "Swooned, Tantra, as well as the old Yāmala texts (Rudrayāmala, Brahmayāmala, etc.). mercury ... drives away disease; killed, it revives itself; bound, it affords Generally speaking, muktivāda works take Ayurvedic rasāyana as their

146 147

Chapter Five Tantric and Siddha Alchemical Literature

scientific starting point, are Saiva Siddhänta in their metaphysical and so- of flight.145 This distinction seems to be of a piece with Govinda's general teriological orientations, and feed quite directly into later Ayurvedic rasa outlook, which clearly follows the soteriology of the Mahesvara Siddhas. sastra.142 Many of the major extant works of Hindu alchemy-the Rasabr- Very rarely does Govinda mention jīvanmukti; rather, yogic and alchemical daya Tantra, Rasopanisat, Rasendracūdāmaņi, Rasaprakāśa Sudbākara, Ras- practice produce a stable body with which one may accede to the highest aratnasamucchaya, and Anandakanda-fall into this category. That is, the end of liberation, i.e., absorption into the absolute, through knowledge of preponderance of data found in these works are rogavāda and muktivāda, the absolute (brábman).16 although, as we will see, they are not wholly bereft of references to dhātu- Govinda lists the usual eighteen samskāras (2.1-2) and eight uparasas vāda and debavāda alchemy. (9.5), but only six mahārasas (8.4). He lists nine metals (9.5-6). His descrip- Tantric works emphasizing this-worldly attainments and powers (dhātu- tions of the first eight samskäras147 are very brief, with most of the body of vāda, siddhis) generally take their inspiration from the old "Kriyā Tantras," the work being devoted to the preparation of mineral "essences" (sattvas) works on tantric sorcery. None of the major alchemical works falls into this which, combined with mercury, serve to transmute base metals into gold category, although nearly all contain references to sorcery and attendant or silver. The final chapter is devoted to the priming of the human body practices. Two works that may be classified as Siddha alchemical works are for the ingestion of transmuting mercury, and the preparation of four mer- the Matrkabheda Tantra [MBbT] and Khecari Vidya [KhV]. The former, curial elixirs, in the form of pills (gutikās). Nowhere does Govinda assert rather than being an alchemical Tantra per se, is a Sakta Tantra tout court, that such elixirs lead to bodily immortality: good health, a loving wife, and but one that devotes significant space to alchemy and batha yoga. As its the birth of sons are as much as he promises.148 In many ways, the RHT, name indicates, the latter is a work on the hathayogic technique known as with its sober, taciturn style, is a prolegomenon to the later alchemical khecarī mudrā; however, since the term khecarī ("one who moves in space," tantras, which pick up where it leaves off. the "aviator") also has a number of tantric and alchemical applications, we In the final verses of the RHT, Govinda praises his king, "Srī Madana, include it here. scion of the Lunar Lineage (sītāmśuvamsa) born into the Haihaya clan ... It has generally been held that the RM of Nagarjuna, by one or another lord (nātha) of the Kirātas" as a master of alchemy (rasācārya) whose al- name, is the oldest extant tantric alchemical work. In the preceding chap- chemical gnosis (rasavidyā) was not learned, but innate. "I, bhiksu Govinda, ter, I advanced arguments to refute this misconception. I now reinforce having learned much from that king of the Kiratas, am the author of the those arguments with additional material from other alchemical texts. Tantra called Rasahrdaya."149 A number of historians of Indian medicine a. The Rasabrdaya (Heart of Mercury) Tantra (RHT) of Govinda,143 a maintain that Govinda's Madana was a tenth-century king of "Kirāta-deśa," work in nineteen chapters dating from the tenth to eleventh century, is a blanket term, going back to the epics, for the Bhutan-Assam border very likely the oldest extant Hindu alchemical work. As such, it is unique region; however, a number of Puranas locate the Kirätas in the Vindhya in a number of ways. To begin, Govinda, unlike all later authors, first (1.2) mountain region.150 What Govinda's RHT passage in fact states is that Ma- identifies the combination of sulfur (or mica) and mercury with Visnu and dana is a Somavamsin and Haihaya by birth, lord of the Kirātas by princely Śiva (Hari-Hara), before making the standard identification (1.33) between avocation, and an alchemist by vocation. Now, there did exist, in the ninth sulfur-mica as the menstrual-sexual emission of the Goddess (Gauri) and through thirteenth centuries A.D., a royal family called the Kalachuris who mercury as the semen of Siva. He praises the Rasa Siddhas (1.7), but does traced their lineage back through Haihaya and Yadu to the Moon and who not name any of them. Govinda is himself the twenty-fifth Rasa Siddha therefore considered themselves to be Haihayas and Yadavas of the Lunar in the RM/RRS [a] list; the RM also appropriates his verse 1.2, verbatim, Dynasty. The Kalachuris ruled from Tripura (modern Tewar, north of Ja- as the opening verse of its final chapter: the RHT is therefore older than balpur) over a kingdom whose borders more or less correspond to those of the RM.144 modern-day Madhya Pradesh. Their influence, however, extended well to Govinda's aphorism (1.3) concerning the miraculous powers of mercury the west (into modern Rajasthan, Maharashtra, Karnataka) and east (Bihar, is shorter than those found in most other works and is unusual in that it Orissa) of their boundaries.151 From the tenth century onwards, there ex- equates bound mercury with liberation (mukti) rather than with the power isted, in the same region (Orissa), a religiously latitudinarian royal family

148 149 Chapter Five Tantric and Siddha Alchemical Literature

known as the Somavamsis, who were patrons of Buddhism as well as of Vaisnavism and Saivism. Finally, two dynasties that ruled in Assam from A number of passages from the RA are nearly identical to passages from the RHT; these may either be paraphrases of the latter on the part of the ca. A.D. 800 to 1100 traced their descent from a king of the "barbarian former, evidence for the oral transmission of the alchemical gnosis, or in- Cinas" and the Kiratas.152 I am therefore inclined to locate Govinda's dications of the existence of a common source from which both borrowed Somavamsi-Haihaya king Madana, lord of the Kiratas, at the eastern ex- material.158 The last appears to be the case, for example, with the descrip- tremity (eastern Madhya Pradesh, Orissa) of the Kalachuri sphere of in- tion of a preparation that stabilizes and strengthens the body tissues prior fluence (which extended into the western Indian regions I have termed the to the ingestion of mercury. The RA (18.14) copies the Astangasamgraba heartland of tantric alchemy). As for Govinda himself, he could have lived (6.50.245) verbatim, whereas the RHT (19.19) appears to paraphrase that anywhere within the boundaries of this Kalachuri sphere of influence. A work.159 tenth- to eleventh-century date for this text is plausible, in the light of the As we have already noted on a number of occasions, the RM borrows dates of later works that refer back to the RHT. widely from the RA.160 That the contrary is not the case is proved time and b. The Rasārnava (Flood of Mercury) is the oldest and most impor- again by the fact that the RM either unnecessarily embellishes passages tant of the alchemical Tantras and may be dated to the eleventh century.153 from the RA or changes the order of passages in ways that do not make the It is an anonymous work in eighteen chapters, cast as a dialogue between Bhairava and Devi. It has been attributed by a number of scholars to a same sense as they do in the RA. A prime example of the former phenome- non is the case of the aphorism concerning the powers of mercury. In no certain Bhairavananda Yogi; "Bhairava," the twenty-ninth Rasa Siddha in the later lists, may be a reference to this text's divine revealer. It is possible other source is the matter stated more succinctly and powerfully than in the RA (1.19): "O Goddess! Swooned, mercury, like the breath, drives away that later texts also refer to its author as Manthanabhairava, for reasons disease; killed, it revives itself; and bound, it affords the power of flight." that will be discussed momentarily. The RA cites no other author or work Compare the derivative, embellished, and thereby weakened aphorism (with the possible exception of the Damara Tantra)154 by name. A compari- of the RM: "When bound, [mercury] affords the state of flight; it removes son with the early eleventh-century A.D. Vimalaprabha commentary to the the heap of diseases when swooned; calcinated it is a destroyer of birth and Kalacakra Tantra reveals a number of striking similarities in both language and subject matter with the material found in the former work, which ap- worldly existence; progressed and colored it affords pleasure and libera- tion."161 Elsewhere, the RM's list of twenty-seven Rasa Siddhas appears to pears to be slightly more archaic than that of the RA.155 Nowhere in the RA do we find a comprehensive accounting of the eigh- be an expansion on a group of twenty-four (Rasa) Siddhas mentioned (but

teen samskäras. In fact, the first eight operations, by which mercury is puri- not named) in the RA.162

fied prior to being applied as a transmuting agent to other minerals and The latter case in point is demonstrated by the treatment the RM gives of the most elaborate alchemical operation found in the entire RA. This metals, are reduced in the RA to two: purification (sodhana) and rubbing (mardana). This is possibly due to the fact that the RA followed a rapid passage, which constitutes the powerful conclusion of the RA (18.208-28), is dramatized in a myth related by Alberuni, in about A.D. 1030, on the Skull-Bearer method (kāpālika-yoga)-as opposed to the slower Lunar subject of the silver threshold of the government house in Dhara (Dhar (saumya) method-for these primary operations.156 The RA is far and away district, Madhya Pradesh). Alberuni's account is reproduced at the begin- the alchemical text that most frequently evokes these kāpālika methods, for the liquification of gemstones, the coloration of mercury and metals, etc.157 ning of chapter three of this book; its close similarity to the procedure described in the RA is a further indication that this is an eleventh-century Like most alchemical works, the RA lists eight primary (7.2) and secondary text (7.56) rasas, but lists only six metals (7.97), as opposed to the nine found In the RA, "diamond-bound" mercury, which has proven its efficacy by in most sources. The RA (7.57-66) is the earliest text to provide an origin restoring cripples, freaks, and mutants to wholeness and transmuting ten myth for sulfur; only in later texts, however, does one find a mythic account million times its mass of base metals into gold, is to be employed in the of the origins of mercury. The abundant tantric elements found in the RA ultimate alchemical Work, the transformation of the entire person of the will be discussed at length in the next chapter. alchemist into an Alchemical Man. This he will do by plunging himself-

150 I51 Chapter Five Tantric and Siddha Alchemical Literature

after an appropriate worship ritual-into a cauldron of superheated oil into tire point of this operation-i.e., the creation of an Alchemical Man- which pellets of diamond-bound mercury have been placed. At this point, were it not for the fact that he refers, in his aside, to the "Great Work the RA explains the principle behind this operation: the five gross ele- called the 'Five Nectars' (pancamrta)." This is precisely the term employed ments-"earth, water, fire, air, and ether"-of his material body will be seri- in the Jayadrathayamala for the tantric "extraction of the essence" of the ally transformed into their subtle correlates (tanmatras), beginning with five gross corporeal elements as the means to generating a Siddha body that of earth and ending with that of ether. The process is at once a con- possessed of the power of flight. Most fascinating here is Nāgārjuna's men- crete application of the principles of (1) Tantric bhūtasuddhi; (2) "extracting tion of the texts entitled Manthanabhairava and "Demon Ravana." While the essence (pancamrta)" of the five elements, as described in the Jayadra- the latter can only refer to Ravana's Kumara Tantra (which is based in part thayamala;163 and (3) the basic Siddha principle of telescoping the gross upon the tantric sorcery of the Kriyakalagunottara Tantra),169 the former into the subtle as a means to generating a siddha body.164 gives rise to thought. Once the alchemist has plunged himself into the cauldron, his labora- There are two candidates for the Manthanabhairava to which Nāgārjuna tory assistant serially adds the alchemical equivalents of the five elements refers. The first, and less likely of the two, is an eponymous alchemical text to the mix, culminating with kha (ether, but also mica), which is to be by this name, which would then account for the name of Manthanabhair- placed inside the alchemist's skull (presumably all that remains of his gross ava as the thirty-second Rasa Siddha and for the numerous references to body at this point): it is noteworthy here that, as with the hathayogic prac- Manthänabhairava (especially to an elixir called manthānabhairava rasa) in tice of kbecarī mudrā, this operation climaxes with ether in the cranial later alchemical works. In fact, all mentions of Manthänabhairava postdate vault.165 Then, "pumping the bellows [until the mix has] the look of molten the RA, by which logic one might posit that Nāgarjuna is referring to the gold, [the assistant] should add an alkaline substance (ksāra). No sooner RA as the (work revealed by) Manthanabhairava. However, the preparation has this been done than he [the Alchemical Man] rises up with a mighty called manthānabbairava rasa, as it is described in the RM itself, as well as roar: 'Hum!'" Then follows a description of the Man's alchemical apotheo- in many later works, is nowhere to be found in the RA.170 sis, in which he sports in the world of the Siddhas with a hundred thousand Moreover, while the RA is obviously Nagarjuna's source here, at no time Siddha maidens.166 does that text refer to this or any other operation as the Great Work called The RM's treatment of the same operation is punctuated by breaks, in- the "Five Nectars." The second, and more likely, source to which Nāgār- consistencies, and asides, which indicate a more or less random appro- juna may be referring is the twelfth-century Manthanabhairava Tantra, one priation of the RA passage. The RM concludes its description of the mir- of the core texts of the Western Transmission, a source to which the RC aculous healing powers of diamond-bound mercury with the statement: seems also to refer.171 (While Nagārjuna appears to be the sole alchemical "Thereupon, one becomes [takes on] another body: there is no need to author to ever have cited this seminal work of the Western Transmission, speculate on this matter." Then, in a very interesting aside not found in the the RA itself contains at least one passage that indicates its author's famil- RA, the RM continues: "This is the Great Work which is called 'Five Nec- iarity with that tradition's cult of the goddess Kubjika.172 We will return to tars' in either the Manthanabbairava, or then again mentioned in the the impact of the Western Transmission on the tantric and Siddha alchem- demon Ravana['s Kumara Tantra].167 All the manifold [life forms] in these ical traditions in the next chapter.) three worlds, including all that moves and all that does not move are [com- The question then arises as to whether the citations, in later alchemical posed of the following]: Earth, water, fire, air, and ether." It then abruptly works, of Manthanabhairava all refer to a Rasa Siddha by that name, or at segues back to diamond-bound mercury, which is capable of transmuting times to a work of the Western Transmission; and, if the former is the ten million times its mass of base metals into gold, at which point it follows case, whether Manthanabhairava the alchemist is to be identified with the the RA description of the operation, with minor variations and one im- Bhairava who revealed or authored the RA.173 portant omission: the RM leaves out the pentultimate operation of placing The RM copies or paraphrases the RA in scores of other places, most kha in the skull of the Alchemical Man.168 especially in its fourth chapter, in the course of which it samples approxi- One might think that Nāgärjuna, author of the RM, had missed the en- mately sixty-four verses from the RA, even repeating itself in the same

152 153 Chapter Five Tantric and Siddha Alchemical Literature

places as those in which the RA is repetitive (which only occurs in the RA's method or expedient of the "fluid" or "power substance" (dravyopāya):18! final, eighteenth chapter). Ironically, it is in this fourth chapter that we "Without the power substance, there can be no realization (siddhi), and no find the "Buddhist" dialogue, set atop Srisailam, between Nāgārjuna, the enjoyment or pleasure. Without the power substance, men become dis- goddess Prajnaparamitā, and a host of other luminaries, that convinced embodied ghosts (pretas) when they die. They who employed inferior sub- Ray and others that this was a Mahayana text. In fact, the passage we have stances [i.e., who followed inferior doctrines] in life roam about blindly in just compared with RA 18.206-228 is cast in the RM is a dialogue between the world of the dead." Nāgārjuna and Ratnaghośa!174 Following this introduction, however, the KCM suddenly becomes very c. The RA is copied, paraphrased, or cited in a significant number of much a Kriyā Tantra, plunging directly into a very long description of a later works, with a particularly long passage (18.278-337) on the subject number of pills (gutikās) which render the user highly attractive to divine of magical waters being borrowed into the Kākacandisvara Kalpa Tantra females (nymphs, yoginīs, Gandharva maidens, and even Bhairavī herself) (KCKT) (8.2-49).175 This work, which deals mainly with tantric healing and eminently capable of satisfying their sexual desires. In this, it closely through the use of botanicals, has little or no relationship to the Kākacan- resembles discussions of magical "attraction" (akarsana) and "the acquiring deśvarīmata (Doctrine of the Fierce Crow[-faced] Goddess) (KCM),176 an of a heavenly nymph" (yaksiņī-sādhana, strīvasyam) found in works on tan- alchemical text in which tantric elements abound. Here too, a number of tric sorcery. The highly evocative names of these gutikas are found in a nearly identical passages in the RA and the KCM would appear to indicate number of later texts; the earlier RA describes one of these, the amara- borrowing from an unknown common source.177 sundarī ("Immortal Beauty") pill in some detail. Of much greater interest As its title indicates, the KCM is Sakta in its sectarian orientation. More to the RA, however, is a pill that affords magical flight (kbecarī gutikā); and exactly, Kākacandeśvarī's corvocephaly is of a piece with such Kaula cults the KCM follows its instructions for the preparation of this and other of Kālī as the Kaula Mata and the cult of Guhyakāli, which has been popu- aphrodisiacs with its own discussion of the same.182 From here, the KCM lar in Nepal since at least the eighth century. In both of these cults, zooce- abruptly changes its orientation once again and launches into a long dis- phalic or avicephalic goddesses have figured prominently.178 The Śāktism cussion of gold making (dhatuvada) through the use of mercury that has of the KCM is explicit from the outset: while it is the god Bhairava who been treated with mica, herbs, and sexual fluids before returning, inexor- is doing the revealing, he is nonetheless ringed, atop Mount Kailash, by ably, to its discussion of aphrodisiacs.183 Next, it lists the best nights of a circle of "yoginīs," named Kāpālī, Kālarātrī, Kākacaņdā, Ambikā, Karālī, the year for meeting divine maidens (devakanyā), discusses the miraculous Kālakarņī, and Kākacaņdeśvarī, who are accompanied by Gaņanāyaka panacea called "rock water" (sailodaka), and describes the use of orpiment (Ganeśa). In the course of this work, Bhairava refers to Kākacandeśvarī (haritala) and other mineral and plant substances in the preparation of by a number of names, including Kākacamundā, Bhairavī, and Cāņdālī. elixirs.184 Bhairava himself is called Śiva, Sadāśiva, Īśvara, and Sarvajña; and a refer- To this point in the text, it would be extremely difficult for an alchemist ence is made in the text to the worship of Batukanatha (the "Boy Master") to carry out any of the operations described in the KCM for the simple Bhairava.179 Like the RA, the KCM contains an abundance of references to reason that none of the basic principles of alchemy have been systemati- worship, mantras, and sexual practices that place it squarely within the tan- cally explained: terminology, materiae primae, apparatus, procedures are all tric tradition. Which tantric tradition is another question, however; the taken for granted. Now, over halfway through the work, the author seems KCM appears to be as much inspired by the old Kriya Tantras as it is by to have second thoughts and undertakes a systematic exposition of the the alchemical tradition. eighteen samskäras and a number of other fundamentals of his discipline. The opening verse of the KCM resembles those of the Kulananda Tantra The samskära list is unusual inasmuch as neither transmutation nor bodily of Matsyendra and the Buddhist Lalitavistara.180 Following this and a very transubstantiation is enumerated; in their place, it names two operations powerful opening statement (second only to that found in chapter one of not counted as samskāras in other works: vidhāna ("apportioning") and the RA) on the inability of the various philosophies and religions to alle- mārana ("killing"). Following this, it singles out eight of the eighteen oper- viate human suffering, the KCM comes directly to its raison d'être, the ations as the "Eight Works." Its enumeration of these resembles that found

154 155

Chapter Five Tantric and Siddha Alchemical Literature

in RA 10.10-12.185 In its expanded discussion of these eight procedures, thirteenth-century works on tantric sorcery: the Dattātreya Tantra (1.5), the KCM describes the "class of enemies" (arivarga), i.e., minerals that at- the RRA (5.1.7), and the KPT (1.9, which merely copies the RRĀ), all of tack and "kill" each of the eight metals. Similar classifications are found which refer to it as the Kalacandesvaramata. At the same time, an apparent in the later RC (4.50; 14.14 [ariloha]) and ÄK (7.28).186 Next follows a dis- reference it makes to a fourteenth- to fifteenth-century work entitled the cussion of the purification of the secondary rasas, of which six are listed: Rasasāra and the presence of quite modern vernacular terms in its conclu- there is no systematic description of the primary rasas.187 At this point, sion make it clear that this is a text that was reworked over a period of a commentary is inserted into the text of the KCM,188 which continues several centuries, prior to taking its definitive form.194 intermittently until the conclusion of the work: a similar tippana is inserted d. Like the RHT the Rasopanisat (RU)195 is a Mahesvara Siddha work, at a similar point in the RM. concentrating more on transmutational and therapeutic alchemy than on Devotional praises of mercury and sulfur (or mica) in combination fol- the goal of jivanmukti through the use of alchemical elixirs. It is unique in low, a commonplace that most alchemical tantras place in their opening comparing mercury to the ätman, the individual soul (15.50), and in its chapter.189 A brief description of the four colors of mica is innovative in- praise of liberation (mukti) as the highest goal (1.5). Cast in the form of a asmuch as it classifies the four colors (varnas) of mica according to "caste" dialogue between Mahädeva and Pārvatī, it is datable to the twelfth to thir- (varna): white mica is brahman, red is ksatriya, yellow is vaiśya, and black teenth century. It is a southwest Indian work, composed in Kerala.1% It is sūdra. A number of other works classify mercury in this way; the KCM is purports to be an abridgment of a work entitled Mabodadhi;197 both it and the sole early text to do so for mica.190 the Mahodadhi are named in the BPb (1.114), also a twelfth- to thirteenth- Hereafter, the KCM once more becomes a work on tantric sorcery and century work. A number of other textual sources are cited in the same healing. A section on "skewing Time" (kāla-vañcana) describes a number chapter (1.14): these are found in no other work.198 It describes a prepara- of means by which to predict the future and revive the dead or dying. The tion discovered by Srīman Nāgārjuna-muni (16.10): chapter colophons to use of a mixture of male and female sexual fluids (kundagolaka) constitutes the thirteenth- to fourteenth-century RM call that text the work of Śriman a portion of a discussion on a rite through which the yogin may gain super- Nāgārjuna. Elsewhere, the RU paraphrases a passage from the RA, in natural powers and immortality. (The MBhT also gives a prominent place which mercury is compared to a wild elephant in rut, which can only be to the kundagolaka and other such cocktails; the RA and RM also use the brought under control by the "elephant goad of mercury" (rasānkuśa). In term, but desexualize it into a combination of the herbal essences of the the case of the RA, the elephant goad is a mantra identified with the al- cāndālī and rāksasī plants).191 The KCM states that this rite renders the prac- chemical goddess Rasānkuśī; in the RU, it is simply stated that mercury titioner a kavi;192 we will return to Siddha uses of this term in chapter ten. can be stabilized only through decoction (ghanät). The "elephant goad" is The work concludes with a long section on tantric healing through the a commonplace tantric mantra, whose use makes sense in the context of use of herbs and mantras. One such mantra contains a long evocation of the RA; borrowed as it is into the nontantric RU, the metaphor loses most rsis (Kaśyapa, Kāpațu), gods (Sūrya), kings (Sugrīvarāja, the monkey ally of its impact.199 Rather than the usual eight, the RU (4.4-6) names only of Räma), bhūtas, and all manner of other creatures (water-goers, sky-goers seven primary and secondary minerals and seven metals. Its presentation . . . piśācas, dākinīs, etc.).193 In every manuscript version of the work I have of the alchemical samskaras is also highly idiosyncratic, with only six of consulted, these final folios are written in a curious mixture of Sanskrit and its samskäras corresponding to the standard terminology employed for the a vernacular in which many Hindustani words are recognizable (cāval, ser, alchemical operations. Another six operations, which appear to correspond subāg, ātāi, lasanu [= lesūn?], thambbana boi, etc.). Given the heterogeneous to more familiar samskāras, are presented under the heading of kūrpa, a nature of this work, it is extremely difficult to date. While the concluding term found nowhere else in the alchemical canon.200 The RU makes no portion (which is likely a commentary by a later hand) is certainly quite mention of the final samskära of transubstantiation. All of these idiosyncra- recent, the KCM's treatment of much of the alchemical and other material cies arise from the fact that this is a work that is relatively early and rather appears to be quite archaic. In any case, some version of the KCM was peripheral, both in geographic and sectarian terms, to tantric alchemy. already extant in the twelfth century, as it is cited in three twelfth- to e. As already stated a number of times, the most problematic element

156 157 Chapter Five Tantric and Siddha Alchemical Literature

in the Bhūtiprakarana (BhP) (Production of Supernatural Powers)201 is its so-called Goraksa Sambitäs is, as has so often been the case in the annals of title. To this point I have generally referred to this work as "the so-called tantrism, a case of ex post facto attribution or "postscriptural systemati- alchemical Goraksa Samhita"; this I have done because the sole edition of zation." On the other hand, an attribution of this work to a Rasa Siddha this work was published under that title, by Janardana Pandeya, in 1977. named Svacchandabhairava ought not to be rejected out of hand: in its In fact, the chapter colophons to Pandeya's edition read: "here ends ... opening chapter, the BhP intimates that it is an abridgment of a much the Siva Sūtra, contained in the Bhūti-prakarana of the Goraksa Sambitā in longer Rasa Tantra revealed by Svacchanda.206 100,000 verses, brought down [to earth] by Svacchanda-Sakti." This is in This is a work in nine chapters, presented in the form of a dialogue fact an abridgment of the colophon to what I have termed the "Sakta Gor- between Śrīkantha and Umā. The latter's name is significant in this work, aksa Sambita," a work that is nearly identical to the Western Transmission given that it refers to sulfur as umāyoni ("Umā's womb"). Also original, in work known as the Srimatottara Tantra. The colophon to this work reads: certain manuscript versions, is the use of the term śarvaja ("born from [Ru- "here ends ... that section of the Kula-kaulinī-mata (in 90 million verses, dra,] the Arrow").207 The mercurial Siva is also referred to as Paradeśvara brought down [to earth] at the yoginīs' secret Vidyāpītha by Śrīkantha) (5.27) and Raseśvara (5.289). The three Trika goddesses, Parā, Aparã, and known as the Kadi section, which is included in the latter portion of the Parāpară, are also mentioned (1.106), as are the divine Mothers (mātrkās) 100,000 verse Goraksa Sambita which was brought to earth by Svacchanda- (6.281). The alchemist is to worship Bhairava and the circle of yoginīs, as śakti, who had received its seventy million-million [verses] from the most well as Ganesa, the regents of the cardinal directions, and brahmin maidens excellent Maha-manthana [-bhairava]."202 Here, we are in the presence of (3.7-8). Ritual sex is also prescribed.208 textual attribution carried to an exponential level. The BhP bears little or The BhP makes vague references to a number of tantric works and cor- no connection to the Sakta work of the same name, save for the fact that pora. It mentions the three srotas of the tantric corpus, as well as (a work the two are revealed by the same form of Siva (Srīkantha). entitled?) the Uttarottara Mätrtantra.209 In addition to the Svacchanda for- It is, like nearly every alchemical work, a Saiva treatise (Siva Sūtra); mulae, it describes a number of other mantras, including one that appears however, this mention is missing from chapter colophons of most of the to refer to Matsyendra.210 It mentions the alchemist Nandi (9.134) and, manuscript versions that I have consulted. As for the claim that this work in at least one manuscript, also makes mention of the [Rasa-] Ratnākara is a portion of the "Goraksa Sambita in 100,000 verses," a single verse (RRĀ) (however, the RRA [4.4.100] names Goraksanāth and his disciple (9.136), found at the end of Pandeya's manuscript, states that "the Nath Gahaninäth).211 It also cites the Rasopanisat and the Mahodadhi, both of made this [work] manifest for the benefit of king Simha," and quotes "the which it attributes to the goddess Umā, and goes on to give titles of nonex- immortal Näth [named] Goraksa," and but does not make it clear that Gor- tant alchemical works authored by a number of divine alchemical Siddhas aksa is the author of the work. These verses, moreover, are not found in (of which two-Sukra and Brhaspati-are also mentioned in the RU).212 every manuscript of this work.203 A rare alchemical reference to Gorakh is It is cited, under the title of Svacchandasaktyavatāra, in the fourteenth- found in the ca. sixteenth-century Rasakāmadhenu (2.362-63), which cites century Rasarājalaksmī of Visnudeva.213 These data would place the BhP in a work entitled the Goraksamata (The Doctrine of Goraksa) in its descrip- the twelfth to thirteenth century, which squares with an apparent reference tion of the alchemical uses of a flower called hamsapada: this passage, how- to it by Somadeva, when he cites Srikantha in his thirteenth-century RC ever, is nowhere to be found in the BhP.204 (7.1). Its sole geographical references are to the Vindhya range (6.60), to a The Svacchanda Sakti who is said to have brought the Goraksa Sambitā site "to the southeast of the Himalayas" (2.56), and to a (probably mythic) down to earth is referred to but once in the work itself; however, a number location called "Virgin Island" (kumarīdvīpa), where divine herbs abound of references to male Svacchanda deities and Svacchanda mantras indicate (7.113). that this work belongs to the Svacchandabhairava canon (the cult of Svac- In nearly every respect, the BhP is a garden-variety alchemical tantra. chandabhairava was especially important in Kashmir, and the SvT was the Its oft-repeated goals are siddhis, Sivahood, and sporting in the heavens principal work of the so-called Mantra Pitha).205 The title of this work, with divine maidens;21+ and its means to that end are a blend of alchemy, then, is Bhutiprakarana, and any connection between it and the two other tantric worship, and devotion to mercurial Siva lingas and the Siddhas.215

158 159 Chapter Five Tantric and Siddha Alchemical Literature

Significant in this regard is its quite frequent reference to mercury as dra- drinking liquor" (1.8-10) to its paraphrase of that work's aphorism on the vya, the fluid or power substance, without which the use of mantras and powers of swooned, bound, and killed mercury (1.28) and its discussion other forms of worship are vain. In this, the BhP resembles the KCM, of the devotional cult of mercury (1.30-51). Chapter two's discussion of which it also follows in its numerous preparations in which menstrual initiation, including instructions for the establishment of images of Rasa- blood is used.216 It also resembles the latter work in its frequent use of bhairava and Rasankusī in the midst of an alchemical mandala (2.4-50) vernacular terms, of which one, khānepāne (4.77), used as a synonym for closely follows RA 2.4-8, 44-76. No fewer than forty verses in chapters siddbi, is also found in the RM, Matsyendra's Akulavīra Tantra, and the Dat- ten through fourteen are borrowed from the seventh chapter of the RA.222 tātreya Tantra.217 Its aphorism on the powers of swooned, killed, and bound Somadeva is, nonetheless, original on a number of counts. His is the mercury (1.51) is a paraphrase of that found in the RA. earliest work to contain the origin myth of mercury, as well as the descrip- The BhP devotes much of its seventh chapter to alchemical uses of the tion of a fantastic procedure for the extraction of mercury from its sub- sixty-four divine herbs (divyausadhi), which it identifies, after an alchemical terranean "wells" (15.4-22). He is the first alchemical author to identify commonplace, with the sixty-four yoginīs. These are said to grow in those the nine gemstones (12.1) with the nine heavenly bodies (grahas), and the places where Siva and the Goddess had sexual intercourse and are to be first to describe five types of alchemical transmutation, which correspond culled and used at certain specific times of the lunar month and the solar to the five elements (4.107-11). Curiously, he devotes very little attention year.218 Its description of the construction and layout of the alchemical lab- to the ninth through sixteenth samskaras, preferring to concentrate on the oratory and the qualities of laboratory assistants generally follow those ultimate operations of transmutation (vedha) and transubstantiation, which found in the RA.219 The eight primary and secondary minerals (2.19, 28) he terms sevana (15.28). Apart from the anonymous author of the RA, and eight metals (2.31), as well as the eighteen alchemical samskāras (3.4- Somadeva is the most stylish of the alchemical authors. His work bears 6), listed in the BhP correspond to those found in most alchemical sources. the stamp of a powerful personality. f. The Rasendracūdāmani (Crest-jewel of the Lord of Rasas) (RC) of So- g. The Rasaprakāśa Sudbākara (Effulgence of Mercury, Wellspring of madeva, a work in sixteen chapters, cites a great number of tenth- to Nectar) (RPS) of Yasodhara Bhatta is a work in thirteen chapters devoted thirteenth-century alchemical authors: Manthānabhairava, Nandi, Nāgār- to medicinal uses of mercurials and other alchemical preparations. Yaso- juna, Bhairavācārya, Govinda, Bhāskara, Bhāluki, Srīkantha, etc.220 The RC dhara's name figures in all the Rasa Siddha lists,223 and he mentions the is mentioned in the thirteenth-century RRA (1.10.51-61) and RPS (7.37; twelfth- to thirteenth-century Somadeva (9.12) and opium: therefore his is 9.11), and copied extensively by the thirteenth- to fourteenth-century a thirteenth-century work.224 Yasodhara states that he is a resident of the RRS: it therefore dates from the twelfth or thirteenth century. Somadeva Saurashtran town of Jirnabhidha, which likely corresponds to the modern calls himself the lord of Karavala Bhairavapura, a name that corresponds toponym of Junagadh.225 In the opening verses of his work (1.1-4), he to no known Indian toponym, either medieval or modern; however, Soma- evokes a series of deities-Sāradā, i.e., Sarasvatī; Gaņapati, i.e., Gaņeśa; deva's mention of both Nāgārjuna and Bhāskara, as well as a familiarity and Harihara. This last name is of course a reference to the conjoined with Mount Abu "in the Gujarat region" (11.111) and the local Saurashtran Visnu-Siva, a god whose name is also evoked in the opening verses of the term for a specific type of clay (11.49) indicate that he was from Gujarat.221 RHT (1.2). Given that Yasodhara identifies his father as a Vaisnava named Somadeva states, on a number of occasions, that he is one of the greatest Padmanābha ("Lotus-Navel," i.e., Visnu) of the Gaudīya (Śrīgauda) brah- alchemists on the face of the earth, and that certain techniques are known min lineage,226 we can be nearly certain that the author of the RPS was a to none save Siva, Nandikeśvara, and himself. In spite of his self-assurance, Vaisnava, rather than a Saiva or Sakta alchemist. He refers to himself as a Somadeva, like many before and after him, copies or paraphrases large por- kavi and a kavivara (13.14-15).227 tions of earlier works, especially the RA. Chapter one closely follows the In spite of Yasodhara's probable Vaisnavism, his account of the divine first chapter of the RA in structure and content, from its evocation of the origin of mercury (1.5-12) is based on the standard Saiva myth of the birth Kaulas (1.6-7, 11-15) and its discussion of "eating the flesh of the cow and of Skanda. Enumerations of the eighteen alchemical samskāras (1.30-164)

160 16 Chapter Five Tantric and Siddha Alchemical Literature

and the eight principal (5.2-3) and secondary (6.1) rasas are unexceptional. Chapter eight includes descriptions of the preparations called svacchanda- [and] the aged in the [fourth] Rasāyana [division]. Supernatural power of

bhairava rasa (8.132-33) and manthānabbairava rasa (8.173-74). In chapter an alchemical order (rasa-siddhi) is proffered for the sake of [tantric] advi-

nine, Yasodhara's enumeration of the sixty-four divine herbs (divyausadbīs) sors in the [fifth] Mantra division."233

is based, by his own admission (9.11), on Somadeva's RC. An unusually The Rasa and Vada divisions of the RRA, composed of ten and twenty

long list of forty yantras is presented in chapter ten; chapter eleven is chapters or "teachings" (upadesas) respectively, treat for the most part of

devoted to gold making. Chapters twelve and thirteen are devoted to the operations proper to transmutational (dhatuvāda) alchemy, while the

Ayurvedic techniques of vājīkarana. In addition to Somadeva, Yaśodhara Rasendra division, seventy-one chapters in length,234 treats of therapeutic

mentions Bhairava (7.26), Nāgārjuna (5.107), and Nandi (6.73). (rogavāda) alchemy. The Rasāyana division, eight chapters in length, is a

h. Because of the corrupt nature of the manuscript he consulted, P. C. discussion of elixir (dehavāda) alchemy, save its concluding chapter, which

Ray mistook the RM for the Rasaratnakara (Mine of Gemstones and Rasas) is a panegyric of the alchemical wonderland of Śrī Parvata (Srisailam) and

(RRÅ), which he attributed to Nāgārjuna. As already noted, the thirteenth- the alchemical practices specific to that site.

century RRA was authored by Nityanatha Siddha, who says as much Much of the final Siddha or Mantra division of the RRA appears to have

(1.1.25) and who names Nagarjuna in an opening verse of his work been borrowed, more or less piecemeal, into the KPT of Siddha Nāgārjuna.

(1.1.17a). In the same passage, he also mentions the Rasārnava of "Sam- This is made manifestly clear in KPT 1.13a, which states: "This is called

bhu."228 It is also in the RRA (3.1.66-70) that we find an alternate (to the the Mantra division (mantra-khanda) [which is offered] for the welfare of

RM/RRS [a]) list of Rasa Siddhas, a list that notably includes Carpati practitioners." The same half verse is found in the Mantra Khanda of the

(whose name is absent from the RM/RRS [a] list); the RRA (1.1.17b) also RRÅ (5.I.IIa), where the statement makes perfect sense. The twenty-seven

cites a work by Carpati, entitled Svargavaidya-kāpālika.229 The RRĀ is a chapters of the Mantra division become condensed into twenty chapters

later work than the KCM as well as the RC, both of which it cites;230 but is in the KPT; however, the chapter order of the two is quite identical. Even

earlier than the RAK, which cites it,231 and the RRS and AK, which borrow if it is divided into fewer chapters, the KPT is substantially longer than the

from it extensively. Mantra Khanda of the RRA, again indicating that it is a later expansion

I have already indicated the ubiquity of Nityanātha Siddha, "son of Pār- Nityanätha's work.235 Given that the RM cites the KPT, we must conclude

vatī," as the tantric author of works on alchemy, hatha yoga, and tantric that it too is a later work than the RRA: therefore, the Rasa Mangala named

sorcery, and as a name figuring in both Rasa Siddha (RRS) and yogic Sid- in RRÅ 1.1.16b cannot be the same work as the RM.

dha (HYP) lists. I have also surmised that Nityanatha Siddha may have Nityanātha Siddha makes a number of statements that are more or less

been a Näth Siddha polymath, after the fashion of the founder Gor- unique to this alchemical source. He mentions both the necessity of be-

akhnāth, whom he names in RRA 4.4.100. (He also names Gahananātha,232 longing to an order (sampradaya), and states that his work is a blend of

who, if he is the same figure as the Gahaninatha who was Gorakh's dis- his guru's teachings and his own practical experience (1.1.2ra, 22a). Else-

ciple-and Jñāneśvara's guru's guru-in Maharashtran Nāth Siddha lin- where, he states that the alchemical teachings were received directly, by the Siddhas, from Siva's mouth; this mystical technique has been kept hidden eages, offers further proof that this is a late thirteenth-century Näth Sid- dha text, likely from Maharashtra). The contents of the RRA, while they in the writings of their order, which he has himself surveyed (3.1.9-10).

betray the author's interest and expertise in the fields of tantric alchemy Then follows an oft-repeated aphorism, of which Nityanatha appears to be the original source: "Neither sequence (oral teachings) without written and sorcery, as well as Ayurvedic therapy, make no reference to hathayogic techniques. Nityanäth sets forth the agenda to his very prolix work in his sources nor written sources without sequence [are acceptable]. Knowing

opening verses (RRÃ 1.1.4b-6a): "This treatise in five sections (khandas) the written sources to be conjoined with sequence (oral teachings), the per-

is offered for the benefit of practitioners; [for the benefit of] Ayurvedic son who then practices [alchemy] partakes of the siddbis."236

physicians (vaidyas) in the [first] Rasa division; [that of] the afflicted in the Nityanātha's discussion of Srisailam (4.8.1-185) is by far his most inno-

[second] Rasendra [division]; theoreticians in the [third] Vāda division; vative contribution to the alchemical literature: that this is the Saiva center in western Andhra Pradesh is clearly proven by references to the jyotirlinga

162 163 Chapter Five Tantric and Siddba Alchemical Literature

of Mallikārjuna (4.8.2, 5, 129, 135, 173) that is the heart of the cultus there. His catalog of alchemical wonders at a number of sites (a grove called Ka- text (4.28) mentions only Himalayan and eastern Indian toponyms.239 Its

dalī Vana, pools, wells, etc.) is by far the most comprehensive of its genre. Śaiva Tantra. title and strong emphasis on the cult of the Goddess make this a Sakta-

Elsewhere (1.1.12a-13a), he expands on an aphorism first introduced in RA 10.29 and compares various types of mercury with the gods Brahmā, Visnu The MBhT, a work in fourteen chapters, is cast as a dialogue between

(Janārdana), Rudra, Maheśvara, and Sadāśiva. These last two names for Śankara and the goddess Candikā. Of its twelve chapters, three (chapters

Siva would appear to indicate that Nityanatha's pantheon was that of or- five, eight, and nine) are devoted, in the main, to alchemy. Here, however,

thodox Saiva Siddhanta; yet his alchemy, concerned with jīvanmukti as an the use of mercurials is not presented as an end in itself, but rather as a

end in itself rather than a means to the orthodox end of moksa, betrays a means to the higher end of ritual devotion to Siva and the Goddess. At the

more heterodox orientation. Elsewhere, he also mentions Manthānabhair- heart of this devotion is the worship of Siva in the form of a mercurial

ava (4.3.197). Explicit references to the perfection of the human body as linga, the mercury of which has been stabilized and calcinated with men-

the alchemist's highest goal (4.1.2), to sexual relations with menstruating strual blood. No alchemical Tantra is as attentive to the miraculous powers

women as a suggested adjunct to work in the laboratory (4.1.4a; 4.3.197- of uterine or menstrual blood as the MBbT. It classifies this fluid (dravya)

199), and to cremation ground rites (4.4.113) reinforce this emphasis, as into six types-according to the age, marital status, and sexual experience

does the entirety of the Mantra or Siddha Khanda, which is wholly devoted of the woman in question-and invokes its use in a number of practices,

to the subject of tantric sorcery. both ritual and alchemical.240

Nityanatha appears to follow the lead of the RA in his discussions of The use the MBbT makes of the term dravya-to refer to both sexual

alchemical worship and initiation, the ideal guru, disciple, and female labo- emissions and mercury as "power substances"-parallels usages we have

ratory assistant, etc .: he is in turn copied by Vagbhatta II, author of the already noted in the RA, KCM, and BhP. Chapter four of the MBbT is de-

RRS. The RRA's rendering of the classical aphorism concerning the pow- voted to the worship of Camundā Kālikā, whose name evokes that of Kāka-

ers of swooned, killed, and bound mercury (1.1.24) may be the source from camundā, one of the names by which Kākacandeśvarī is identified: there is,

which the RM borrows most directly; the RM also appears to follow Nitya- moreover, a certain symmetry between the Candika of the MBbT and the

nätha who, in the final verses of this chapter, discusses the dosas of mercury Kāka-cand-eśvarī of the KCM. The MBbT is thoroughly conversant with

and suggested quantities of mercury to be employed in the alchemical op- the subtle body of hatha yoga and includes extended discussions of the

erations (1.1.25b-33a). cakras, the "nine doors" (9.20), and the kundalini. Here, the MBbT follows

The RRÅ lists no fewer than twelve metals (1.8.1-2), among which it mainstream tantric discourse when it states that the alcohol, flesh, fish, etc.

includes three alloys and iron rust. On the basis of Nityanātha's name, evo- ritually consumed by the practitioner are offerings to the Goddess who,

cations of Gorakhnath and Gahaninath, his emphasis on the practitioner's dwelling within the subtle body in the form of the kundalini, rises up to

adherence to a sampradaya, his inclusion in the HYP, and the other yogic the practitioner's tongue, in order to enjoy them. The mūlādhāra cakra,

and tantric works attributed to him, I am inclined to think that Nityanath in which the kundalini dwells, is identified with the Goddess's pitha of

was a Nath Siddha who lived slightly later than Gorakhnath but earlier Kāmarūpa.241

than the BhP that is attributed to the latter.237 Worship of Siva in the form of a linga is central to this work, with the

i. The Mätrkabheda Tantra (MBhT) is, on at least two counts, a singu- mercurial linga being the optimum type. Neither worship, nor ritual, nor

lar work in the annals of tantric alchemy. On the one hand it is, in the ful- alchemy is effective without the use of mantras, to which the MBbT de-

lest sense of the term, a Siddha alchemical work, combining the standard votes much attention.242 The work also discusses the importance of the

theoretical, yogic, ritual, and behavioral components of practice with al- guru, the establishment of lakes, reservoirs, and temples, and various rules

chemy.238 On the other, it is the sole Hindu work containing significant of worship-the standard subject matter of a Tantra. The practitioner who

alchemical data to be a clearly eastern Indian product. Most manuscripts follows its instructions gains siddbis and becomes a "Siddha who is the

of the MBbT are found in Bengal, and a list of sacred waters found in the manifest equal of Siva."243 The MBbT mentions the eighteen Puranas, and cites the Kāli Tantra, Nityā[șodašika] Tantra (= Vāmakeśvara Tantra, which

166 167 Chapter Five Tantric and Siddha Alchemical Literature

the tenth century. On the other hand, "Nāgarjuna" may have been-as was some space to "skewing Time" (kāla-vancana).263 It makes mention of the the case with Matsyendra-a title applied to alchemists who had reached tantric worship of a virgin and of the dravya "power-substance" to be em- a certain level of expertise, in which case we should not be surprised to be ployed therein.264 confronted by a host (gana, the title of chapter eight of the RM) of Nāgār- Following chapter four in the Paris-Bikaner manuscripts of this work junas! In any case, the premise for all of these speculations-i.e., that the are several folios which neither correspond in their contents to any of the Nagarjuna in the text of the RM was a historical figure-may itself be final four chapters of the work nor have a name given to them in the final groundless. colophon of these manuscripts. Contained in this "appendix" are an auto- Following its table of contents, the RM launches into a praise of mer- commentary (tippana),265 a verse borrowed from the RHT (1.2), and the list cury, stating that "through the use of the mercurial method (rasendra- of the twenty-seven Rasa Siddhas which is borrowed directly into the RRS. yogat), one becomes a superior yogin on this earth, surpassing even the Lord As noted above, this auto-commentary is similar to that found in the KCM [Siva] (nātha-atitulya)": Śiva is called "Lord" (nātha) in a number of pas- and is basically a companion to the work as a whole, giving the five names sages in this work.252 Hereafter, the work lists the impurities present in for mercury (including śarvaja,266 a term also found in the BhP, but no- mercury and gives an incomplete list of the alchemical operations, which where else), types of female laboratory assistants, etc. Until the critical edi- it calls the "eighteen Works" (karmānyastādaśa), and gives the suggested tion of this work appears, the authenticity of this final section is open to quantities of mercury to be used in these operations.253 doubt. Chapter two begins with a description of the "Kāpalika method" for the k. The Rasaratnasamucchaya (Amassing of Rasas and Gemstones) (RRS) liquification of gemstones, a technique this work borrows directly from the of Vägbhatta II is a thirteenth- to fourteenth-century work in thirty chap- RA.254 The same chapter concludes with another seminal teaching, cribbed ters, of which fourteen (1-11, 28-30) are alchemical or iatrochemical and from the RA (7.151): "killed metals become adequate for internal use."255 sixteen (12-27) are Ayurvedic. In spite of its great prestige this is, at least This is precisely the subject of the third chapter, i.e., therapeutic uses of in its alchemical content, a highly derivative text. Its first chapter is, for herbal and mercurial preparations and elixirs, most of which bear the the most part, a patchwork of passages cribbed from the RA, RM, and espe- names of Saiva or Sākta deities.256 Chapter three concludes with a long cially the RHT.267 Of its remaining alchemical chapters, much of their con- section on eye medicines (añjana), for which, it will be recalled, a ca. fifth- tent is a transparent adaptation of the RC of Somadeva and the RRA of century Nāgārjuna was renowned.257 Nityanätha.268 Its mention of the Yädava king Singhana (d. A.D. 1220) and It is in the very long chapter four of the RM that we find nearly all of of opium support a mid-thirteenth-century date; if, as some commentators the "Buddhist legend" and the borrowings from the RA discussed above.258 maintain, it borrows eleven verses (5.132-142) from the ca. A.D. 1375 This chapter, on the subject of mercurial gutikas, gives a description of Rasaratnapradīpa of Rāmarāja, then it is a late fourteenth-century work.269 the same "Immortal Beauty" pill referred to in the RA and KCM; the RM 1. The Anandakanda (AK) (The Root of Bliss), the most encyclopedic preparation differs slightly from that of the two other texts.259 It is also in work of the entire Hindu alchemical canon, is a massive text, over six thou- this chapter that we find a list of the eight primary rasas, although lists of sand verses in length. In its colophons, it calls itself the work of Bhairava; the secondary rasas and metals are lacking.260 A long section in this chapter, the apparent referent here is the tantric god, given that the entirety of the devoted to the use of "leech-bound" mercury (i.e., treated mercury whose text is cast as a dialogue between Bhairava and Bhairavi. The AK is divided consistency is that of a leech), placed in the vagina as a female aphrodisiac, into two sections, curiously called "recreations" or "places of repose" (vi- appears to be an expansion on a similar discussion found in the KCM.261 śrantis), which are subdivided into a total of thirty-six chapters and an The latter portion of this chapter has already been discussed at some appendix. The second section, entitled "work" (kriya) is essentially devoted length. Following the RA, it stipulates that mercury that has been cal- to the chemistry of metals and gemstones. The first section, entitled "nec- cinated six times in sulfur is optimal for transmutation and transubstantia- tar making" (amrtikarana), may be termed a Siddha alchemical text, con- tion.262 Like the KCM, the KJiN, and most other Siddha works, it devotes taining as it does chapters on both alchemical and hathayogic (chapters 20

168 169

Chapter Five Tantric and Siddba Alchemical Literature

and 21) techniques for the realization of bodily immortality, a chapter (12) rowed directly from the RC (6.1-72). Its discussions of the use of mantra, on the alchemical wonders of Srisailam, discussions of initiation, etc. As yantra, nyāsa, etc. are likewise more prolix than those found in other al- already noted, it is in this work that the sole attempt is made to fuse the chemical sources. Its chapters on hatha yoga make no mention of Gorakh- Nāth Siddhas together with the Rasa Siddhas: this is, from a history of näth or any other Siddha; the content of these chapters is, however, clearly religions standpoint, the AK's greatest innovation. inspired by the Goraksa Sataka and the other works of "six-limbed" yoga In spite of its dual emphasis on alchemy and hatha yoga, the AK is, prop- that it spawned.272 Its description of the meditation hut (kuti) is similar to erly speaking, less a work on Siddha alchemy than an encyclopedic work, that of Ayurvedic rejuvenation therapy, inasmuch as it is a threefold struc- which incorporates and synthesizes (i.e., copies and paraphrases) all that ture (1.21.4-5). As in many other alchemical sources, the AK lists its Rasa falls within its fourteenth-century purview. So it is that the AK's anony- (and Nath) Siddhas in a worship context: they are the recipients of water mous author also incorporates extended discussions of Ayurvedic elixir offerings in the "conventional initiation" (samaya dīksa) section of the therapy, seasonal dietary regimens, and a number of other ancillary topics work.273 Graphic representations of the nine Naths and sixteen Siddhas are

into his work. also to be figured, together with the eight Puranic (and not Tantric) Bhaira- It has generally been maintained, mainly on the strength of the fact that vas, the sixty-four yoginīs, and other deities, on the outer walls of the medi- its three original manuscript sources are housed in south Indian collec- tation hut (1.21.71-74). tions, that the AK is a south Indian work.270 This is further supported by m. The small gem entitled Khecarī Vidya (KhV) (The Aviator's Science; the inclusion, in the AK's Siddha lists, of names whose origins are clearly or The Arcane Science of Flight) is, in spite of its relatively late date, a south Indian (Mūlanātha, Revaņa, Konkaneśvara, etc.). If this is a south paradigmatic text of the Siddha alchemical tradition. Unlike the AK- Indian work, however, it likely emerged from the Deccan plateau rather whose encyclopedic content encompasses alchemy, batha yoga, and nearly than either of the southern coastal regions: the seasonal cycles it describes everything else under the medieval sun-the KhV never strays from its in chapter nineteen correspond to a temperate and continental (versus a single object. The object in question, the siddhi of human flight, is treated tropical and coastal) climate. It is also possible that this chapter, like so in an extensive and generally derivative way: the KhV draws, for the most much of the rest of this work, is a wholesale borrowing from external (and part, on preexisting hathayogic traditions. What is original about the KhV

northern) sources. is its triadic focus, on hathayogic, erotico-mystical, and alchemical tech- The AK is most ingenious in its ability to inflate the content of passages niques for the realization of its end, a focus I have previously qualified as it borrows from other alchemical works, most especially the RA, RRA, RC, the "Siddha distinction." and RRS. Of these, the RRA is the AK's prime source. So, for example, the The KhV, a work in four chapters and 285 verses attributed to Ādinātha, AK chapter (1.12.1-200) on Srisailam is borrowed nearly piecemeal from generally calls itself, in its manuscript colophons, a portion of a work en- RRA 4.8.1-185.271 It contains the standard origin myth of sulfur (1.13.3-24) titled the Mabākālayogasāstra. Two late medieval commentators, Brah- as well as of mercury (1.1.8-23: an expansion on that found in RRS 1.60- mananda and Narayana, refer to a work of the same title as a treatise on 66); its mythic description of the extraction of mercury (1.1.53-62) from batha yoga: they are presumably referring as well to the KhV.274 This not- its well in "Cinnabar-Land" also expands on the RRS (1.85-88) account of withstanding, every manuscript of this work that I have found calls itself the KhV275 The sole works that the KhV itself cites are called "Mahākala" the same. Instructions for the worship of Rasabhairava, Rasānkuśī, the ten Dūtīs, the eight Vidyeśvaras, the yoginīs, and Kubjikā (1.2.165) rely heavily and "Martandam Viveka." The latter of these is the Yogamārtanda or Vivek- on the rasamandala of the RA (2.53-72). amärtanda of Gorakhnäth, a work that is, in the main, an expansion on the Whereas most alchemical works list six to nine metals, the AK lists Goraksa Sataka. As for the former, if it is not the Mabākālayogašāstra itself, it may be the Mahākāla Sambitā, a "Kāpālika" work also attributed to "Ād- twelve (2.1.9); it lists twenty-three, rather than the standard eight uparasas (2.1.4-8). Its list of the "eighteen" samskaras does not include "body work" inatha"!276 There exist south Indian recensions of the KbV, which comprise (sarīra yoga); appended to it are a list of subsidiary operations. The AK's only the first chapter of northern versions; this first chapter is moreover (2.9.1-97) discussion of the sixty-four divine herbs (divyausadhīs) is bor- often reproduced as the second chapter of the Yogakundali Upanisad. Else-

170 Chapter Five

where, three verses of the KhV are reproduced in the fifteenth-century SIX HYP (3.33-35),277 in the light of which evidence we may conclude that the KhV is a fourteenth-century north Indian work. The KhV begins with a praise of itself (1.1-28), followed by an account Tantra in the Rasārnava of the kbecari mantra (1.29-40). The balance of the first chapter is devoted to a discussion of the mechanical preparations for and hathayogic practice of the so-called khecarī mudra (1.41-74).278 The second and longest chapter is an extended description of the siddhis, as well as of the subtle body-the cakras, nādīs, kalās, etc. A portion of this chapter (2.81-97) describes the effects of hathayogic practice on the humors (dosas) of Ayurvedic theory. Chapter three discusses the raising of the kundalini, her union, in the cra- 1. Chapter One of the Rasarnava: The Text in Its Tantric Context nial vault, with Siva, who is figured as an internal linga, and the realization, on the part of the practitioner, of bodily liberation. This is described in As we have already noted, the use of alchemical imagery was, for several terms of the reintegration, or telescoping back into their subtler essences, centuries, a pervasive fixture of both Hindu and Buddhist tantrism. This of the five elements (3.49-52). chapter looks at this phenomenon from another perspective and examines It is, however, by virtue of its short final chapter that the KhV stands as what the alchemists themselves had to say about Kaula and Sakta tantrism. a paradigmatic Siddha alchemical text. This emphasis is clearly stated in No alchemical source has more to say on tantric theory and practice in the chapter's opening verse: "I will now speak of the 'divine herbal medi- general, and the Kaulas in particular, than does the eleventh-century RA, cines' (divyānyausadhāni). Without herbal medicines, absolutely no siddbis decidedly the greatest work of tantric alchemy. can be realized." A short way into its discussion of these "herbal medicines," More than any other, this source attests to the highly cosmopolitan na- however, the KhV recommends the consumption of "mercury, sulfur, ture of the alchemical tradition. Written in a quite grammatical Sanskrit, orpiment, and realgar," the mineral mainstays of elixir alchemy. The work this text, while it was likely authored in western India, appears to be, like concludes (4.13) with the following statement: "Having risen at daybreak, many of the alchemical classics, the production of a quite homogeneous- one should eat a decoction of mercury, the sap of the silk-cotton tree, and albeit geographically widespread-tradition. This homogeneity may be at- sulfur, [blended] together with the three sweets [honey, sugar, and butter]. tributed in part to the sociology of alchemical knowledge to which I al- In the space of six months, one becomes unaging and immortal." This is luded in previous chapters: the authors of these texts were most probably precisely the regime of the Yogis described by François Bernier, a regime court-based brahmin physicians to Hindu, Buddhist, or Muslim rulers. which, taken in the morning, "restored the body to perfect health."279 What sets the RA apart from all other alchemical works, indeed, from the great majority of the Hindu Tantras in general, is the rhetorical flourish its author gives to its opening chapter. I therefore reproduce, as much for their style as for their content, a number of excerpts from this introductory diatribe, in which the alchemical gnosis is favorably compared and con- trasted to the other philosophical and religious schools and movements of the day. Our author begins to delineate his position in the fourth verse of this chapter. The Goddess (Devi), Bhairava's interlocutor, addressing him for the first time, says, "O God among gods, O Great God, O incinerator of Time and Love, O leader [śāsaka] of the Kaula, Mahākaula, Siddhakaula, etc. lineages!"2 This introduction immediately reminds us of the sectarian pedigree of Matsyendra: a member of the Siddha Kaula and heir to the

171

172 173 Chapter Six Tantra in the Rasārnava

Mahakaulas, he was the founder of the Yogini Kaula: the author of the RA fluids) is crucial to the alchemist's practice.12 However, "the fluid" (dravya) is thereby claiming a similar pedigree for his work.3 He nuances his posi- is to be withheld from "other women" (paradārāt).13 tion, however, in a number of later verses, in which he unambiguously con- It is, however, with our author's mention of the heterodox tantric mat- demns what he considers to be excessive practices on the part of some ters of eating cow meat (gomamsa) and drinking the liquor of immortality tantric groups. (amaravaruni) that the light truly dawns. Such expressions are common- So, for example, in verse 10 he states that "if liberation is to be identified places of the "code language" of hatha yoga. As a commentator explains on with the excitation of the female genitalia, would not even donkeys be lib- the subject of an identical passage in the fifteenth-century HYP, these two erated? Indeed, why are rams and bulls not liberated?"+ And two verses terms actually refer to the exalted practice of kbecari mudra, whereby the later: "If liberation came from utilizing one's semen, urine, and excre- yogin internally absorbs his own semen, which has been transmuted into ments, which of the races of dogs and swine would not be liberated?"5 And nectar through yogic practice.14 Neither meat nor alcohol is actually being in verse 24: "For those people who have lost their powers of reason consumed; here, the reference is to the hathayogic complement to alchemi- through indulgence in liquor, flesh, sexual intercourse, and the male and cal operations, a complementarity that is the hallmark of "the [alchemical] female organs, the mercurial science is exceedingly difficult to realize."6 Work in two parts."15 The two verses that immediately follow (1.25-26) are placed in counter- The tantric alchemist tempers his tantrism and sublimates certain Kaula point to all that has preceded: "However (tu), mercury is not perfected by erotico-mystical practices with hathayogic and laboratory techniques (even those who lack the teaching of the [true] lineage (kula) and who have no if many of these are called kāpālika techniques).16 This sublimation, of the desire for the true doctrine. They, O Goddess, are [like men] drinking a male and female sexual essences so vital to the yoginī cults of the Vidyā mirage.7 He who eats cow meat [and] who drinks the liquor of immortality, Pītha, into mineral essences to be manipulated by the alchemist, is clearly him I consider to be one of the kula [and] a knower of rasa. Other rasa stated by Siva: "You, O Goddess, are the mother of all beings, and I am the experts are inferior."8 eternal father, and that which was generated from the great sexual union At first blush, we would appear to be in the presence of a text that con- of us two, that is rasa."17 tradicts itself. The truth is of a somewhat more subtle order, however. The The hathayogic component of tantric alchemy will concern us through- attacks on the excessive indulgence in erotico-mystical practices and the out much of the balance of this book; let us first, however, take a longer fire makaras are just that: Siva's alchemical revelation is, generally speaking, look at those elements of the RA which may be designated as properly a "post-Trika reformation" one. Following the lead of Matsyendra and per- "tantric." As I will presently demonstrate, the RA remains mainstream in haps Abhinavagupta in toning down the more extreme mortuary and erotic its tantrism, in its use of mantra, yantra, initiation, worship, etc. At the elements of earlier tantrism, this alchemical Tantra presents itself as a same time, it purports to be superior to both the orthodox Siva Siddhanta "meta-Śākta-Saiva" work, superior not only to orthodox Saiva Siddhānta, and the more heterodox Kaula forms of Tantra in the results it promises. but also to the practices of the earlier Kula and prereform Kaulism. For, whereas the former can offer liberation only at death and the latter As noted in a previous chapter, the author of the RA does consider the liberation only in life, tantric alchemy offers certain, verifiable bodily im- alchemical school to constitute a lineage, a line of transmission (kula) with mortality in life, an immortality whose effects are enduring.18 a kula-teaching of its own. ° Because the lineage nectar of the alchem- The opening benediction of the RA sets the tone of the entire work: ical lineage (sampradaya) is,10 like that of other tantric sects, transmitted "He in whom everything [dwells], from whom everything [is issued], who through female sexual emissions (called siddhadravya in the RA), he also is everything and everywhere, who is all-encompassing and eternal, salu- prescribes sexual intercourse and erotico-mystical worship as means to al- tations to that universal soul!"1° Following the Goddess's salutation, Siva's chemical transformation (with an exception made for yogins, who are ad- first words speak to the goals of every nondualist tantric practitioner, re- monished to remain celibate in their preparation of mercury)." Elsewhere, gardless of sectarian affiliation: "Eternal youth, immortality of the body, a female "laboratory assistant" (and, especially, her sexual and menstrual and the attainment of an identity of nature with Siva, that is, liberation in

174 175 Chapter Six Tantra in the Rasārnava

the body (jivanmukti), which is difficult even for the gods to attain."20 From female consort. Therefore, the alchemist is to worship a mercurial linga, here, the god launches into a condemnation of all other forms of religious the phallic emblem of Siva, visualizing it as embodying the eighteen-armed practice, reserving his greatest scorn for the six major schools of phi- Rasa-Bhairava and his consort Rasānkusī-Bhairavī. Of course, this mercu- losophy-the philosophical foundations of nontantric Hinduism-which rial linga, the support for such worship, is composed the stuff of Siva him- equate liberation with release from the body (i.e., which assume that libera- tion is realized only after death):21 self, i.e., of rasa, mercury.30 The alchemist's Rasa-Bhairava is an ectype of the eighteen-armed Svacchanda Bhairava, the deity of the Svacchanda Tan- The liberation that occurs when one drops dead22 is indeed a worth- tra (SuT).31 less liberation. [For in that case] a donkey is also liberated when he It is on the subject of this icon and its mercurial content that the RA drops dead. Liberation is indeed viewed in the six schools as [oc- and a number of other alchemical works wax the most eloquent in praises curring] when one drops dead, but that [kind of] liberation is not that bear all the marks of Hindu devotionalism. The most propitious days immediately perceptible, in the way that a myrobalan fruit in the for worship are the eighth and fourteenth days of the bright lunar fort- hand [is perceptible] (karāmalakavat).23 night; in addition to worship on these fixed dates, the observance of a spe- cial "alchemical celebration" (rasotsava) also produces miraculous results.32 The concrete emphasis of the alchemical quest is hereafter emphasized Viewing, touching, eating, recalling, worshipping, and making an offering repeatedly: to mercury yield six types of fruits. The merit one gains from viewing all Liberation [arises] from gnosis (jñāna), gnosis [arises] from the main- of the lingas in the world (including Kedarnath) is gained from the mere tenance of the vital breaths. Therefore, where there is stability, mer- viewing of mercury. By worshipping Siva in the form of calcinated mer- cury is empowered and the body is stabilized. Through the use of cury, with sandalwood, camphor, and saffron, one attains Siva's heaven, mercury one rapidly obtains a body that is unaging and immortal, Sivaloka. Eating mercury destroys the triad of sins (in word, deed, and and concentration of the mind.24 He who eats calcinated mercury thought), afflictions, and morbid states. The combined fruits that one (mrtasūtaka) truly obtains both transcendent and mundane knowl- might gain from worshipping one thousand of Siva's self-generated (svay- edge, and his mantras are effective.25 ambbü) lingas is reaped 100,000-fold by worshipping the rasa-linga of Śiva.33 But woe to those ignorant persons who do have no faith in mercury, Supernatural powers and bodily immortality, the goals of the tantric or who speak ill of it: they will suffer 1,000 rebirths as dogs, 300,000 as practitioner, cannot, however, be realized through alchemy alone. The ab- cats, 100,000 as donkeys, 100,000 as crows, 100,000 as worms, 100,000 as solute, Siva, too plays an active role in alchemical transformation: "So wild cocks, and 100,000 as vultures!34 It is to this cult of mercury that a long, however, as Siva does not descend to block the impurity that impedes the soul's liberation, and so long as one's fetters to this world remain un- number of authors refer when they identify rasacikitsa "chemiatry," i.e., the worship of chemicals.35 cut, there is no way that true discrimination can arise through the use of In addition to such external forms of devotional worship, the RA also calcinated mercury."26 Divine grace implies its human complement of de- discusses external worship of a distinctly tantric stamp, as well as internal votion (bhakti), which is also present in the alchemical synthesis, even if worship, mainly in the context of alchemical initiation, dīks.36 such is unusual to tantrism, which places a greater emphasis on ritualized forms of worship (pūjā, upāsana) than it does upon devotionalism.27 To which gods does the alchemist offer his devotion and worship? Not 2. Alchemical Mandalas in Chapter Two of the Rasārnava surprisingly, the divine revealers of the RA single themselves out, at the conclusion of the first chapter of the work, as the prime objects of the We must not hastily conclude from so much sound and fury on the matter alchemist's worship: these are Rasa-Bhairava, the "Mercurial Bhairava,"28 of the rasalinga that Siva in his mercurial and phallic form was the sole god and Rasankusī-Bhairavī, "Bhairavī, the Elephant Goad of Mercury."29 in the alchemical universe. If the alchemical tradition was a tantric one, Bhairava is, of course, one of the tantric names for Siva, and Bhairavī is his then it too should have had a number, nay a pantheon, of supporting deities

176 177 Chapter Six Tantra in the Rasārnava

in its micro- and macrocosms, and this it in fact does, as the second and Rasabhairava in sexual union with his consort Rasankuśī within his own third chapters of the RA reveal with a multitude of details. body.+ Following the installation of this divine pair, the alchemist con- As in nearly every Hindu tantric sect, all the gods of the alchemical structs his mandala, mentally and concretely, from the outside in,45 starting pantheon are either forms of Siva, his divine consort the Goddess, or one at the bhupura. On either side of each of the four gateways of this enclosure, of their divine attendants. Because the specific names of these deities he installs and worships eight demonic and semidivine Saiva guardians, evoked in the alchemical sources are so many indications of the various while at each of its eight cardinal points within the enclosure, he installs sectarian ties of the alchemical school, it behooves us to dwell on these for and worships an array of seven Puranic gods together with the goddess a moment. The RA compares mercury, inasmuch as it incorporates the five Umā.46 Working inwards, he installs and worships, on the petals of an outer elements, with the five-faced Sadāsiva.37 As already noted, however, Rasa- lotus, the ten female attendants (dūtīs), whose names are nothing other Bhairava, the supreme divinity of the RA, is an ectype of Svacchanda Bhair- female adjectival forms of alchemical processes and potencies.47 On the ava (whose cult is mentioned explicitly in the BhP).38 Indeed, the RA's icon- petals of the same lotus he worships the eight principal mineral reagents ography of Rasa-Bhairava, including his enthronement upon the mahāpreta (mabārasas)-all "male" in gender-that are employed, in combination Sadāsiva strongly indicates that the cult of Svacchanda Bhairava is its im- with mercury, in the processes of transmutation.48 On the petals of an inner plicit model. As noted in the last chapter, however, the RA and such texts lotus, he installs eight Rudras-who are to be identified with the eight as the BhP, MBbT, and KCM are also possessed of an important Sakta com- "Lords of Wisdom."+ On the pericarp of this lotus are placed the four ponent, referring on numerous occasions to the circle of the Mother God- śaktis-Mālinī, Parā, Aparā, and Parāparā-who are identified, respec- desses and Yoginīs and the erotico-mystical ritual associated with this con- tively, with the four alchemical energies (saktis): the energy of gold, of sul- figuration. The worship of the Yoginis and tantric forms of the Goddess fur (bala [for vali?]), of diamond, and of magnetite.50 In the midst of this are also evoked at a number of points in the RA.39 group of four and at the center of the mandala is Rasabhairava. Seated upon A highly refined current of nondualist Sakta tantrism was the Trika the shoulders of a preta, he engages in sexual intercourse with the goddess Kaula of the Eastern Transmission, the pūrvamnaya, so called for its cultic Rasankuśī, who is seated on his lap.51 Both of these deities are described in focus on a triad (tri-) of goddesses named Para, Aparā, and Parāparā.10 No detail. Rasabhairava is crystalline, eighteen-armed, five-faced, and three- doubt taking its lead from the Trika, the RA incorporates these three god- eyed, with flaming hair, tongues, eyes, etc., and a crescent moon in his desses, together with a fourth, Mãlinī,11 into a complex mandala, which it piled-up ascetic's chignon.52 He wears a tiger skin and has a serpent for his instructs the alchemist to construct for the external and internal tantric sacred thread. As for Rasankusī, her complexion is the color of molten worship of Rasabhairava and the goddess Rasānkusī."2 This diagram, whose gold, and she is clad in yellow-colored clothing. She has one face, three concrete support consists of a square stone slab measuring ten hand widths eyes, and four arms, which bear the elephant goad (ankuśa), rosary, and on each side, serves a dual purpose. It is to be used both for the conse- noose and show the gesture of protection. She is to be worshipped with cration of the newly constructed alchemical laboratory and for the sake of the five Vidyā seed mantras (bījas).53 gaining alchemical siddhis .* 3 Its form is that of a lotus composed of two This diagram, with its "superenthroned" divinities, is our key to the concentric sets of petals ringing a central pericarp, with the whole sur- theological and sectarian stance of the RA's author.54 In such a system, that rounded by the standard four-sided, four-gated bhupura enclosure. In the which is closer to the center of the mandala is closest to the ultimate reality lotus's pericarp, as well as on each of its petals and at the articulations of of the divinity at the heart of the mandala, from whom all radiates out- the bhupura, are placed the entire host of the RA's alchemical pantheon, wards, while that which is further from the center is lower, in terms of the into the heart of which we now descend. order of the divine revelation it represents. Working from the outside in, The alchemist begins his construction of this diagram, at the heart of the eight divinities of classical Hinduism are far removed from the center his laboratory, with the installation of a rasalinga, composed of mercury of this tantric system and therefore from the immediate concerns of the (rasa). Simultaneously with his installation of this concrete image, he uses alchemist. On and around the two concentric rings of lotus petals, the ten the "Elephant Goad of Mercury" mantra to establish the alchemical god female attendants (dūtīs) are a probable evocation of the Vidya Pītha's old

178 179 Chapter Six Tantra in the Rasārnava

and terrible cremation ground-based cult of the yoginis that was reformed twelfth century, this cult had spread south to the Tamil country, where it by Matsyendra's Yoginī Kaula cult.55 The basic configuration of the inner lotus is that of the old Saiva Siddhanta system, surmounted by that of the continues to thrive.61 Our author, who probably lived midway between Kashmir and Tamil Nadu (i.e., in the region of Maharashtra) in the elev- Trika goddesses; these are, in turn, superseded by the supreme divinities of the alchemical system, who are iconographic ectypes of Svacchandabhair- enth century, was therefore adapting the prevailing Saiva cults and doc- trines of his day into an alchemical "superenthronement" of the images of ava and Bālā Tripurasundarī, respectively. Rasabhairava and his consort Rasankusī over those of the supreme deities So it is that the eight Rudras located on the inner lotus are identified of other tantric sects and cults. with the eight Vidyeśvaras ("Lords of Wisdom") who ring Sadasiva in the worship mandala of the Saiddhāntika pūja and who stand as mediating cate- gories, in the dualist Saiva Siddhanta, between the "pure" and "impure" worlds, and as beings through whom the lower order of beings may attain 3. Mantra and Initiation in Chapters Two and Three of the Rasārnava to higher evolutionary stages.56 As for Sadasiva himself, he has become demoted, as he is in the iconography of Svacchandabhairava, to the func- Following its description of the rasamandala, the RA turns to a detailed tion of a mere vehicle: he is the preta upon whose shoulders Rasabhairava discussion of mantra, of those "tools for thought" by which the tantric is installed.57 In a later description, in which the RA (2.110-16) elaborates practitioner, like the Vedic ritualist before him, empowers himself to ma- on the mystic body of Rasabhairava himself, we find another adaptation of nipulate the divinities whose acoustic being is nothing other than the ag- SuT traditions. Here, the RA identifies various members of Rasa Bhairava's body (knees, loins, heart, mouth) with the five faces of Sadasiva, of which gregate of the phonemes of the Sanskrit alphabet. In a universe that is vi- bratory in nature, it is through these primal vibrations that one may most the last, Isana, is located between the deity's eyebrows. Upon his forehead efficiently return to the absolute source of all vibration. Mantras are thus is a half moon; above this is a dot or drop (bindu), surmounted by the nada. indispensable for the alchemist who, like every other tantric practitioner, Higher still, located at the fontanelle, is the goddess Sakti. Above her is has for his ultimate goal the conscious realization of a transcendent self the goddess Unmana ("Beyond Mind"), above whom is the triadic void or (i.e., identification of oneself with the Self of godhead). ether (trisunya). This configuration is identical to those found in SvT and At the culmination of its description of its alchemical mandala, the RA Netra Tantra descriptions of the yogic meditative utterance (uccara) of the identifies the mantra of Rasesvara, of (Siva) the "Lord of Mercury" as the syllable Om.58 In yet another description of its principal deity-in this thirty-two-syllabled aghora mantra of Svacchanda Bhairava "with at- case, his thirty-two-syllabled aghora mantra-the RA (2.68) further identi- tributes."62 The mantra of the goddess Rasānkusī, the rasānkuśa mantra fies Rasa Bhairava with Svacchanda Bhairava and his cult.59 As for Rasa- is given in encoded form.63 The RA also instructs the alchemist to make bhairava's consort, Rasankuśī, she is patterned not after Svacchandabhaira- use of a number of other garden-variety tantric mantras: these include va's śakti Aghoreśvarī, but rather after Bālā Trīpurasundarī of the Śrīvidyā the pranava (Om),64 the so'ham hamsa,65 the aghora, the mrtyunjaya,66 the Kaula.60 One might choose to see in these iconographic data a reflection seventeen-syllable kālī mantra,67 the kālapāša,68 mabāpāśupatāstra,69 etc. of sectarian developments occurring within the Saiva fold, in tenth- to These are to be variously used before, during, and after every alchemical eleventh-century Kashmir, where the cult of Svacchandabhairava was the operation to ensure success. dominant Saiva cult, as it continues to be today. During the same period, Other mantras, perhaps peculiar to this work, are the mala ("garland") the dominant Saiva doctrine in Kashmir was the Saiva Siddhanta, whose and mūla ("root") mantras which, like the Raseśvara mantra, combine seed particular dualistic interpretation of the SuT was countered by later Trika mantras with evocations of Bhairava-Siva.70 These latter mantras are em- nondualism, in the form of a (later authoritative) commentary on the text ployed in the practice of mantra-nyāsa, the imposition of mantras-each by Abhinavagupta's pupil Kșemarāja (fl. ca. A.D.1000-50). The new nondu- of which is identified with one or another tantric deity-upon the prac- alism also entered the Śrividyā Kaula cult of the goddess Tripurasundarī, titioner's own body, by which process that body is divinized.71 This divini- which became popular in Kashmir during the eleventh century. By the zation of the body is crucial to two components of tantric practice. These

180 181 Chapter Six Tantra in the Rasārnava

are initiation (dīksa), by which a teacher renders a disciple capable of enter- māmnāya. Kubjikā is, of course, the cult Goddess of the Western Transmis- ing upon the tantric path; and worship, because all tantric worship assumes sion. So too are the six yoginis: chapter 15 of the KM and chapter 24 of the an identification of self with the divine, in order that one may worship ȘSS are entitled "Sadyoginyah,"80 and there is in this tradition a fixation on one's self as the divinity. In both cases, this imposition of mantras effects the number six that one finds nowhere else in the Hindu corpus. As for the the transformation of the body, purging it of its gross elements and filling four Siddhas, these are to be identified either with the gurucatuska, the the void left by their implosion with divine essences, in the form of pho- "Four Gurus" (Mitranātha, Oddīšanātha, Șausthanātha, and Caryanātha) nemes.72 of Western Transmission tradition81 or with the four yuganathas, the Lords In the latter of these practices, this procedure is doubled by the simulta- of the Ages (Khagendranātha, Kūrmanātha, Meșanātha, and Macchanda- neous preparation of a concrete worship site, in this case the alchemical nätha) of the Siddha Cakra of earlier Kaula tradition.82 The eighteen Puru- laboratory. This practice, called bhutasuddhi, "purification of the elements," sas mentioned in the RA evoke the eighteen Nathas of the Kubjikanityāh- is a commonplace of tantric propitiation.73 We now turn to this practice, nikatilaka and Samvartārthaprakāśa, figures whom the SSS specifies are as it is delineated in the rather obscure language of chapter three of the RA. purusas, i.e., human males, as opposed to the divine (three or) four Siddhas, Here, Bhairava begins his exposition by saying that he will now describe of whom they are the sons and disciples. These eighteen are said to be "another acoustic image" (punaranyam mantramürtim) of the goddess Ra- "distinguished in couples"; that is, they are a redoubling of the Nine Näths sānkuśī.74 He begins his discussion by speaking of the fifth element, space who were intermediaries between the divine and the human in the Western or ether which is called the fifth "house." This he distinguishes from the Transmission.83 four other houses, which are "placed in front" (pradhana); that is, they are Following his installation of Kubjika, the alchemist of the RA "be- arranged like the arms of a cross, with the empty fifth house standing as a smears" the site with the Candaghanta ("The Goddess of Furious Bells," a central "courtyard." Because the element ether is the substrate of sound, tantric name of Durga), the mantra of the Goddess in the eastern house; this fifth house is also the "house of the hamsa [mantra]," in which "the takes up cow dung in order to smear the site with the Candabhairavī ("The siddhis, five makāras, etc. stand." It is within this space that the practitioner Furious Goddess of Terror"), the mantra of the Goddess in the southern will install the goddess Rasänkusi, in the form of mantras.75 house; and sprinkles water over it with the Candakāpālinī ("The Furious With this, the kalapāsa ("noose of Time") mantra of the "house of Goddess of the Skull),"84 the mantra of the Goddess in the northern house. the hamsa" is given: om siva brim krom hamsa. The practitioner is then Thereafter, he installs the seventeen-syllabled Kāli mantra in the stone instructed to repeat this mantra-called the "mercurial goad" (rasan- mortar, the thirty-two-syllabled Aghora mantra in the pestle, etc.85 While kuśa)-one hundred thousand times as the "doorkeeper" (pratibara) at the it is difficult to say whether the author of the RA was attempting, in this beginning of the worship. This recitation of "doorkeeper" mantras at the passage, to systematize the tantric āmmāyas or pīthas in his placement of opening of a worship is standard in Sākta-Saiva tantrism.76 the goddesses of the four houses around the central bamsa house, we can Next, "using that same [kālapāśa] mantra, he should take up the dung of maintain without hesitation that his was a syncretistic endeavor. Chapter a young calf, and using that alone, O Goddess, he should take up the ashes three contains mantras from Mantra Pītha (Aghora), Paścimāmnāya (Kub- of a funeral pyre,"77 to ritually prepare the "most excellent worship site jikā), Uttarāmnāya (seventeen-syllabled Kālī), and Śrīvidyā (Tripurā- in which Lord Mercury (Rasendra) dwells" and empower the alchemical bhairavī) worship traditions.86 apparatus he will use in his experimentation.78 He then sweeps the site with Now that the alchemist's body and laboratory have been ritually puri- a brush he has empowered by installing and worshipping Kubjika-the fied, the narrator Bhairava turns to the alchemical apparatus, consecrating goddess of the Western Transmission-in it (with the thirty-two-syllabled each with a particular mantra, until all that the alchemist will employ in kubjika mantra) together with her full sequence of subordinates. These sub- his ritual manipulations of elements has been purified and transformed.87 ordinates, which are so many seed mantras, are termed the four Siddhas, From this point on, all of these mundane instruments, chemicals, herbal the eighteen Purusas, the five elements, and the six Yoginīs.79 Here, we preparations have been transformed into so many tools (yantras) with appear to be in the presence of a number of elements proper to the paści- which the alchemist may master the one divine energy that surges through

182 183 Chapter Six Tantra in the Rasārnava

both the universe and his own body. He is now prepared to embark upon the alchemical life-cycle rites (samskaras) that will ultimately render him a penetrated by mercury, attains the state of gold, so the soul, penetrated by initiation (dīksā), attains the state of Siva." second Siva. It would appear from the above discussions that the matter of initiation In a sense, the relative paucity of material on diksa in the alchemical

(dīksā) into alchemical practice is quite entirely omitted from the opening sources may be explained by the fact that alchemical transmutation (vedha) chapters of the RA. Yet we find that the second chapter of the work is en- served as the original model for tantric initiation, especially that most pres-

titled "rules for initiation" (diksa-vidbāna) and that verses 36 to 83 of this tigious form called the "initiation of penetration," vedha[-mayī] dīksā.91 chapter give a lacunary account of dīksa.88 Many if not most of the standard Similarly, the purificatory mantras of the broader tantric discipline (bhūta- elements of diksā are present: a description of a suitable site for initiation, śuddhi, etc.) are also rendered redundant, inasmuch as purification is a con- i.e., upon which to build an alchemical laboratory (2.37-43); a description crete enterprise in alchemy.92 Lastly, the all-important yantras and manda-

of said initiation hall/laboratory (2.44-47a); information on the auspicious las used as supports for tantric meditation are themselves abstractions of

time for initiation (2.37b-48a); the preparatory nityapūja by the teacher alchemical apparatus.93 In the alchemical (and to a lesser extent the hatha-

including the visualization and worship of the supreme deities in the midst yogic) sphere, purification, identification with the divine, and initiation are of the elaborate alchemical mandala discussed above (2.48b-74); "terrible so many techniques for the concrete transformation of the "power sub-

offerings" (aghorabali) to demonic beings who would otherwise obstruct stances" that are the human and divine-and mineral, in the case of al-

the ritual (2.75-76); worship of the "elephant goad" or rasānkusī (2.77a); chemy-sexual fluids. Much of the balance of this study is devoted to de-

oblatory offerings (2.77b-78a); establishment of the consecration vase lineating the myriad ways in which the language of such transformations

(2.28b-79); annointment (abhiseka) of the initiate with water impregnated was a common language, shared by the interpenetrating alchemical, hatha-

with mantras and poured from that vase (2.80-81); teaching the mantra to yogic, Siddha, and tantric syntheses of medieval India.

the initiate (2.82a); and the celebratory all-night cakrapūjā for Kumārīs, Yoginīs, and Sadhakas (following Kaula, but not orthodox Siddhanta tradi- tions; 8zb-83). What is missing in this account is the diksa, or initiation proper (as opposed to abbiseka, annointment, by which one is empowered to give initiation to others), that intervenes, between the nityapūja and the abbiśeka ritual, in standard tantric procedure. Nowhere in this portion of the chapter do we find explicit references either to diksa in fire in which the initiate's karmas are destroyed with oblations into a consecrated fire or to the internal, yogic "initiation by penetration" (vedhamayī dīksā) of the Kaulas.89 What reasons might we adduce for the relative paucity of material in the RA on the transformative heart of diksa? It may be that the concrete details of initiation were too secret to be committed to writing and were passed on via oral tradition. Alternatively, it may be that the alchemical practices themselves were considered to be the realization and culmination of the process set in motion by dīksa. This latter explanation is supported in a quite explicit way in a number of nonalchemical sources, which identify mercurial transformations with diksa itself. So the tenth- to fourteenth- century Kulārnava Tantra,90 a seminal Kaula source, states that "just as iron,

185 Corresponding Hierarchies

SEVEN On the one hand, we find in the late classical Upanisads the origins of the mystic subtle physiology of the human body. In these works, we find early vague references to yogic techniques by which to generate trans- Corresponding Hierarchies: The Substance formative inner heat in tandem with the meditative, gnoseological realiza-

of the Alchemical Body tion (jnana) of the identity of individual soul (atman) with universal soul (bráhman). Together, these practices served to burn away and thereby ne- gate the accumulated fruits of prior acts (karma-phalam), thereby affording liberation (moksa) from the cycle of rebirths (samsāra). In this context, rasa was the term employed by the Maitri Upanisad to designate the high- est emanate of the highest guna, the strand known as sattvā, "essence" or I. Rasa and Rasāyana in Indian Systems of Thought "purity."s Early on, the internal dynamics of hatha yoga came to be viewed as so Rasa (from the same Indo-European root as the English word resin) has many interactions between lunar (candra), solar (sūrya), and vital (prāņa) one of the broadest semantic fields of any term in the Sanskrit language. principles, a transparent variation on the Vedic triad of rasa-agni-vāyu. In Originally employed in the Vedas1 to signify the waters and liquids in gen- later hathayogic sources, as well as in more general tantric works on yogic eral-vital fluids, animal juices, and vegetable saps-applications of the and psychological integration as the technical means (sädhana) to the real- term rasa have proliferated over the millennia to embrace such fields as ization of total autonomy, we encounter the important notion of samarasa. Ayurvedic medicine, batha yoga, alchemy, and Indian aesthetics. More gen- Literally, "of even rasa" or "of the same rasa," the term implies, according erally, rasa was and remains the "fluid essence" of Indian thought. If the to the numerous systems in which it is employed, a state of "fluid equilib- universe is a great pulsating flow of essence and manifestation, rasa is the rium," a condition of stasis in which the emanatory and resorptive impulses

fluid "stuff" of that flow. of the Absolute are balanced within the human microcosm.6 In the hatha- Already in Vedic speculation, rasa was homologized with water (ap), se- yogic system of the Nath Siddhas, the rasas in question are portrayed as men (retas), the vital fluid (ojas), herbal remedies (osadhi, ausadbi), nectar male and female "drops" (bindu), which are lunar and solar, seminal and (amrta), and soma as the moon, the Moon god (also named Candra), and sanguineous, Siva and Sakti, respectively. Realizing a state of equilibrium the mythic liquor of immortality.2 In its most general sense, rasa was, in between the two members of this pair is tantamount to the formation of a the sacrificial system of the Vedas, a term that could be applied to any "great drop" (mahābindu), a yogic zygote of sorts, from which the new, lib- oblation offered into fire (agni). The essence of said rasa was thence con- erated, all-powerful, and immortal self of the jīvanmukta emerges.7 veyed, via the wind (vayu), up to heaven where it was enjoyed by the gods.' Coeval with these original Upanisadic syntheses was the scientific disci- This triad, of rasa-agni-vayu, is one that has been reappropriated and pline of traditional Indian medicine, Åyurveda. It was in this field that an reformulated throughout the sweep of Indian history. With the internali- important new application of the term rasa was promulgated at a very early zation of the sacrifice, the major conceptual and practical breakthrough date. Here, the term rasa came to be applied to the product of the first of the Aranyakas and Upanisads, rasa became identified with the "body as stage of the digestion process: rasa was chyle, food that had been moistened oblation" whose fluid essences were cooked and transformed over the well- and broken down by saliva prior to swallowing. In a more extended sense, tempered fires of ascetic ardor (tapas), fires that were fanned by the winds rasa retained its signifaction, in Ayurvedic usage, as bodily fluid.8 This was of the vital breaths (prana) .* Once the bodily microcosm was transformed an echo both of Vedic identifications of rasa with the waters and of emerg- into the seat of the sacrifice (to the detriment of external sacrifice, which ing metaphysical systems that identified each of the five elements with a had been on the wane ever since the seventh century B.c in India), interest sensory organ and field of sensory activity. In Sämkhya, the element water, in the internal workings of the body became greatly expanded. Speculation rasa, was identified with the sense of taste. This notion gave rise, in Āy- and research took two directions, the one mystical and the other medical. urveda, to the system of the six rasas or tastes-sweet, acid, saline, hot

184

187 186 Chapter Seven Corresponding Hierarchies

and pungent, bitter, and astringent-with each of the tastes arising from a of this siddhi in these early sources. Here, rasa-rasayana was but one of eight mixture of water together with the other four gross elements, in varying magical powers (listed together, for example, with padalepa, the smearing proportions.° Emerging out of this Ayurvedic matrix, India's classic theory of the feet with a cream that produced the power of flight) fantasized by Buddhists and Hindus in the first centuries of the common era.1 of aesthetics, i.e., of taste (rasa)-through which the cultivated spectator reinterpreted the raw emotions (bbāvas) portrayed in drama, dance, and It is only later, with the alchemical Tantras of the tenth century and after, that laboratory methods for transmutation first come to be discussed literature into their corresponding cultivated rasas-was developed some systematically, under the heading of rasayana. It is also in this later period centuries later.10 Drawing on the Vedic tradition which maintained that it was water that we find references to alchemical practitioners as Rasa Siddhas and which upheld all plant and animal life, Ayurvedic theory takes rasa to be to the alchemical doctrine that Madhava terms "Raseśvara Darśana." What the support of all of the bodily constituents (dhatus). As we saw in chapter rendered this later form of Indian alchemy tantric was the pivotal notion two, digestion is conceived in Ayurveda as the serial "cooking" (pacana) of that alchemical mercury was nothing other than divine semen, and that its the bodily constituents on the constituent fires (dhatvagnis),11 with the end principal reagents (sulfur, gandhaka; mica, abbraka) were the uterine blood product of this process being semen in men and uterine blood in women. and sexual emissions of the Goddess.15 Working from these basic identifi- These in turn combine in conception to give rise to the human embryo. cations, the tantric alchemists were able to construct an integrated concep- Rasāyana (the "way of rasa"), the seventh branch (anga) of Indian medi- tual framework for their art, a framework that mirrored the general cine, is the most holistic and prestigious of all Ayurvedic systems of heal- worldview of Hindu tantrism. ing, taking the body to be an integrated whole, the microcosmic reflection Once again, bodily fluids, and the rasa-agni polarity of the archaic triad of the universal macrocosm. Its prestige also lies in the results it promises: we have already referred to on several occasions (reproduced here as fluid rasãyana is rejuvenation therapy which, combining clinical practice with mercury and fiery sulfur) stand as the basic categories of this new synthesis. the internal use of elixirs, affords long life, whence the classical statement So it is that Siva states in the RA that "because [mercury] is the rasa [vital of the Caraka Samhitā (6.1.7-8): "Long life, heightened memory and intel- fluid] of my body, one is to call it rasa [mercury]." A more elaborate etymol- ligence, freedom from disease, a healthy glow, good complexion, a deep, ogy is found in the RRS: "Because of its power to assimilate (rasanat) all powerful voice, great bodily and sensory powers, the capacity to see one's other metals, it is known as rasa. Or, indeed, it is considered rasa because it is effective (rasyate) in the destruction of aging, disease, and death."16 pronouncements realized, respectability, beauty-all these does one obtain from rasāyana. It is called rasāyana because it is a means to replenishing the The term rasa may also be employed in the plural in Hindu alchemy, in rasa and other dhatus of the body."12 which case the notion of "essential element" becomes operative.17 Here, In the same Ayurvedic context, the rasāyanas, in the plural, are the elixirs the sixteen principal mineral reagents are generally divided into eight ma- the physician employs in rasayana therapy;13 and it was out of this back- barasas and eight uparasas, primary and secondary rasas that are so called ground that the term rasāyana emerged as a blanket Sanskrit term for al- because they participate in the transformative powers of mercury, the rasa chemy. Here, the "way of rasa" was a mercurial path since rasa, the semen par excellence.18 It is in this context that the RC can portray mercury as Lord Rasa (raseśvara) who leads the [mahā-]rasas and uparasas to victory in or fluid essence of Siva, was identified with mercury, the transmuting ele- battle against disease and death.19 This same source also provides an Ay- ment par excellence. While we do not know at what time this absolutely crucial identification-of the vital sexual fluid (rasa) of the phallic god with urvedic usage of the compound rasa-rasäyana, which it uses to designate a mercurial elixir that is highly effective both therapeutically and alchemi- quicksilver, the fluid metal-came about, we can be sure of another, prior, cally (a usage not far removed from that of rasāyana in a number of alchem- "alchemical" use of the term rasāyana. This was the compound rasa- ical sources). rasāyana, which was employed, in Hindu and Buddhist sources alike, as far back as the second century A.D., to signify the supernatural power (siddhi) While rasāyana has been used as a generic term for alchemy in India since the tenth century, the means and ends of the discipline are more of alchemical transmutation and bodily transubstantiation. One finds no description whatsoever of a practical method (sadhana) for the realization properly identified by two other terms. These are dhātuvāda,20 "the doc-

188 189 Chapter Seven Corresponding Hierarchies

trine of elements" and dehavada, "the doctrine of the body." These two with the semen of Siva, and of sulfur (gandhaka), red arsenic (manahsilā) or branches of tantric alchemy form a unified whole: dhatuvada alchemy, mica (abhraka) with the menstrual blood (khapuspa, rajas, śonita, ārtava) or which concerns itself with the transmutation of base metals into gold, is sexual emission (virya) of the Goddess. This identification of bodily fluids the necessary propaedeutic to debavāda alchemy, the alchemy of elixirs of or compounds with the metals did not originate with the yogic or alchemi- bodily immortality. Their complementarity is underscored in a classic pas- cal traditions. The moon is identified in the Vedas with the draft of immor- sage from the RA: "As in metal, so in the body. Mercury ought always to be tality, i.e., soma or amrta, as well as with the vital fluids of living beings employed in this way. When it penetrates a metal and the body, [mercury] (rasa):25 this last term comes to be employed to signify mercury in alchemi- behaves in an identical way. First test mercury on a metal, then use it on cal traditions. The divine origins of gold, the subtlest of the metals (dhātus), the body."21 The RRA explains rasendra, the "lord of rasas" (one of the five is described in the Satapatha Brähmana, which states that it arose from the names for mercury), in similar fashion: "It is called rasendra because seed of Viśvarūpa, whose body, after he was slain by Indra, was shattered through its proper use, both metals and the body become possessed of into fragments by Tvastr, the divine smith of the Vedas: "From his seed rasa."22 In the end, all is but a continuity of the same fluid (rasa): divine his form [rūpa] flowed and became gold."26 In the same early source we semen transmutes base metals into gold and transforms human semen into find a reference to gold refined and produced from the body of Prajāpati nectar by which the mortal practitioner is rendered an immortal superman, through the heat of his austerities.27 Brahma, a later form of Prajāpati, is a second Siva,23 whose bodily secretions, like Siva's own, transmute base born from a primal union of water and the seed of Agni, called the Golden metals into gold. Egg (hiranyagarbha).28 And of course, the statement "gold is immortality" For reasons unknown to us-although we may hazard a guess that mer- becomes a leitmotif of the brahmanic sacrifice.29 cury poisoning had a part to play-there was, in the north Indian heartland In these prealchemical traditions, gold is considered to be the ultimate at any rate, a gradual shift of emphasis, from the thirteenth century on- product of a long period of germination or gestation within the womb of wards, away from the goal of bodily immortality and towards a more thera- the earth. This notion, also held by the Roman Seneca (ca. A.D. 60) and, peutic use of mercurials and other "elixir" preparations. Here, rasa šāstra much later, by Arabic (the eleventh-century Avicenna) and Chinese think- ("mercurial science") came to apply the scientific-and some of the mys- ers,30 is already voiced, in India, in the pre-second century A.D. Rāmāyana.31 tic-discoveries and techniques of tantric alchemy to the Ayurvedic disci- In this source, the six-headed child god Skanda-Karttikeya is born from pline. So it is that the internal application of mercury and other mineral the seed of Agni (Fire) who ejaculates into every opening of the goddess and metallic rasas would come to constitute a subdivision of Ayurvedic Earth when he sees the goddess Ganga "in her magnificent splendor." rasāyana. It is in this subordinate form, as Ayurvedic pharmacy, that tantric Earth generates an embryo, which becomes the metals gold and silver; and alchemy-which gave up nearly all pretension, by the fourteenth century, from the splendor of these metals the divine child arises. From the various to being a path to immortality-has persisted over the centuries and con- rays emitted by this effulgent child, as well as from various residues pro- tinues to thrive down to the present day throughout India.24 Siddha al- duced in his childbirth-afterbirths of sorts-are produced the metals chemy, with its persistent emphasis on bodily transformation and immor- copper, iron, tin, and lead. Taken together, these are the classic "six metals" tality through the combined disciplines of yoga and alchemy, has remained of Indian tradition, first listed in the Yajur Veda.32 more faithful to the original spirit and goals of tantric alchemy than has This myth is important for two reasons. First, metals are treated in it as rasa šāstra. living substances, with distinctions between animal, vegetable, and mineral blurred: "bodies" of whatever composition or configuration may interact with and even be generated from one another. In this case, fire, fluid, and 2. Myths of the Origins of Minerals from Divine Bodily Fluids earth give birth to a god, the six metals, and the natural coloring of certain trees and plants. Here, it is the emanatory dynamic of the proto-Vedānta The most concrete point of intersection between the yogic and alchemical metaphysics of the Upanisads-a system that is very similar to the emana- traditions in India lies in the identification of mercury (rasa, pārada, sūta) tion and participation of Neoplatonist thought-that facilitates such anal-

190 191

Chapter Seven Corresponding Hierarchies

ogies between the animal, vegetable, and mineral kingdoms. The universe cury) identifies the six metals with (the seminal emissions of) six gods: gold in all its parts is a single organic entity, with all that exists on the great arises from Agni, silver from the moon (Candra), copper from the sun chain of being the internal flux of a divinely constituted whole, to which (Sürya), tin from Indra, lead from the serpent king Vāsuki, and iron from all emanated form necessarily returns in the fullness of time. As such, all Yama, lord of the dead.39 Other sources, such as the twelfth-century Loba- in the universe is shot through, "like the scent in a flower,"33 with the divine sarvasvam and the late nineteenth-century Rasendra Bbāskara, trace the ori- essence. Moreover, since all exists on the same continuum of this divine gins of a great number of other elements and metals back to the bodies, outpouring, all is comparable, even identifiable. The corresponding hier- blood, semen, sweat, and tears of a still wider array of gods and demons.10 archies of the elements, body fluids, metals, tastes, cakras, etc. are not dis- Nath Siddha sources identify the indestructible body of the Siddha with crete and distinct from one another; rather, they are so many expressions mercury41 and a hierarchy of holy men with the hierarchy of metals.42 of a single unified whole. Thus they are interchangeable, even interpene- trating. In such a quite nearly "pantheistic" context, metals are not merely comparable to body fluids, they are body fluids.34 3. Blood and Semen The second point that arrests our attention in this origin myth of gold is the god who issues from it: this is a very early account of the birth of By far the richest and most elaborate homologies drawn between human Skanda-Kārttikeya.35 This is a myth that comes to constitute a perma- and divine body fluids on the one hand and elements from the mineral nent fixture in every Saiva Purāna, where Skanda (also known as Kārtti- world on the other are that between mercury and the semen of the phallic keya, Murugan, Subrahmanian, Kumāra, etc.) is depicted as arising from god Siva and that between sulfur and the uterine or menstrual blood of his the seed Siva emits into the mouth of Agni (fire) and thence into a bed of consort, the Goddess. It is in the context of the myth of the divine birth of reeds on the shore of the Ganges, where he is suckled by the six daughters Skanda, once again, that we find the myth of the origin of mercury, in the of Daksa, the Krttikās (whence his name Kārttikeya and his six heads), in RC (15.4-12), RRS (1.23-29), AK (1.8-15), and a number of other alchem- preparation for his victorious battle against the demon Taraka. In a wide ical sources.43 The novel element in these accounts is that it is mercury array of other Saiva myths, Siva's seed is said to be fluid gold.36 rather than gold that is produced from that portion of the semen emitted Once this central metaphor was established-i.e., that gold (and, as we by Śiva which did not contribute directly to the generation of Skanda. will see shortly, mercury and sulfur) were the quintessential reproductive As in the gold origin myth, Agni has a role to play in the alchemical fluids of the great tantric gods-the floodgates were thrown open, as it account of the origin of mercury, as it is he who drops from his mouth the were, to similar myths of the origins of the lesser elements. So, we learn semen that becomes the transmuting element: that diamond (vajra) arose from drops of amrta that fell from the gods' mouths to earth after they had churned the Ocean of Milk (RA 6.65-66). In a hidden cave of the Himalayas, Siva and the Goddess were Tourmaline gems (vaikranta) are the drops of blood the Buffalo Demon engaged in love-play when the gods came to plead that they produce shed when he was slain by Durga (RA 6.123-25; RRS 2.56-57). Blue vitriol Skanda in order that the demon Taraka, who was threatening the or copper sulfate (sasyaka) is the poison and nectar vomited by Garuda after entire universe, might be destroyed.4 The god Agni [Fire], having his theft of Soma (RA 7.39; RC 10.71-72; RRS 2.119);37 and copper pyrites taken the form of a pigeon, peeked in through the round window of (maksika) the blood that flowed from a wound in Krsna's foot (RA 7.3-4). the apartment within which they had closed themselves in order to Natural gold (prakrta-svarna) arose from the abstract quality of activity generate a son. Upon seeing Agni, Siva, filled with shame, shed his (rajoguna) that set the universe in motion; another form of "natural" gold seed. That seed, of blinding brilliance, fell into the mouth of Fire. (svabaja-svarņa) is the golden placenta-deposited atop Mount Meru- Fire, unable to bear the heat of that seed, spit it into the waters of the from which the god Brahma, the golden embryo, was born at the begin- goddess Ganga [the Ganges river]. She too, overcome by the inten- ning of our cosmic eon.38 sity of that semen, pushed it with her waves to her shores, where a The ca. sixteenth-century Rasakāmadhenu (Wish-fulfilling Cow of Mer- child, Skanda, was born.

192 193 Chapter Seven Corresponding Hierarchies

In those places where Agni dropped that seed from his mouth45 longer lying in that place, the tide having carried them out into the [on his way to the Ganges], it burrowed into the ground to form five wells of mercury, 100 yojanas [about 900 miles] in depth. It is there middle of the Ocean of Milk.

that that semen is found today, in the form of mercury.46 During the churning of the Ocean of Milk, that blood rose to the surface, together with the nectar (amrta). All the gods and the anti-

It is by drinking this elixir that the gods themselves enjoy immortality. gods were pleased with the aroma (gandha) that wafted on her blood.

Jealous of humans who might do the same and so threaten their hegemony, The gods and antigods thereby said, "May this be called gandhaka

the gods requested that Siva adulterate it with impurities, and so it is that ("aromatic": sulphur). May it be used in the calcination and fixing of

the mercury found on earth comes in a variety of colors. The mercury mercury. May those qualities that are found in mercury also be found

of the eastern well is white, that of the western well yellow, and that of in this sulphur." Thus [sulphur] is called gandhaka here on earth.

the southern well, controlled by Nagas, is blue. The well at the center of this pentadic mercurial mandala contains "mixed" mercury, called miśraka, This myth appears to have both an alchemical and a geological explana- tion. First, the opening verses of the Mātrkabheda Tantra (1.9-14) seem to which is said to be mayura, variegated like the throat or tail feathers of a evoke the myth of the origin of sulfur in a description of an alchemical peacock. Of the five wells in which this mercury is said to have established itself, however, the RRS and AK single out the northern well, in which operation that involves the heating of sewn cloth and menstrual blood to- gether in a vessel of milk.52 Elsewhere, this account has been interpreted mercury is red and pure, as superior to the other four.47 This northern well is implicitly identified with Darada-desa (Dardistan, as a mythic description of a submarine volcanic eruption, from which sub-

located to the north of Kashmir),18 a likely overland source of mercury for marine deposits of sulfur would have formed. Similarly, the Puranic my-

India in the age of tantric alchemy. The name of this land evokes the prime thology of the submarine mare (badava), and the doomsday fire (bādava)

Sanskrit term for cinnabar,49 which is darada. So too does the color as- that issues from her mouth, may refer to the phenomenon of submarine

signed to the mercury of this well, given that cinnabar (a compound in gas vents or fire jets. As such, badava would be the oceanic equivalent of

which one atom of mercury is chemically bonded to one atom of sulfur), naturally occurring terrestrial gas vents (jvalamukbi), which are considered

the most common naturally occurring mercurial ore, is red in color. in India to be geothermic manifestations of the Goddess. The most popu- lar pilgrimage site in all of Himachal Pradesh is Jvālamukhī, a blue flame Sulfur too has its origin myth, which parallels that of mercury inasmuch as it is the sexual essence of the Goddess that gives rise to this primary that burns within a natural freshwater spring, identified with the tongue of the Goddess. It has also long been an important center for the Nath alchemical reagent, the female counterpart to male mercury.50 This myth Siddhas.53 is framed in the Puranic account of the churning of the Ocean of Milk, from which the gods gained (through the intervention of Dhanvantari, the In the light of these glosses on this myth and the hard fact that India

divine physician and revealer of Ayurveda) the amrta that rendered them possesses no measurable deposits of mercurial ores, the following state-

immortal and capable of defeating the antigods, the asuras, in their sempi- ment, from the Shorter Oxford Economic Atlas, appears to be significant: "al- most all the world's mercury is obtained from the red sulphide mineral ternal war for universal supremacy.5 cinnabar, HgS, but a little of the metal occurs naturally as small globules

Once, when the Goddess was sporting together with the female Wiz- associated with cinnabar in certain rocks in the neighborhood of old volca-

ards (vidyādharīs), Siddha-maidens, Goddesses, and celestial nymphs nic regions where the minerals were deposited from hot aqueous solutions during volcanic activity. Many hot springs in such regions are still depos- in the Ocean of Milk off the coast of White Island, blood issued from iting some cinnabar."54 her body. That wonderfully scented blood was captivating. Because This statement implies that there may have been and continue to be it stained her clothes, she left these on the shore, and bathed to purify traces of mercury in regions of geothermal activity in India. This would herself. When she wished to return to the town, her clothes were no explain, for example, the purported existence of a "mercurial pool" located

194 195 Chapter Seven Corresponding Hierarchies

in the vicinity of a hot sulfur spring in the region of Kedārnāth, an ancient (oxide of lead) to color the bones of the dead, as a means of ensuring them Saiva pilgrimage site in the Garhwal Himalayas. In the light of these data, an eternal afterlife. Here, the "element metal" cinnabar, identified with the myth of the origin of sulfur may be read as a reference to those sites at living blood, was long employed as an elixir of immortality for the living which alchemists garnered the materia prima of their art. as well.61 Cinnabar, mercuric sulfide-composed of mercurial semen and sulfu- In India, where gold and other metals are also employed in the cult of rous uterine blood-is a mineral hierophany of the sexual union of Siva the dead,62 vermilion (sindura: red mercuric oxide or synthetic cinnabar) and the Goddess, a union that, according to Hindu tradition, creates and has long been employed as a substitute for or a complement to blood offer- sustains the universe. Indeed, the Tamil, Malayalam, and Sinhalese term ings.63 Of course, blood is also closely identified with fertility and procre- for cinnabar is lingam, a clear evocation of its sexual valence.55 Now, cinna- ation, and so it is in India that married women wear vermilion in the part bar is also one of the eight amalgamated elements (astabandha) used in the of their hair as a sign of connubial felicity. The term saubhāgyam refers installation of divine idols, and it is here that the obligatory installation of both to this happy state and to the lead peroxide or vermilion that is the a mercurial linga at the heart of the alchemical laboratory becomes most mark thereof: here, the term hingula (another common term for cinnabar) meaningful. Here, the rasalinga is formed of either an amalgam of gold56 may also be employed.64 Also in India, red arsenic (manabśila) is identified and mercury, or of sulfur and mercury,57 i.e., of synthetic cinnabar. Else- with the uterine blood of the Goddess: this is especially the case at the where, the RA (11.4) compares the mortar in which mercury is pounded to "seat" (pītha)65 of the goddess called Kāmākhyā (Gauhati, Assam), at which the yoni-shaped chasing (pitha) in which the image of a Siva-linga is set, site the goddess's yoni (in the widespread Puranic myth of the dismember- and the mercury in question to the linga itself. In another passage (11.102), ment of the goddess Sati) is said to have taken the form of a great block of the same text notes that mercury that has been calcinated in a particular red arsenic where it fell to earth.66 mixture of gemstones, minerals, and herbs becomes phallomorphic (lingā- In the more abstract schemata of yonipūja, the tantric worship of the kāra). This appears to be the north Indian cognate to the production, in female sexual organ, the yoni is represented as a downturned triangle, at south Indian alchemy, of "mercurial phalluses," called gulikās (the equiva- the heart of which is, once again, Kamarupa, the abode of the goddess lent of the Sanskrit gutikā, "pill").58 Kāmākhyā, who is identified with the kundalinī and with feminine materi- Here as well, the MBbT substitutes human menstrual blood for its di- ality (prakrti) in the form of menstrual flux (pusparūpini). At the pītha of vine mineral equivalent, as if to emphasize the sexual symbolism of this Kāmākhyā itself, Assamese tantrics identify their "lineage nectar" (kulām- operation: "Place tamarind (cincini, the botanical equivalent of the goddess rta) with the Goddess's menstrual fluid (or the commingled sexual fluids of Kubjikā) and mercury together on the support. Mix these together ... so Siva and the Goddess); it is at the time of the Goddess's menses, in August- that the mixture resembles mud ... Having shaped it into a linga, one September, when the water that oozes from this stone becomes reddish in should then harden it [in the following way]. One should tie it up inside color, that their annual gathering takes place.67 cloth that [has been soaked] with menstrual blood [and place it] over a The powers of the Goddess's menstrual blood are directly linked to its fire [fueled by] cow dung. Some heating will be necessary in order that it redness. Persons suffering from leucodermia (pānduroga) come to Kāmā- become hard."59 khyā to smear their bodies with the Goddess's menstrual blood, thus "col- The red color of compounds of mercury and sulfur inspires all manner oring" themselves with the ooze of her red mineral hierophany. The ocher of homologization in Hindu traditions. Indeed, the compound cinnabar is dye (geruā) with which the Näth Siddhas color their traditional garb (kan- naturally red in color; sulfur and mercury both redden when heated, the tha) is said to originate from the blood of the goddess Parvati, who resorted latter taking on the ruby (manikya) red color of the rising sun when oxi- to self-mutilation to dye the robe of Gorakhnäth. Cloth dyed in this color dized at a temperature of 300° C.60 This redness, identified with that of is considered by the Näth Siddhas to aid in yogic semen retention (virya- blood, has long been associated with the life force or vital energy in both stambha śakti).68 This red mineral manifestation of the Goddess at Kāmā- India and China. The ancient Chinese used both cinnabar and red minium khyā is also capable, according to the Kālikā Purāna, of transmuting base

196 197 Chapter Seven Corresponding Hierarchies

metals into gold. According to medieval legend, a sealed cave, containing the "womb-cave" of the Goddess. We will return to the symbolism of this a pot of mercury (rasakumbba) the drink of which confers immortality, is cave in the next chapter.71 located in its vicinity.69 In the alchemical laboratory, such homologies come to be practically As we have noted, accounts of such conjoined mineral manifestations applied through techniques that involve the commingling of human, of Siva and the Goddess in the form of mercurial "wells" and sulfurous or divine, and mineral blood and semen. Here, the point of convergence be- arsenious phenomena may have their basis in geological fact. Of the five tween these interpenetrating systems is the person of the alchemist's sacred "alchemical" sites of India, three-Hinglāj Devī, Gaurī Kund (near female laboratory assistant, who is described in the RA and a number of Kedārnāth), and Kāmākhyā-combine a mineral manifestation of the other sources. This assistant may be of four types: kākiņī (a woman who Goddess with a mercurial pool or source. The two other sites, Girnar (Ju- menstruates in the dark half of the lunar month, i.e., when the moon is nagadh district, Gujarat) and Srisailam (Kurnool district, Andhra Pradesh) on the wane), kīkanī (who menstruates in the middle of the lunar month), appear to lack the feminine complement to masculine mercury. Hinglāj kāñcikācinī (who menstruates in the bright half of the lunar month), or pad- Devī is, in a sense, a geographical and mythological mirror to Kāmākhyā. minì (who menstruates either on the full moon or the new moon).72 The Located at the westernmost fringe of the Indian subcontinent, on the edge names for this assistant (kākiņī means "cowrie maiden"; padminī means of Baluchistan, this shrine is said to be that site at which the goddess Sati's "lotus maiden") appear to be direct references to her sexual organ, which fontanelle, her brahmarandhra, fell. Like Kāmākhyā, Hinglāj features a is also described in these texts, in ideal terms, as "resembling an aśvattha number of caves and wells, of which one, known for its geothermic activi- (Ficus religiosus) leaf."73 In fact, everything about the alchemist's female ties, is called Candrakūpa, "Moon Well." Hinglāj is clearly named for cin- assistant is ideal: she is young, beautiful, raven-haired, doe-eyed, perfectly nabar (hingula-ja); and one of the many local names by which she is known proportioned, fair of speech and light of laughter, gentle when she kisses is "Red Goddess." and embraces, a lover of dairy products, and a devotee of Śiva.74 The keepsake which pilgrims take with them from Hinglāj Devī is not, It is, however, her sexual organ as well as the menstrual blood that flows however, a piece of cinnabar, but rather a lump (or lumps, polished and from it that seems to be at the center of interest here, and this for a very strung together in a rosary) of nummelite, a type of yellow limestone, concrete reason, which the RRS (6.34) explains. "She who menstruates [lit- called thumra. These lumps are variously described as the petrified seed of erally, "she who flowers," puspavati] in the dark half of the lunar month is the Creator or a combination of cereal grains (khichri, a rough rice and most excellent for the fixation of mercury [rasabandha] in alchemical prac- lentil dish favored by Siva) and the blood of a demon. As we will later show, tice." In what way is the assistant useful to the alchemist's craft? "For rice often doubles for semen in yogic legend, so the combination of semen twenty-one days, she is to eat sulfur [mixed with clarified butter] ... Her and blood is indeed present in the thumra of the Hinglāj Devī pilgrimage. Indeed, it is with a large thumra stone that Nath Siddhas at the Gorakh- menstrual blood [then] becomes efficacious in the fixation and calcination of mercury."75 Other sources instruct the alchemist to place said mercury, pur monastery ensured easy delivery to women in the former half of this century: the stones were washed in water and the liquid then given to the wrapped in a piece of cloth, deep in his wife's vulva, to the same end; or to macerate sulfur in a woman's menstrual blood in order to increase its women to drink.70 Elsewhere, a number of "womb-caves" of the Goddess, which are legion potency.76 The BhP, which terms sulfur umayoni-samsarga ("that which issued from throughout south and east Asia, are considered sacred by Hindus, Bud- the vulva of [the Goddess] Uma"), states that mercury can be bound and dhists, and Taoists alike. One such cave, called Ca-ri, in extreme southeast- made complete (samskrta) only by entering into that yoni, i.e., sulfur. Re- ern Tibet, has long been revered by Hindus and Buddhists as the abode of productive symbolism is explicit here: mercury, of which one name is sūta Siva Maheśvara and the Goddess (or, in Tibetan Buddhism, of Cakrasam- ("that which is born, generated"), and which transmutes through the vara and Vajravarāhī). This cave, entered through a narrow fissure identi- agency of bijas ("seeds") of noble metals in combination with sulfur or fied as the Goddess's yoni, is said to contain a lake of sindūra (red mercuric oxide), identified with the commingled sexual fluids of divinities inhabiting mica, here enters into the womb of the Goddess (sulfur) to become acti- vated.77 Alternatively, the same source states that the alchemist may bind

198 199 Chapter Seven Corresponding Hierarchies

or stabilize mercury by placing it in his urethra (pu[m]randhre) "together Here, the ingestion of the mineral equivalent of sexual fluids stands as with the menstrual blood of Gauri." Here, Gauri may be taken as a name a metaphor for the consummation of love. In the alchemical sources, such for the Goddess, in which case it is sulfur that is being manipulated here; emissions, both divine and human, are taken to be elixirs of immortality. or it may be taken adjectivally to stand for any "fair woman," in which case So the RA states: "Until such time as one eats Siva's seed, that is, mercury, it is human menstrual blood that he is commingling with his own seed where is his liberation, where is the maintenance of his body?"8+ The RHT (tadbījena), in all probability through the yogic technique of urethral suc- elaborates: "A divine body is afforded through the union of the [sexual] tion (vajroli mudra).78 In this latter case, we are once more in the presence emissions of Siva and the Goddess."85 And, in yet another source, "a wom- of the "power substances" of early tantric practice. A reference in this an's flower mixed with semen, eaten for a year" is proposed as an elixir.86 source (BbP 1.77) to the khecari-cakra ("sky-going circle") of ritualized tan- Here, the flower (puspa) in question is menstrual blood (especially that of tric sex, as well as to the kundagola[ka] as an elixir in other alchemical a virgin, a woman who has yet to be deflowered),87 and we should not fail sources, further documents this connection between ritual and alchemical to note here that sulfur in its purest form is known in this tradition, as it is uses of the products of human and divine sexual intercourse.79 in the west, as "flowers of sulfur" (puspita gandhaka).88 At the end of such processes, which serve to stabilize and fortify mer- Such analogies carry over into the vegetable kingdom as well, in which cury, this divine semen in its mineral form may at last be ingested by the the "divine herbs" (divyausadhi) are said to grow in those places in which alchemist in the ultimate samskāra of sarīra yoga, transubstantiation. Here, Śiva and the Goddess once made love.89 A case in point is the cāndāli plant, the RA stipulates that sexual intercourse is essential to the activation of the undoubtedly named for those outcaste women whose menstrual blood has mercury the alchemist has ingested (and adds that said mercury turns him perennially been prized by tantrics for its transformative powers. The root into a sexual animal, whence, no doubt, this text's insistence on his female of the candali exudes a red milk (ksīra) that is used, like sulfur and menstrual laboratory assistant's many charms).80 blood, for the fixing of mercury.90 Elsewhere, Alberuni's celebrated account This remarkable interplay-between human sexual fluids and their di- of Vyadi's discovery of the alchemical elixir portrays this alchemist as vine and mineral counterparts-spills over into the genre of courtly lit- searching in vain for an essential ingredient called raktamala, which he as- erature as well, as in the ca. A.D. 1540 Padmavat of Muhammad Jayasī,81 sumes to be red myrobalan, a plant with red, bloodlike sap. It is only when who allegorically portrays the star-crossed love of his hero and heroine in he bumps his head and bleeds into his alchemical cauldron that his elixir alchemical terms. Here King Ratansen, who has taken up the celibate life "takes." The same source relates an account of a Siddha who employs the of a Näth Siddha, is first questioned by the pining Padmavatī's female com- red milk of the plant called thohar to transmute the body of a shepherd and panions: his dog into gold.91 A moment ago, I mentioned the practice known as vajroli mudrā, Where did you lose the fair plant [biravā lonā: Padmavatī] that makes through which the male practitioner enabled himself to tap directly into the silver and the gold? Are you unable to blend yellow arsenic [Pad- the fluid "power substance" naturally occurring in his tantric consort, the mavatī] with mercury [Ratansen]? Why have you forsaken sulfur yogini. In technical terms, vajrolī mudrā is urethral suction or, more prosai- [Padmavatī] for the cold fare [of the celibate yogin]?"82 [Ratansen re- cally, the "fountain pen technique," by which the male practitioner, having plies, as if to Padmavatī herself]: " ... If I were but able to blend the ejaculated into his female partner, withdraws his own semen, now catalyzed mercury with the yellow arsenic [i.e., to abandon my vow of celi- through its interaction with her sexual essence or uterine blood, back into bacy], I would give up my life just to see the sulfur ... I who was his own body. In so doing, he also draws back into himself, along with his [transparent, i.e., dispassionate as] mica, having been turned over the own refined seed, a certain quantity of that female essence which may in fire [of love], am become [red, impassioned as] cinnabar. My brass turn serve to catalyze the yogic processes (the raising of the kundalinī, etc.) body can become yellow gold, if you but choose to make it so ... by which his semen becomes transmuted into nectar. If the mercury cannot swallow the sulfur, how shall I call my life A number of sources note that the woman may do the same as the man, yellow arsenic?"83 i.e., retain his seed within herself to thereby catalyze her own yogic trans-

200 201 Chapter Seven Corresponding Hierarchies

formations.92 Such techniques on the part of the woman, which are de- is made to double for male seed in certain preparatory techniques for scribed only in the hathayogic sources, appear to be redundant when vajrolī mudrā, as described by a contemporary practitioner, the Aghori Vi- viewed from the erotico-mystical perspective of a number of Kaula tradi- malananda: tions. It is a basic assumption among these traditions that the fluid lineage or clan nectar, the subtle fluid essence of liberating consciousness, is natu- To learn Vajroli you must first thoroughly clean out your body ... rally present in women, and it is precisely for this reason that the male The urinary passage is cleansed by sucking water through the penis tantric practitioner engages in sexual intercourse with her. This was the into the bladder and through the ureters into the kidneys and then basic doctrine of Matsyendranāth's venerable Yoginī Kaula: women, be- releasing it ... After you have been able to suck up water through the cause they are embodiments of the Goddess and because it is through their penis and hold a bladder full of water for three hours, you proceed to "wombs" that the lineage is perpetuated, have something that men do not; use milk, to cool the genital organs. Then ... clarified butter to lu- it is therefore necessary for males to tap into the female in order that that bricate. Honey next; it is very sticky and hard to make flow upward. boundless source of energy be activated within them. This fluid power sub- Finally you do it with mercury, which is extremely heavy ... When stance (dravya) or lineage nectar (kulamrta), also simply known by the term you can hold a bladder full of mercury for three hours without spill- "true being" (sadbbäva)-the purest substance found in the human body- ing a single drop, you have reached the level of the first qualification is unique to women in their multiple roles as sexual consorts, practitioners [for Vajroli]. A woman prepares for Vajroli in exactly the same way.97 of yoga, and biological mothers.93 Let us consider for a moment the etymology of this rather obscure term All three roles are present in the fluid exchange involved in the practice vajrolī mudrā. While we can be certain that mudra means "seal" and that of vajroli mudra. When the sadbhava of a tantric consort is conjoined with vajra means "penis," the -oli suffix (which also appears in the names of the the semen of her male partner, a "great fluid" (mabarasa) is produced, out related practices of amarolī and sahajolī mudrās) is more problematic. In at of which a perfect yogic child (yoginibhu) is born. Because women are natu- least one other context, -oli appears to have the sense of "sphere" or "ball": rally endowed with a greater abundance of vital breath (prana) than men, one manuscript of the SSP (1.73) calls the developing embryo garbh-oli.98 the raising of the kundalini is easier, even natural, for them; thus, when her By extension, the -oli suffix may refer to the spherical form of the womb of partner emits his seed into her womb, it commingles with her vital breath the consort, which would give a reading of "the seal of the penis in the and, transformed through her yogic energy, becomes pure mahārasa. In womb." erotico-mystical practice, it is originally in the female partner that the pure In the concomitant hathayogic techniques associated with urethral suc- substance [the mahārasa] resides. It is this that is transmitted to the male part- tion, the upward rise of the combined sexual fluids culled by the male prac- ner and is returned to the female in an endlessly renewed exchange effected titioner through vajroli mudra does not culminate in his lower abdomen. through the "two mouths"-i.e., the sexual organs of the partners: "This Urethral suction is but the beginning of a process, which in fact raises these knowledge beyond duality ... is rightly said to be transmitted from mouth combined sexual essences, via the medial channel of the susumņā nādī, all to mouth ... it goes from the chief mouth [that of the yogini] to [the ad- the way up to the cranial vault. In this light, the fluid essence of the subtle ept's] own mouth and vice versa. Bestower of immortality and youth, it body travels along a single channel, which runs from the tip of the penis is named kula, supreme."9 This practice becomes internalized in certain to the top of the head. We are aided in our considerations of this technique hathayogic sources, in which one internally drinks the "brilliant white- by descriptions from two Buddhist sources, the fifth-century A.D. Mahāyā- red" nectar.95 nasūtrālamkāra Sūtra09 and a Tibetan Buddhist alchemical source, 100 which The common end result of both the yogic and the erotico-mystical describe what we have been calling vajrolī mudrā as maithunasya parāvrtti, techniques is identical to that obtained through the ingestion of the al- "retroverting [the product] of intercourse," and bcud len, the "extraction of chemical cognates to these male and female power substances: white hair the essence," respectively. Here, a drop of fluid containing the combined and wrinkles disappear, and one becomes rejuvenated.96 Indeed, mercury male and female essences is carried upward along the susumnā to the cranial

203 202 Chapter Seven Corresponding Hierarchies

4. A Cosmopolitan Alchemical Theme vault. In these two Buddhist contexts, the practitioner's realization is of a gnoseological order: concomitant to the raising of this drop, he comes to These multifarious identifications and interactions between human and realize for himself the nondifference of samsāra and nirvāna, of suffering mineral equivalents of divine sexual fluids are brought to their analogical existence and the extinction of suffering, and thereby experiences the great conclusions in a truly astonishing set of instructions for the extraction of bliss (mabāsukha).101 These Buddhist examples find their direct homologue in the Hindu tan- mercury from its mercurial well in Darada country:109 tric notion of samarasa or samarasya, in which the process initiated by vaj- Upon seeing a well-adorned maiden who, having bathed after first roli mudra is brought to culmination in the yogin's cranial vault.102 While coming into season, [rides by] mounted upon a horse, mercury, the term becomes exalted to the point of being identified with samadhi and which is found in wells, [becomes] possessed of a desire to seize her, other enstatic, beatific states, the concrete and original sense of the term [and] rushes up out [of its well]. Upon seeing it, she gallops away. appears to be that employed by the Näth Siddhas, for whom samarasa is The mercury pursues her for the distance of one yojana [eight to nine the commingling of two drops (bindus). One of these, solar and red103 is miles]. [When] that [mercury which is] born of Siva then quickly re- identified with Sakti, while the other, lunar and white, is identified with turns to the well, it is caught in troughs dug in its path. That mercury, Siva; combined into a great drop (mabābindu) of "white-red nectar," they because of its heaviness, fell from the mouth of Agni in Darada-deśa. form a yogic zygote of sorts.104 That mercury, absorbed into the surface of the earth there, came A south Indian alchemical source, the Kongana 3000, identifies the male to remain in that country. By placing that [mercury-rich] ore in a drop (which it calls bindu) with Siva and mercury, and the female drop pātana-yantra, one kills the mercury. (which it calls nada) with Sakti and sulfur, which, when combined, produce What makes this description of the extraction of mercury from its bodily immortality in a process called kalpasādhanai: this is the homologue northern well all the more astonishing is the fact that this is an account of the Näth Siddhas' käyakalpa, a general term meaning "bodily regenera- that is found in at least two other alchemical traditions. One of these is a tion."105 It may be that the HYP (4.96) is referring to the same yogic union, seventeenth-century Chinese encyclopedia, the Ho han sans ts'ai t'ou bui, in similar alchemical terms, when it states that "the mercurial mind, when whose section entitled yin-shui ("silver water, quicksilver"), gives the same calcinated by the nada of sulfur, becomes bound and immobilized." information, but locates its mercurial well "in the land of Fou-lin"-i.e., The fruit of this union, of the yogin's commingled male and female es- Syria-"far to the west."110 This detail is corroborated by the earliest ex- sences, is nothing less than the new, supernatural, immortal self that will tant source that we have for this account: Syriac recensions of the alchem- now emerge from the "husk" of the gross body, as the result of this yogic ical works of Pseudo-Zosimus, dated to the fourth to fifth century A.D., process.106 This is the siddha- or vajra-deba, the yogin's perfected or dia- describe how mercury is induced to rise up out of its well when a beautiful mond body, his innate immortal essence which he has now restored to its naked maiden walks past it and then runs quickly away. Young men attack pristine perfection by burning away the gross and refining the subtle ele- the flowing metal with hatchets and cut it up into bars."" This account is ments within himself. This is a body possessed of all the yogic siddhis, further corroborated by the Syrian toponym Bir es Zeibaq, which means including the ability to transmute base metals into gold with one's ex- "Well of Quicksilver."112 crements, spittle, etc.107 It is this new self that is the subject of the first of The implications of these three parallel accounts are staggering. First, the banis of Gorakhnath: "In the empty vault [of the cranium] a child is they attest to the very cosmopolitan nature of the world's classical alchem- making sounds. How can one give him a name?"108 Such a child of yoga is ical traditions. We can well imagine that the Silk Road, which was India's to be placed in parallel with the yoginībhu mentioned a moment ago; it also pipeline for the raw materials of alchemy, would also have served as a con- evokes the transformation, even the procreation, of a disciple by his guru duit for alchemical knowledge and legend, of the sort alluded to here, be- through initiation, a process I discuss in the penultimate chapter of this tween the Mediterranean world and east and south Asia. More than this, book.

205 204 Chapter Seven Corresponding Hierarchies

both the western Ptolemy and the Indian Puranas located the Daradas we may glimpse, behind the language of this extraction technique, refer- (called Daradrai by Ptolemy) was the Indus River region. In his version ences to another very cosmopolitan tradition-that of the unicorn. The of the Indo-Persian account related above, Somadeva calls this country presence of a virgin, a horse, and the theme of capturing an elusive "game" are all present in the alchemical account. While the western unicorn leg- Parada-desa, "Mercury Land," rather than Darada-desa. The Pāradas were end was very probably born out of the Indian myth of Rsyasrnga,113 it is also located, by ancient and medieval Indian and western sources, in this only in the west that the hunting of the unicorn-with the aid of a virgin, same general region as the Daradas, i.e., to the northwest or west of India, who tamed it by grasping its horn-and the transmutative properties of its in either Afghanistan or Baluchistan.118 Now, the Indus River valley is also horn are brought to the fore.114 that region from which there have been unearthed a great wealth of clay This same theme is treated in Persian alchemical legend as well, in an seals, many of which are figured with what have to be the world's most account of a wondrous creature called the Physician of the Sea.115 The ancient images of unicorns (as distinguished from rhinoceri, which are also Physician of the Sea is described as having a golden stone set into its fore- represented). But unicorns are not all that this region has in common with head which, when removed, transmutes base metals into gold and cures all the ancient and medieval west. The goddess Hinglaj, whose shrine stands diseases. The Jabirian corpus of Persian alchemy, generally dated to the astride the coastal range dividing the Indus River plains and the highlands of Baluchistan, also appears to be possessed of a venerable pedigree. In ninth to tenth century, describes the capture of one of these fishlike crea- tures which, upon being netted and brought aboard a ship off the coast of fact, she is referred to by her many Muslim devotees as Bībī Nānī, the "lady an island called Sindiyyat,116 shows itself, after the fashion of a mermaid, grandmother," a denomination which evokes a wide array of cognate names to be a beautiful woman. She remains on ship, bears a son by one of the for other ancient mother goddesses, most particularly the Persian Anahitā, sailors, and later jumps back into the sea, where she becomes a great sea the Nanaea of the apocryphal Book of Maccabees (2 Mac 1:13-15), the Chaldean Nãna, and others.119 monster who swallows the entire ocean during a great storm. These legends present more problems than can possibly be treated here. While there is no well of mercury per se at Hinglaj, the principal feature Not the least of these is the direction of transmission of alchemical legend, of her pilgrimage is Candrakupa, the "Moon Well," a place of impressive lore, techniques, and raw materials throughout the first fifteen centuries of geothermic activity, whose waters do indeed periodically gush up out of its depths and which also is said to belch fire. Pilgrims who come here drop the common era. As Joseph Needham has demonstrated, China stands, according the best evidence, as the primal source for the world's transmu- pellets of bread into this pool and interpret their destiny in terms of the tational and elixir alchemy. According to Needham's historical reconstruc- eruptions that follow. Similar descriptions-of an Indian "Well of Proof" whose waters contained realgar and were considered to possess magical tion, the first-century A.D. Chinese technique of kim or chin, "aurifaction," would have been carried west to the Mediterranean world in perhaps the properties and of a nearby fiery crater that gave off a lead-colored flame but which never overflowed-are found in the third-century A.D. Life of third century A.D .. This Chinese term would then have been transliterated, by Pseudo-Zosimus, as chymeia or chemeia, later arabicized into al-chymeia, Apollonius of Philostratus.120 The site also features rock-hewn images of the and introduced into European traditions as alchymia, alchemy.117 If Need- sun and moon, the description and location of which correspond, most intriguingly, to data found in the fourth-century B.c. Indika of Ctesias, a ham is correct, then Syria, which received its alchemy from China in the third century A.D., would have "exported" its legendary extraction tech- work that was written during a protracted stay in the Persian Achaemenian nique back to the east, via our thirteenth-century Indian sources, to China court of Artaxes Mnemon II. Sun and moon are, as well, absolutely pivotal images in the hathayogic system of the Näth Siddhas, who are the peren- in the seventeenth century. It is impossible to say with any certainty that this was in fact the route nial pilgrim's guides to and custodians of Hinglaj and whose illustrious founders are said to have come from points as far west as Khorasan, in that this tradition took. It is equally impossible to determine the ways in eastern Iran. We may therefore see in the isolated shrine of Hinglaj- which the many alchemical exchanges of this long period, effected along which remained, at least down into the middle of the present century, only the Silk Road, may have occurred. Another detail of the Indian alchemical accessible by a twenty-day camel journey-a possible crossroads for the account is, however, tantalizing in this regard. One of the regions in which

z06 207 Chapter Seven Corresponding Hierarchies exchange of alchemical knowledge and imagery in the ancient and medi- wholes as well as parts. We may take, by way of example, the hierarchy of eval world.121 the five gross elements in Samkhya and the parallel case of the five metals It is possible to dig down to a still deeper mythological stratum and see in alchemy.125 In both systems, that which is higher on the hierarchy ema- in these accounts variations on the quite widespread Indo-European theme nates, differentiates, or penetrates (vyapana) into that which is lower, and of a well whose fiery liquid contents erupt in pursuit of a woman but are is likewise capable of resorbing (laya) that which is lower back into itself. eventually channeled and thereby neutralized. If this is an Indo-European That which is higher is subtler than-and therefore capable of encom- mytheme (the myth is attested in Rome as well as Ireland and India), it passing, even imploding-that which is lower back into itself without be- would necessarily date from the third millennium B.c. or earlier.122 In this ing modified.126 Thus, each of the higher elements telescopes into itself the case, Pseudo-Zosimus's alchemical gloss would have been a much later cumulation of all elements lower than it, together with their characteristic addition. properties and qualities. In Samkhyan metaphysics, the interfaces of the five lowest differentiates of Spirit (purusa) with materiality (prakrti) are the five gross elements (ma- 5. Pentads in Metallic and Yogic Hierarchies of the Elements babhūtas) that make up the material world as perceived by the five senses: these are, in descending order, ether, air, fire, water, and earth.127 The ha- The RA (12.83) states that "Sadasiva exists [in the state of] mercury, [which thayogic subtle body respects the same hierarchy as it incorporates these is] composed of the five elements."123 One scholar has analyzed this state- five elements into the five lower cakras, which are aligned along the spinal ment in the following highly literal terms: "cinnabar is mined from the column from the perineum up to the throat. In Ayurveda as well, the body earth, on roasting yields mercury, which is a liquid metal, which boils to is imagined to be composed of the five elements: the element earth corre- form a gas."124 While this explanation makes perfect sense, it is probably sponds to that which is solid in the body, water to bodily fluids, fire to body not what the author of the RA had in mind when he composed this verse: heat, air to breath, and ether to the bodily orifices.128 he was rather referring to the five elements as the warp and weft of a nature Also in Ãyurveda, the process of digestion involves the progressive that was the self-manifestation of the god Siva, the Absolute. refinement of food from chyle (rasa) to semen (sukra) through seven in- Drawing on a map of reality whose first grid may well have been the tervening stages, which are called dhatus.129 It is semen, the end product Vedic fire altar, the alchemical and other later Hindu traditions, both eso- of digestion, that becomes the raw fuel that the techniques of hatha yoga teric and exoteric, assumed the cosmos to be pentadic in its structure but progressively refine into nectar (amrta). Dbatu is also the term employed unified in its essence. However many constituent parts the universe had, in the alchemical tradition for certain of its basic elements: the dhatus are these were always reducible to familiar sets of five, fives that radiated out- the metals.130 Like the elements in Samkhya, the metals may be reckoned ward from a common divine center of gravity like so many spokes on a as five in number; this figure is often rounded up to six or seven (the same wheel. Sämkhya and Vedänta, the most perennial and pervasive metaphy- as the number of dhatus in Ayurveda and the cakras in yoga) through the sical systems of India, emerged in the same Upanisadic period as did the addition of iron and one or another alloy. In the RA, six metals are listed: earliest traditions of Ayurveda. Indeed, some scholars argue that Āyurveda gold, silver, copper, tin, lead, and iron. These metals already appear, in a was the source-rather than a reflection-of a number of Upanisadic (and hierarchized schema, in the Chāndogya Upanisad.131 Buddhist) doctrines and theories. Also in this period there appeared, in The most explicit connections between these parallel systems are those such classical Upanisads as the sixth-century B.c. Kāthaka (6.16-17), early made between the five mabābhūtas of Sāmkhya and the five cakras of hatha rudimentary descriptions of the physiology of the subtle body. yoga on the one hand and the five metals of alchemy on the other. In this These three systems-the Upanisadic, the Ayurvedic, and the yogic- latter case, as well as the alchemical synthesis that would follow some centuries later- were all hierarchical systems. More than this, Indian thinkers were quick the absorption of herbs takes place in lead. Lead is absorbed in tin, to tabulate all manner of correspondences between these systems, between tin in copper, copper in silver, silver in gold, and gold in mercury.

208 209 Chapter Seven Corresponding Hierarchies Mercury that has been calcinated in sulfur is capable of integrating which gives [-da] the far shore [para-] [of immortality]"), that from the the body and is capable of "cooking" all the elements. In the same southern well is rasendra ("lord of rasas"), that from the western well is sūta way that earth is absorbed into water, water is absorbed into fire, fire ("that which was engendered [by Siva]") and that from the central well is into air, air into ether, ether into the soul (atman), and the soul into miśraka ("mixed; that which mixes with all other rasas").137 the Absolute (brahman). Therefore, just as earth and the other ele- A number of sources hierarchically order these naturally occurring ments associated with it, by virtue of being gross (sthūla) are absorbed forms of mercury in the mineral world. This they do by assigning a "caste" into water, they and all that follow [water, on the hierarchy of ele- (albeit in an idiosyncratic way) to the mercury of each of the four direc- ments] are absorbed into the most subtle (sūksma) element, which is tions: white pärada, located to the east, is identified with the brahmin caste; Bráhman. In the same way, wood and other herbal elements are ab- blue-black rasendra, to the south, is ksatriya; yellow sūta, to the west, is sorbed into the more subtle element lead, such that they and all that vaisya; and red rasa to the north is sūdra. No value is given for the central follow [lead, on the hierarchy of metals] are absorbed into mercury.132 well, which is said, in this text, to be the source of the mercury found in all the other wells.138 The KCM assigns castes to four types of mica (abbraka); The south Indian discipline of Siddha medicine goes one step further the RA does the same for four types of diamond (vajra),139 and expressly identifies the five elements with five metals. Here, gold is Mercury is again cast in the hierarchical mold of the five gross elements identified with earth, lead with water, copper with fire, iron with air, and when the alchemical texts discuss the five "[modes of] going away" (gatis) of mercury.140 The gatis are in fact the five ways in which mercury is lost zinc with ether.13 In hatha yoga, the raising of the yogin's energy, figured as the female through physical or chemical reaction.141 The first of these, mala-ga kundalinī serpent, through the system of the cakras, also effects a resorption ("dross-gone") is mercury that is lost in solid form; jala-ga ("water-gone") of gross into subtler elements. Thus, when the kundalinī rises from the is mercury lost in liquid form; hamsa-ga ("goose-gone") is mercury lost müladbara cakra (located at the level of the perineum and identified with into air, through evaporation; and dhūmra-ga ("smoke-gone") is mercury earth) to the svädhisthana (located at the level of the sexual organs and iden- lost through fire. The final gati of mercury is jīva-ga,142 "soul-gone" mer- tified with water), the element earth becomes resorbed into and encom- cury, which is the loss of mercury's vital essence into ether, space. Whereas passed by the element water. Likewise, water is resorbed into fire in the the first four gatis of mercury may be remedied through laboratory pro- third cakra, the manipura (at the level of the navel); fire into air in the anā- cedures, the fifth and "invisible" loss of mercury can be countered only bata cakra (at the level of the heart); and air into ether in the visuddhi cakra through the use of the invisible and mystic techniques of mantradbyāna, (located in the throat).134 As in Sämkhya, hatha yoga, and the other hier- mantric meditation.143 It is this subtlest form of mercury, which dwells archical systems, so too in alchemy: that which is higher encompasses, ab- "within the pericarp [of the lotus] of the ether (vyoma) of the heart," which sorbs, that which is lower. By way of demonstration, I cite the ca. tenth- the RA instructs the alchemist to recall in order that he may be cleansed of century RHT:135 "Woody plants are absorbed into lead, lead into tin, and all sins of present and past lives.144 tin likewise into copper. Copper [is absorbed] into silver, silver into gold, The same hierarchical principle applies in discussions of the ways in and gold is absorbed into mercury."136 which mercury transmutes (vedha) metals. Here too, the hierarchy of the Earlier in this chapter, I identified a set of pentads unique to the alchem- five elements is fully respected: lepa [-vedha] ([transmutation through] ical tradition itself. When mercury first comes into the world, it divides "smearing") corresponds to the solid element earth, ksepa ("casting, [into itself into five parts, burrowing down into five deep wells, in each of molten metal]") to the liquid element water, kunta ("dart[ing]" to air, dhū- five color-coded directions: the mercury in the northern well is red, while mra ("smok[ing]") to fire, and sabda ("speech, speaking") to ether.145 This that in the eastern well is white, southern blue, western yellow, and the last form of transmutation occurs when the alchemist, holding a mercurial pill in his mouth, blows upon a base metal.146 The parallels between these center variegated, like a "peacock's tail." The five classic names or varieties (bhedas) of mercury correspond to these five wells: mercury originating two hierarchized mercurial pentads, that of gati and of vedha, is especially from the northern well is rasa; that from the eastern well is pārada ("that clear with respect to the last two members of each of the hierarchies. In

210 2II

Chapter Seven Corresponding Hierarchies

both cases, the fourth member is dhumra, smoke, while the fifth element, for the element ether, which is identified, in Samkhya, as the substrate of corresponding to the element ether, sound, and hearing, involves mystic the sound tanmatra as well as of the sense of hearing. Like vyoma, gagana speech (sabda). It is in this context that the alchemical use of mantras, de- and kha are also terms that signify both mica and ether. Located as it is at tailed in the last chapter, is to be understood. the summit of the hierarchy of the five elements, ether is, both temporally The alchemical tradition knows of a great number of other pentads and logically, the first of the elements that emanates from the second guna, which, while not arranged hierarchically, nevertheless bear witness, by the rajas. As such, it constitutes a turning point in the play of nature (pra- simple fact that they are presented as groups of five, to the pervasiveness krti):159 it is the most subtle differentiate to be found in the visible world of Samkhyan metaphysics. These include the five gemstones (ratnas), the and the least subtle differentiate to be found in the invisible world. five salts (lavanas), the five greater and lesser poisons (visas), the five animal It is fitting that ether should be identified, through the term vyoma, with oils (vasās), urines (mūtras), biles (pittas), excrements (vitas), as well as the the sexual essence of the Goddess. In Samkhya, and even more so in a classes (vargas) of the five white, red, and yellow substances.147 Here too, number of tantric classifications of the tattvas-cosmic essences and meta- alchemy is a particular case of a general rule: other realms of tantra know physical categories-the pivot between pure essence and impure creation of the five faces of Siva, the five makāras,148 the five amrtas,149 the five prin- is the maya tattva, which is identified with the creative power and activity cipal pithas,150 the five clans (kulas), the five transmissions (amnāyas),151 the (rajas,160 the second of the three gunas) of the female principle, the God- five primordial Nath Siddhas,152 the five colors, tastes, and dispositions,153 dess, who is the manifestation, or act of reflection, of Siva, consciousness. and so on. The position of ether is therefore an elevated one and is translated, in the subtle physiology of hatha yoga, into the identification of the cranial vault-called the "sphere of the void" (sūnya-mandala)-with vyoma[n], 6. Bird and Serpent gagana, and kba, i.e., ether. Here, however, ether is as much an extremely subtle element located In general, it is the upper and lower poles of a hierarchical system that conceptually at the acme of a hierarchy as it is a locus of empty space, a define the nature of the entire range of interrelationships that obtain be- cavity, hollow, or void (which one is to fill with the "stuff" of mystic experi- tween that system's constituent elements.154 So it is that we should expect ence)161 located spatially at the summit of the subtle body. It is in both of to find the mineral or metallic equivalents of ether and earth, the highest these senses that ether is associated with hamsa, a term translated as "swan," and lowest of the five elements, playing salient roles within the alchem- "goose," or "migratory bird." Since the time of the Vedas, the hamsa has ical system. It is not, however, gold, the highest of the metal dhatus, that been the bird of predilection for authors wishing to discuss the movements doubles for the element ether in tantric alchemy, but rather mica (called of the vital breath (prana, a term etymologically related to atman, soul or abbraka, gagana, vyoma or kha),155 one of the three major mineral manifes- spirit). Indeed, the Rg Veda (4.40.5) itself calls the ether (kha) the "seat" of tations of the Goddess. Mica is already identified with the Goddess's sexual the bamsa, and a series of later sources, continuing down to the Tantras, emission in the RA (6.1-3) and RRS (2.2), and a short account of its origin identifies inbreathing and outbreathing with the syllables ham and sah. is found in the Rasakamadhenu: 156 "One day the hillborn Goddess saw the Hamsa is at once the sound that the breath makes when one inhales mind-boggling Hara; the 'semen' (virya) she shed produced brilliant mica." and exhales and the vibratory resonance (nada) of the Absolute that the The Rasendra Bhaskara reproduces this account with minor emendations, practitioner hears internally in the course of the spiritual exercises that adding that "because it fell from the firmament (gaganat) it is also called lead to samadbi, total yogic integration.162 In the subtle body, the bamsa is gagana; because it wandered (abhramat) through the clouds (abhra), it is identified with the empty (sunya) medial channel through which the vital called abhraka." 157 These are the two principal Sanskrit terms for mica. energy, breath, and consciousness descend in the individuation of the Ab- Another name for mica, the Goddess's sexual emission, is vyoma. Now, solute into an individual being.163 In all of these traditions, breathing is vyoma is also, at least from the time of the Susruta Samhita,158 a term used tantamount to identifying the individual soul with the absolute: hamso 'ham

212 213 Chapter Seven Corresponding Hierarchies

is a palindrome that can be read either as "the goose! the goose!" or "I am semen he had shed into the mouth of Agni (Fire), who had taken the form That," i.e., "I, atman, am That, brabman." The cosmic goose, honking in of a pigeon. Somadeva then goes on to say that the void, thus becomes a metaphor for the resorption-of individual breath, sound, and soul-into the Absolute. those dhätus, in overflowing abundance, were indeed the same as nec- The void, the empty space of the heart, susumņā nādī, or cranial vault, tar .! 71 That which dripped from Siva's hand truly became mercury is also termed kha-hole, cavity, empty space-whence statements in ha- (pārada). From the dross [of those dhatus] born from the immortal thayogic sources concerning the free-floating state of the meditative mind, and unaging supreme god, there arose the metals (dhatus), eight in intellect, or consciousness in the ether: this is khecara, "moving in the number. And that rasa which fell from the mouth of Agni [who had ether."164 This principle becomes concretized on an alchemical level as taken the form of a pigeon] was indeed used by the Lord of Serpents, well. Mercury, when bound, is said to become khecari, possessed of the who wished to leave behind old age and death. But that [mercury] power of flight, a power it transmits to the alchemist who holds a capsule which, while being drunk [by the snake] fell to the ground, did-due of said mercury in his mouth.165 If, however, it is heated before it has been to its heaviness-fall through the ground into one hundred yojana- properly prepared, or too rapidly, it can be lost to evaporation. Here, it is deep wells. said that until its wings have been clipped (paksa-ccheda)166 through various This interaction-between a bird (Agni, in the form of a pigeon, from stabilizing techniques, unrefined mercury becomes subject to flight, i.e., whose mouth drops of divine semen/nectar/mercury fall) and a snake (the evaporation (patana), and may be "gone like a goose" (hamsaga).167 Here, Lord of Serpents, who drinks up what has fallen in order to gain immor- mercury behaves like the hamsa of the vital breaths: unless these are teth- tality)-harks back to mythic themes from both the Vedas172 and ancient ered by yogic discipline, the breath will eventually fly up out of the body Mesopotamia.173 In these archaic accounts, a bird is identified with the to be lost forever, and the person will die. The ideal thus becomes one of theft of an elixir of eternal life, of which a serpent (who recovers its youth fettering168 that modality of the yogic or alchemical subject which has a every time it sheds its skin) would once have been the custodian. The tendency to volatilize; by so doing, the wholly integrated subject may enjoy counterpart-or more exactly the archenemy-of this bird who flies the power of controlled flight, rather than disintegrating, i.e., losing parts through the ether of the heavens to perch atop the tree of life, to name but of itself that would otherwise fly off. This is the difference between khecara, one tree, is the serpent who lives at the foot of the same tree. This theme "moving in the ether," and hamsaga, being "gone like a goose." Here again, is, moreover, the starting point for a multitude of Indian variants on the the original inspiration appears to be Upanisadic. Referring to the soul, epic myth of the bird-god Garuda's battles with the serpent descendants of the Kāthaka Upanisad (2.21) states: "seated, he travels afar, and recumbent Kadru.174 So it is that we find in the alchemical appropriation of certain goes everywhere." perennial themes from Hindu mythology, as well as metaphysical catego- As we have shown, vyoma, as both mica and ether, is identified with the ries from Sämkhya, a return to one of the most ancient "creative opposi- Goddess's sexual emission. Mercury, when it is identified with the soul tions" known to humanity. (jīva), when it transmutes through speech of which the substrate is ether Indeed, there is much of the serpent in Hindu imagery of death and (śabda-vedha), and when it becomes possessed of the power of flight (khecar- eternal life. As early as the Brbadāranyaka Upanisad (4.4.7), the immortal a[tā]), is itself associated with ether.169 In this way, both mercury and mica, self is said to discard its mortal body "like a snake sheds its slough." The the sexual emissions of Siva and the Goddess, located at the summit of the slough that a snake periodically sheds is an important image in alchemy as alchemical hierarchy, are likened to the ethereal goose. well. Here, mercury in its unrefined state is said to be covered with seven Siva's seed is further associated with another sort of bird-as well as "sloughs" or "sheaths" (kañcukas). These are coatings of various base met- with a serpent-in Somadeva's account of the origin of mercury, which als and other impurities, which float on the surface of mercury (whose spe- varies significantly from the chronologically later RRS account presented cific gravity is greater than that of all the metals, save gold), and which are earlier in this chapter.170 Like the RRS, the RC states that Siva first cast the said to be the creations of jealous gods who feared that mercury would

214 215 Chapter Seven Corresponding Hierarchies

render human alchemists their equals. Mercury, the semen of absolute place of the feminine in the alternation of the divine between self- godhead, must be purged of these coatings if it is to be used with success concealment and self-discovery, may be elicited here. In describing the in alchemical operations.175 mercury produced from Siva's semen, the BhP states that this next com- No doubt working from the Upanisadic image evoked above, a number bined with the substance emitted from the Goddess's sexual organ to pro- of Saiva schools and sects came to apply the notion of the kañcukas to a duce four colors and two types of mercury. Mercury, the mineral form of much broader range of ideas. In a number of these systems, a group of five the body of the wholly transcendent Siva, is mediated in the world by the kañcukas ("sheaths") become an important metaphysical construct em- Goddess, whose interposition, in the form of her sexual emission, renders ployed to explain the difference between God in his transcendence and said mercury impure.178 God in his immanence, between essence and existence, the unity of the In alchemy, the image of the serpent is not limited to the mercurial kañ- divine, infinite self and a plurality of finite selves. Such is especially the cukas. It is also widely employed as a signifier for the lower end of a system case in the orthodox Saiva Siddhanta, according to whose Kiranāgama that is crowned by an ethereal bird: the lowest metal on the alchem- creatures are provided with the five kañcukas while in the "bosom of Māya," ical hierarchy is lead, most commonly called nāga, "serpent" or sīsa[ka], an the material cause of the impure worlds. These impure worlds are said to allomorph of the name of the cosmic serpent Sesa; or, more rarely, abirāja, owe their existence to Ananta (or Anantesa) the first of the eight Vidyeśv- "serpent king."179 The Rasakāmadhenu (2.1.4) and Rasendra Bhāskara aras. Now, Ananta is one of the names of the Hindu cosmic serpent, whose (4.108) state that lead arose from the semen of Väsuki, the king of a mythic relationship to the kañcukas is an obvious and immediate one. This being race of serpents known for the great wealth it possessed in its subterranean the case, the intermediate role of the Vidyesvaras appears to double that of treasure hoards. This bird-serpent opposition is made most explicit in the the kañcukas themselves: they constitute the gray area between divinity and ca. sixteenth century Rasakāmadhenu, which identifies gold, at the summit humanity, essence and manifestation.176 of the hierarchy of metals, with semen shed by Agni, and lead, at the base Abhinavagupta, in his discussion of the untraceable boundary between of the system, with the seed of Vāsuki. Let us also recall here the Rāmāyana Siva and his creation, may well have had the mercurial kañcukas in mind myth, related at the beginning of this chapter, in which lead and tin (often when he wrote the following passage, so close are the conceptual parallels used interchangeably) are said to arise from the residue (mala) or afterbirth between the chemistry of mercury and the metaphysics of gnoseology of the generation of gold.180 here: This further recalls the serpent Sesa ("Remains," "Residue")181 who up- The God, whose nature is a free consciousness, whose characteristic holds the golden egg of the universe on his many hoods and whose coiled is the supreme light, due to his own intrinsic nature and as a result of body is composed of the dregs, the calcinated residue of past creations. Of what sort of residue is Sesa composed? According to the Puranas, the cos- his enjoyment of the sport of concealing his own nature, becomes the mic dissolution or reabsorption (pralaya) that occurs at the end of a great atomic, finite self, of which there are many. He himself, as a result of his own freedom, binds himself here by means of actions whose na- age (mahāyuga) of some 4,320,000 human years is a two-phase process. The first of these is a universal conflagration, in which Siva, in his destruc- ture are composed of imagined differentiations. Such is the power of tive Kālāgnirudra form, 182 incinerates all the gross, inert matter located the God's freedom that, even though he has become the finite self, he once more truly attains his own true form in all its purity. inside the cosmic egg (while preserving the subtle souls of liberated beings in the ether of the highest levels of the cosmic egg, well above the confla- Here, the dividing line between the impure path (asuddhadhvan: the gration). Then follows a great rain and flood, the true dis-solution, which twenty-five Samkhyan tattvas of the manifest world) and the pure path extinguishes the fire and immerses the world in a great ocean.183 There (śuddhādhvan: the five transcendent categories of the Trika Kaula) is com- remains a calcinated or ashen residue from the fire, however, which sinks posed of the intervening six kañcukas. Taken together, these are the thirty- to the bottom of the ocean of dissolution, to coalesce into the serpent Sesa, six (25 + 6 + 5) categories of Kaula metaphysics.177 at the bottom of the cosmic egg. A final alchemical parallel, which also brings us back to the important At the end of a cosmic eon (kalpa), of one thousand mabāyugas in length,

216 217 Chapter Seven Corresponding Hierarchies

the entire cosmic egg-and not merely the lower levels of its interior-is Just as "serpentine" lead, at the base of the alchemical hierarchy of met- consumed, and it is upon the chaotic waters of the ensuing flood that the mabāyogin Visņu Nārāyana, the "Abode of Mankind," sleeps, in a state of als, must, of teleological necessity, fully actualize its potential by ultimately

total yogic integration (samādbi) for a kalpic night.184 (Here, Visnu's yogic becoming alchemical gold, so too the sleeping kundalinī must awaken to remount the hierarchy of emanated being and reunite with the absolute in sleep is the divine model for the samädbi of the human yogin who has with- drawn all breath, seed, and consciousness to concentrate these into a single the "place of the ether," the human cranial vault. The temporal endpoints of these processes-of the ripening of metals and the yogic transformation point of pure being-consciousness-bliss. In Visnu's case, all that is subtle of the subtle body-are spatially located within the hierarchized systems and eternal in the universe has been withdrawn into that single point.)185 What of the gross matter of the previous kalpa? It too is preserved, once for which they serve as fundaments. So it is that we find, in the coils of these symbolic serpents, the nexus of a Hindu time-space continuum. again in the ever-recyclable serpent body of Sesa, who serves, this time, as With this, we turn to the structure of these parallel and interpenetrating Visnu's couch, upon which he sleeps away the kalpic night in yogic enstasis. In this way, Sesa, the cosmic serpent who ever renews himself from one systems.

cycle to another, remains the same serpent even as he is reconstituted from a new mixture of recycled elements. As such, his body is an endless source of raw material for renewed creations. For this reason, he is also called Ananta, "Endless." 186 The serpent at the base of a system is a commonplace of Hindu cosmol- ogy and metaphysics. In both sacred and profane constructions in India, a spike is symbolically driven into the head of this serpent, a head that upholds the world, to ensure a solid foundation.187 Elsewhere, the semileg- endary figures who laid down the foundations of Indian medicine (Caraka), yoga (Patañjali), and alchemy (Nāgārjuna) are all said to be incarnations of great serpents.188 Similarly, the ashes that compose the body of the serpent Sesa also bear a Vedic pedigree. In the logic of sacrifice, there can be no "first sacrifice"; rather, every "new" sacrifice, ignited with the embers of the sacrifice which preceded it, is a reenactment of that prior sacrifice: this holds as well for the primal sacrifice of the Purusa-recounted in Rg Veda 10.90-in which we find that the "primal" sacrifice out of which the universe was created itself arose out of preexisting sacrificial materials! Ashes of prior sacrifices are the seeds of future sacrifices, and so it is that the erotic ascetic Siva can call the ashes of cosmic dissolution that he smears upon his body his "seed."189 Similarly, when the Vedic sacrificer used a fire drill and block to create a spark with which to ignite a sacrificial fire, it was presumed that the fire produced was latent in the wood of the fire block. In the language of batha yoga, the kundalinī serpent is described in the same way: until she is awoken, she is fire that is latent in the fire block; once awoken, she is the spark that bursts into manifestation through the "churning" of yogic practice.190

219 Homologous Structures of the Alchemical Body

EIGHT seed.7 She is twofold, and it is in this perspective that yogic sources speak of this internal female serpent by another name: she is bhogavati, a term that at once bespeaks her enjoyment (bhoga, from bhuj, "partake, enjoy"), Homologous Structures of the Alchemical Body her coiled form (bhoga from bhuj, "coil, curl"), and her female sex (-vatī is a feminine ending). As bhogavatī, she is the serpentine female principle within the subtle body.8 It is when her name is interpreted in terms of bhoga as pleasure that this female serpent's twofold role is brought to the fore in the tantric context. The kundalinī as bhogavatī is a female who both takes pleasure and gives pleasure. In tantric metaphysics, it is the kundalini's coiled body itself that I. The Two Kundalinīs is the turning point between emanation and participation, emission and resorption.9 At ground zero of the self-emission of the absolute into phe- Nāgārjuna's name, in its Tibetan inscription (klu sgrubs), evokes terms that nomenal being is the kundalinī who takes pleasure as she allows the micro- signify, in that language, "confused mass," "a mass in the spiral form of a cosmic life force to drain away into her sleeping mouth. Her sleep is the conch shell," "abyss," "abdominal cavity" (klon), and "valley floor" (klun). sleep of dumb matter, and her head or mouth the sole obstruction to the These associations, of a founding alchemical wizard, a serpent, the abdom- opening of the upward path to her own return, from existence to essence inal cavity, and the base of an enclosed environment, evoke the perennial as it were. Her awakening blows open the "door to the absolute" (brah- hathyogic image of yet another serpent. This is the female kundalini, who madvara) at the base of the medial channel, and her rise along the length is explicitly identified with the serpent at the base of the cosmic egg, Sesa of this channel effects the return-on the part of the yogin in whose body or, alternatively, Ananta, "the endless one."2 she has been awakened-to the wholeness, being-consciousness-bliss, and In every human body, the female kundalini serpent sleeps coiled in the absolute godhead that is his true nature. Here too, we can glimpse the kuņ- place of the "fire of time" (kalagni),3 with her mouth closed over an internal dalini's role as the microcosmic homologue of Sesa, the cosmic serpent and linga in the lower abdomen.+ It is only in the body of a yogin that she is endpoint of a prior creation and the starting point of a future creation, ever awakened, and her awakening corresponds precisely to the initiation who is located as well at the base of a self-enclosed system. of the yogin's progressive withdrawal into total yogic integration (samādbi) We can also see, however, why the kundalinī is, unlike Sesa, a female or fluid equilibrium (samarasa). On a more concrete level, it is the rise of serpent. When she slumbers, "as if stupefied by a poison," in a man's ab- the kundalinī that brings about the transmutation of raw semen into nectar domen, the kundalinī is identified with human mortality, with death-laden in the cranial vault, a locus associated with the ethereal goose. In the last existence, and the bondage of the ignorant,10 which is figured by the in- chapter, I evoked the identification of the hamsa of the cranial vault with cessant drain of semen that she, as woman, effects in man. In this role, the the subtlest channel of the subtle body, the medial susumņā nādi. Like the sleeping kundalinī is identified with the fire of time (kalāgni) because the subtlest of minerals, mercury, the susumna can also double as both a bird mortal who allows her to drain away his semen is doomed to be consumed and a serpent:5 thus this channel is identified, as well, with the kundalini by the fire of time and die. Such is the fate of a number of yogins of Nath legend. Most famous among these is Matsyendranath, who, having had his serpent who, when awakened, pierces its opening at the base of the spinal column "like a key in the lock of a door-panel," to rush upwards to the yogic energy and life force drained away by years of debauchery in a King- dom of Women, is fated to die within three days if Gorakhnath cannot cranial vault.6 The kundalini in the body of the yogin is an incarnation of the feminine reawaken him, i.e., awaken his sleeping kundalinī. in this tradition and thereby incarnates all the perils and joys that women It is when the kundalini is awakened that she becomes capable of giving pleasure-and here too, it is appropriate that she be figured as a female can represent for men. She is divine energy (śakti) and female materiality (prakrti), but she is also a tigress who can drain a man of all his energy and serpent. Her awakening is the beginning of the return or resorption, on a

218

220 221 Chapter Eight Homologous Structures of the Alcbemical Body

microcosmic level, of the phenomenal world back into the absolute from straightens itself out like a stick"; and that the mūla bandha "awakens the which it was originally emitted. The yogin, in rousing her from her slum- kundalinī, who straightens, entering the medial channel, like a snake enter- ber, finds in the rising kundalini a vehicle by which to raise himself from ing a hole."1+ The kundalini's awakening marks the beginning of the yogin's mundane existence to the god-consciousness that renders him a second own withdrawal into his yogic sleep or trance, into the total integration Šiva. that is samadhi. In this light, we can see that in the universal scheme of Indeed, the final aim of the yogin, like that of any tantric practitioner, things, the great yogin, be he named Siva or Visnu, ultimately "awakens"- is twofold." On the one hand, he strives toward liberation (moksa, mukti) pours himself out into mundane being of which the sleeping kundalinī is from conditioned mortal existence; at the same time, however, he also the end- or turning-point-in order that human yogins might find a way seeks to realize for himself the enjoyment (bhoga, bhukti) that the absolute, to genuinely "fall asleep," i.e., enter into the yogic sleep of samadbi. On the Siva, knows in his very being.12 This pleasure that Siva, the absolute, takes microcosmic level, this yogic reintegration affords liberation and bliss; on in his manifold creations, his self-realizations, is often cast as his sexual the macrocosmic level, it is nothing other than the pralaya, the universal union with his female, manifest aspect, i.e., with the Goddess, who is iden- resorption of all mundane existence into the primal and primordial essence tified with vimarsa (reflective consciousness), māyā (creativity), śakti (en- that is the Absolute, God. ergy), and the kundalini. It is a pleasure of the same order as that enjoyed These two poles of the kundalini's mode of being-sleeping and waking, by Siva in his union with the Goddess that the yogin comes to know in taking and giving pleasure, allowing the body to be consumed by the fire awakening and raising his kundalini. It is in this light that the kundalini of time and consuming the fire of time-these mundane and transcendent becomes identified, in tantric practice, with both the male practitioner's poles are identified as her "poison" and her "nectar."15 The kundalinī is sexual partner and a goddess to whom he makes sacrificial offerings. As the poison when she remains asleep in the lower abdomen; she is nectar pre- MBbT explains, the practitioner's bhoga takes the form of the pleasure he cisely when she rises up through the medial channel of the subtle body to enjoys in his imbibing the makaras that precede sexual intercourse with his reunite with Siva, the Absolute, in the yogin's cranial vault.16 In the hatha- partner. The wine that he drinks and the flesh and fish he eats become yogic sources, this union is in fact accompanied by an outpouring of nectar, offerings into the mouth of the kundalini, who rises up to his tongue to which renders the yogin immortal. consume them. His bhoga, when accompanied by the proper mantras, This connection between poison, nectar, and yogic prowess has its re- becomes an inner sacrifice to the Goddess, with whom he thus shares his flection in Hindu mythology, in which the serpent king Vāsuki serves as pleasure. Like his human partner, she becomes his lover, and because she churning rope in the mythic churning of the Ocean of Milk. This churning is the Goddess, his savior as well. The bhoga offered into her mouth be- (which produced sulfur in the alchemical myth related in the last chapter) comes nectar, and her rise transforms his very being.13 Riding the kundalini is best known for its production of the nectar (amrta) of divine immortality. upwards on a wave of mutual pleasure, against the grain of the entropic Another byproduct of this process is, however, the production of the most processes of aging and death, the yogin comes to experience liberation and virulent of poisons in the universe (kālakuta, halāhala), which arose from bliss. Ultimately, he becomes "a second Siva," in eternal embrace with the the Vasuki's venom.'7 This poison, churned out of the ocean, would have Goddess. consumed the entire universe had Siva, whose yogic austerities rendered How does the yogin awaken the slumbering kundalini to reverse the him powerful enough to do so, not swallowed it. It is the trace of this feat order of nature on a microcosmic level? By assuming a number of postures that accounts for the dark blue mark found on Siva's throat in Hindu (āsanas), by breath control (pranayāma), and by means of a number of in- iconography, and his epithet of "Blue-Throat," Nilakantha.18 Lastly, mer- ternal blocks (bandhas) and seals (mudrās) that fan the fire of yoga (yogāgni), cury, Siva's semen, is a poison for the uninitiated who would presume to the fire that consumes the fire of time (whence its synonyms kālāgnirudra partake of it without the proper preparations; for the initiated alchemist, and yamantaka). So, for example the HYP states that the "Matsyendra āsana however, it is the nectar of immortality. This coincidence of opposites is .. arouses the kundalini"; that the mabā-mudra causes the "kundalinī [to] divinized in the Tibetan Buddhist homologue of Yamäntaka-that is, of suddenly become straight just as a coiled snake when struck by a rod Siva as "the Death of Death." This is the Tibetan divinity named "Quick-

222 223

Chapter Eight Homologous Structures of the Alchemical Body

silver" (i.e., mercury) who, although he is a "black poison-faced divinity," the mastery of the forces that sleep in the lower part of the body, forces

is also the "black master of life."19 that, when awakened, transform the yogin's being completely. The twofold kundalini, who incarnates the two prevailing Hindu atti- A moment ago, I evoked Matsyendranäth, the original guru of the Nth tudes toward the manifest world, a world which it always views as femi- Siddhas, who was rescued from death in a Kingdom of Women by his more nine-i.e., the Vedantic view of the world as imprisoning maya and the disciplined pupil Gorakhnäth. It is on this mythic ground that Matsyendra tantric view of the Goddess as a limitless source of energy, sakti-is the guru becomes Matsyendra the disciple: after Gorakh has rescued Mat- but one of a wide variety of redoubled images proper to the yogic and syendra from the Kingdom of Women, he must give his teacher a "re- alchemical body. This dual role of feminine energy-of the Goddess em- fresher course" in hatha yoga. Indeed, it is a leitmotif of Nath legend that bodied in the world, in a human woman, or in the internal kundalini- Matsyendra, the teacher of Gorakh, is much more susceptible to the lures is clearly expressed in a verse from the Spanda Kārikā of Abhinava- of the fair sex, to the world of appearances (maya), than is his pupil, who gupta's ninth-century forerunner, the philosopher Vasugupta: "It is Siva's must constantly awaken him to the reality behind appearances. To be sure, śakti, that is, his power to act, who, dwelling within limited creatures, the erotic rituals of the Yogini Kaula sect of which Matsyendra was the causes bondage; When she is known as herself the path, she is the one who purported founder lie behind much of this mythic imagery. So it is that the makes perfection (siddhi) possible."20 In the balance of this chapter, I ex- yoginīs of Kāmarūpa, from whom Matsyendra received his tantric rev- amine a number of such phenomena in these interpenetrating symbol elation, become the "Plantain Forest" women who ensnare him, in an im- systems, in images that range from human mouths to wells to rivers to portant myth cycle of Gorakh and Matsyendra.24 forests, cities, and even entire kingdoms! Throughout, it will be shown In a curious way, Matsyendra is doubly connected, by his name, to the that each image at once points to the imminence of death, the great equal- lower half of the yogic body, the place of the sleeping kundalini. However, izer, and to its opposite-immortality, the cheating of death (kāla- just as the kundalini sleeps at a turning point in the play of divine mani- vañcana)-on the part of the Siddha who knows how to reverse, on the festation and resorption, so too, the symbol system constructed around mesocosmic level of his subtle body or alchemical laboratory, the being- Matsyendra's name is an ambivalent one. An extensive body of medieval towards-death that is the erstwhile fate of all creatures who are bound to sources connect Matsyendra to his doctrine of the fish belly through a

this world. mythic gloss of his name. Matsyendranatha means "He Whose Lord is the Lord of Fishes" (or the pleonastic "Lord Lord of Fishes"), and his connec- tion with fish is explained through a myth in which the Goddess, seated on 2. Charting the Subtle Body: The Legends of Matsyendranath or near the shore of Candradvīpa ("Moon Island"), has asked Siva to teach her the most secret of all esoteric knowledge, things he has never told her before.25 This Siva agrees to do, but he has barely launched into what will a. Fish Belly turn out to be the essence of the Hindu tantric teachings than does the Like the supreme yogic god Siva, the hathayogin is himself capable- Goddess fall asleep. Siva's words do not, however, go unheeded. Matsy- through his raising of the kundalini-of transmuting poisons into nectar. endra, who has been swallowed by a fish (whence his name in these More properly speaking, that which is poison for mere mortals is, for the sources), draws up to the shoreline and overhears everything. It is in this yogin who has realized, through his practice, a divine "identity in differ- way that the "historical" Matsyendra26 becomes both the link between the ence" view of reality, identical to nectar, amrta.21 Numerous Nath Siddhas divine and human in the transmission of the tantras and the founder of are known for their ability to control (and charm) serpents, yet another numerous sectarian traditions.27 In a number of these sources, Matsyendra metaphor for their mastery of the female kundalini, and for their ability is further identified with, or called the father (or son or brother or nephew to treat poisons as elixirs.22 Likewise, mercury, which is a poison for the or daughter) of, Minanatha, a name which also means "Lord of Fishes."28 uninitiated, becomes an elixir of immortality for the alchemist who knows The earliest mythic account of Matsyendra, that found in the K7nN its secrets.23 In this section, I interpret a number of yogic metaphors for (16.27-56) itself, tells a somewhat different story. This source, which calls

224 225 Chapter Eight Homologous Structures of the Alchemical Body him Macchaghna, "Killer of Fishes"29-an indication that he was a fish- (1.7) of his monumental Tantrāloka, cryptically refers to matsyodara in a erman-makes no reference to his overhearing Siva's teachings from the later passage of the same work: "On the level of the highest kundalin is the belly of a fish. Rather, it relates that Siva has committed the kulāgama to Emissional Power which is beautiful because it contains within itself the writing, the which his son Kärttikeya has thrown, in a rage, into the sea. vibration, there the yogin should repose devoted to the condition of belly of These teachings are then swallowed by a great fish. The god Bhairava takes the fish."38 the form of Matsyendra the fisherman to recover the teachings out of the What this term means is stated in elliptic fashion in the thirteenth- fish's belly, at a site called Candradvīpa.30 Later, it is stated that this teach- century Amaraugha Prabodha of Gorakhanath, who further intimates that ing (sāstra) was brought down at Candradvīpa;31 earlier in the text, how- this doctrine constituted the very first revealed teachings of Siva, which ever, it is stated that it was brought down by Siva and the Goddess, at that god gave to an audience consisting of Minanath and the Goddess, i.e Kāmarūpa.32 on the shore of Candradvīpa. Here, Gorakhnath states: "Holding the As a killer of fishes and the founding guru of a tantric lineage and tradi- breath [when it is] restrained by force (hathat) is [called] swallowing into ion, Matsyendra may be further linked with another tantric founder. This the fish-belly (minodare). He [who is] blessed with detachment is not con- is the ca. ninth-century Buddhist Siddhācārya Luī-pā (or Lo-yi-pā, Lū- scious until he releases it [i.e., his breath]."39 What I wish to argue here is yi-pā, Lū-hi-pāda, Lū-yi-shabs). The original Siddhācārya of the Tibetan that Matsyendra's doctrine of the fish belly-a revolutionary doctrine in- Grub thob33 lists and author of two Bengali carya songs (nos. I and 29), Luï- deed, given the perennial importance that has been attached to it in Hindu pā's name means "Venerable Red-Fish." In Nepal, Luī-pa appears to be tantrism over the past one thousand years-is about diaphragmatic reten- further conflated with a Buddhist divinity. Here, a famous red image of the tion and its effects on the body and consciousness of the yogic practitioner. bodhisattva Avalokiteśvara has been worshipped in the village of Bunga This meaning is greatly expanded in the unexpected context of a de- and the city of Patan since the seventh century. Since the sixteenth century, scription of the holy city of Benares given in a number of Saiva Purānas. this same image has been identified with Rato Macchandernath, the "red As any armchair traveler knows, the city of Benares is situated on a wide Matsyendranāth"; and, indeed, an old Newari bhajan, still sung in Kath- bend in the holy Ganges River, at the last point at which the river turns mandu, refers to this deity as "Luïpada-nāth."3 While scholars generally north, as if to return to its Himalayan source. It is also a city that is said agree that this is a late and spurious identification, there are a number of in certain sources to resemble a fish in shape.+0 Though not because of its piscine elements to the names and roles of this Buddhist divinity that de- piscine outline, the beauty and holiness of Benares are such that Siva, in a serve further attention. First, "Lui" is a vernacularization of lobita/robita, distant mythic past, chose to make Benares, or Kasī, his home whenever i.e., "red": Lui-pa's fish is a red fish, which squares with the color of the he was not meditating atop Mount Kailash.+ For Hindu pilgrims from Patan image.35 Second, Avalokiteśvara has, in at least one case, been repre- time immemorial, it is not the beauty of the site, but rather the liberating sented in a piscine mode: at Ratnagiri (Cuttack district, Orissa), the bust effects of bathing in the waters of the Ganges there-the salvific powers of of a mutilated Avalokiteśvara image was found to be superimposed upon which are enhanced by Siva's presence-that has been the prime reason the image of a large fish, "so as to convert the Bodhisattva into Matsya, for undertaking a journey to that fabled city. the first Avatara of Visnu." The Hindu Matsyendranath has himself been The Ganges (Ganga) is, however, but one of a triad of holy rivers-and portrayed in a manner similar to that of Visnu's fish incarnation, emerging of river goddesses-its two sister rivers being the Yamuna (which joins the out of the mouth of a fish.36 Ganges at Prayãg, the modern Allahabad), and the Sarasvati. There is a Of much greater interest to us is the identification we find of Matsy- mystery to this last river, the earliest river to be identified with a goddess endra or Lui-pa's name with the term matsyodara, which simply means "fish in the Vedic literature: rising in the highlands of southwestern Rajasthan, belly." Tibetan translations of Luï-pa's name, found in the Tanjur, include it is a river that "disappears" into the plains of Kacch before ever reaching na'i-rgyu-ma-za-ba, which means "one who eats the intestines of a fish"; the sea-and it is this hidden nature of the Sarasvati that is brought to the and na-lto-pa, which restored into Sanskrit yields, precisely, matsyodara, fore in the Puranic descriptions of matsyodari-yoga, the "fish-belly conjunc- "fish belly."37 Abhinavagupta, who praises Macchanda in an opening verse tion" of Benares.

226 227 Chapter Eight Homologous Structures of the Alchemical Body

Before we can explain this phenomenon, however, we must first briefly review a commonplace of yogic physiology, according to which the three "itself." The waters of the Ganges that were so channeled therefore ran

principal subtle channels (nadīs) of the life force are identified with these against the river's normal flow, along the underbelly of the piscine outline

three major rivers (nadīs): the solar pingala is the Yamuna River, the lunar of Benares-within which the permanent Matsyodarī Lake was located-

ida is the Ganges, and the medial, "empty" susumna is the "fiery," "hidden" whence the designations of matsyodari for the periodic stream itself, and

Sarasvati.42 In the practice of breath control, prānāyama, it is by pumping matsyodarī-yoga, the "fish-belly conjunction," for those periods in which it

up and thereby emptying the two peripheral subtle channels, the ida and appeared (and for the site at which the overflow of Matsyodarī Lake ran

the pingala, that the theretofore empty susumna suddenly becomes filled into the waters of the flooding Ganges).

with the subtle breath or life force (prana) to become the yogin's internal This anomaly, of a periodic backflooding of the Ganges, is given a yogic

upward-tending channel to liberation. This yogic homologue to these gloss in the ca. tenth-century Linga Purana, a gloss that will at once explain

three river systems of India is so obvious as to not require explanation here. the great yogin Matsyendranāth's piscine associations:

Benares, the greatest pilgrimage site (tīrtha) of India, is said to contain all of the rivers and all of the tirthas of India within its precincts. It is in Pingalā is the name by which the fiery nadi is praised, and it is known

this context that the geography of the city, as well as seasonal floods there, to be that dried up stream (the Asi) [which runs by the place] where

contributed to the unusual riverine phenomenon known as the "fish-belly [the tank named] "the trembling sun"46 stands. Ida is the name by

conjunction," down to at least the sixteenth century. Prior to its urbaniza- which the lunar nādi is praised. It is known to be the Varanā [River],

tion over the past two centuries, the inner limits of riverine Benares were where the [Visnu] Keśava [temple] is situated. That nādī [which runs]

constituted by a string of inland pools, tanks, and lakes. One of these was between the two is glorified as the susumņa and known as Matsyodarī;

named Matsyodarī, the "[Lake of the] Fish Belly." In particularly heavy it is praised as "[a stream] running in both directions" (visuvam).47

rainy seasons, these reservoirs would flood into one another to form a There where [two currents] come together at the Fish-Belly Pool, a

single channel-also called the Matsyodarī-that ran for no less than three bath yields liberation ... It is an extraordinary conjunction when the

miles between the river known as the Varana, the tributary to the Ganges Ganges runs into the Fish-Belly channel to the west of Kapileśvara.

forming the city's northern border, and the stream known as the Asi, which A bath at that juncture yields [the fruits of] a thousand horse sacri-

flowed into the Ganges at the southern end of the city.43 According to folk fices ... That place is indeed praised as the fluvial brabma-syllable.48

etymology, the classic name of Benares-Vārānasī-derives from the fact that that city-state lay between the Varana and Asi rivers. For certain car- In a passage from the somewhat later Kasi Khanda of the Skanda Pur-

tographers of the subtle body, the symbolism was clear :* "The two vessels ana, the Matsyodarī is said to be babir-antaścarā, "flowing both inside and

called the ida and the pingala are the real varana-asi. The space between outside" (i.e., surrounding Benares, via the Ganges' normal outer channel

them is called Varanasi. There it is said that [Siva] Visvanath dwells ... as well as that of the inner Matsyodari), and taking the path of retraction

From the right side of the ajna [cakra] and going to the left nostril flows (sambāra-marga).49 With this, we now hold all the necessary elements for

the ida. It is here called Varana, the upward-flowing [Ganges] ... Rising a symbolic interpretation of the term "fish-belly conjunction," matsyodarī-

from the left side of the ajna lotus45 and going to the right nostril, this yoga. The belly of the fish, and that "path of retraction" at which a reversal

upward-flowing pingala has been called of yore the Asi." of the normal flow occurs, is, assuredly, the opening of the medial channel,

This schematization flows directly into the rainy-season phenomenon I the susumna nadi50 at the base of the subtle body. The two peripheral chan-

have been describing: when the seasonal Matsyodarī opened its periodic nels (the idā and pingalā) are the Varaņā (Ganges) and Asi (Yamuna) rivers,

channel, the normal flow of Benares's rivers became reversed: the Ganges whose flow is reversed such that the "empty" medial channel, the susumņā (the Sarasvati River; and in Benares, the Matsyodari), may be opened and would actually back up into its tributary, the Varana, and flow therefrom filled with the reversed, upward flow of semen, life force, breath, and mind, into the seasonal Matsyodarī channel; the Matsyodarī would in turn drain into the Asi, out of the mouth of which the Ganges would empty back into which yields liberation and immortality. In this schema, the backward- flowing Ganges itself doubles as the "disappearing" Sarasvati River, which

228 229 Chapter Eight Homologous Structures of the Alchemical Body

"reappears" in a feat of yogic reversal when it becomes the Matsyodarī, the and well-documented. As June McDaniel has demonstrated, the Bāuls fish-belly channel.51 Matsyodarī-yoga, the "fish-belly conjunction," is, however, more than a portray

mere macrocosmic reproduction of a structure of the subtle body. It also Isvara or the man of the heart [as] a fish (min-rūpa). This fish [which] refers to a particular yogic practice, specifically to that practice to which swims in the "high tide" of [a woman's] menstrual flow ... is caught Gorakhnath alludes in the Amaraugha Prabodha passage quoted above. by the enjoyer (rasika), the Baul who is full of love. This is the "tide "Holding the breath when it is restrained by force (hathat) ... in the fish time in the river," the overflowing of rasa [sexual fluids]. He catches belly (minodare)" is a description of that most fundamental aspect of breath the fish and causes it to move in an upward direction. The fish must control called kumbbaka, the "potlike" diaphragmatic retention of the be caught at the right time, or the waters dry and he is gone ... The breaths.52 Here, the ida and pingala channels are "pumped like bellows," by right time is once a month, a Mahayoga. It includes the new moon, means of breath control, until the susumna channel suddenly opens. At this the period regarded by the Bauls as the time before the fish emerges; point, the two peripheral channels empty and lie "swooned." What follows this is the time of lust (kama). When the fish is present, it is the time is automatic: the vital breath that rushes into the susumna rises upward, of the menstrual flow designated as full moon. The third day [of a against the normal downward flow of all bodily fluids, affords yogic libera- woman's menses] is the time to catch the fish; he is not present before tion.53 In Puranic mythology, it is precisely by bathing during the period that, and there is the danger of the black crocodile of desire.58 of the "fish-belly conjunction" that the god Bhairava was liberated from the skull of Brahma that had clung to his hand for twelve years. Having In this context, we may read matsyodarī-yoga as a temporal conjunction

thus completed his "skull bearer's vow" (kāpālika vrata), Bhairava estab- in the menstrual cycle of a yogini. Matsyendra the fisherman catches his

lished himself as the guardian deity of Benares, at a temple on the bank of "fish" of the Kaula revelation at that precise moment and thereby becomes

the Kapalamocana ("Liberation from the Skull") tank.54 This god's mythic the revealer of the Yoginī Kaula.

liberation from his terrible vow is tantamount to the wholly unconditioned state of Sivahood enjoyed by the yogin at the culmination of his practice b. Moon Island

of "swallowing into the fish belly." Matsyendra's doctrine of the fish belly would therefore have been, if With this, we exhaust the symbolism of the yogic fish belly; in so doing,

nothing else, a teaching on hatha yoga. This is supported by the fact that however, we bump up against another riddle of tantric parlance. Here, we

Gorakhnäth, in the opening verses of his twelfth- to thirteenth-century must consider that in the times of the fish-belly conjunction, the city of

Goraksa Śataka, invokes Mīnanāth as his guru.55 Now, while it has been Benares, surrounded by the Gangā, Varana, Matsyodarī, and Asi rivers, became an island.59 Here, we are reminded of Moon Island, off the shore maintained that six-cakra yoga was an innovation of the Western Transmis- sion, it would appear that Matsyendra's doctrine of the fish belly (kumbhaka of which Matsyendra, in the belly of a fish, overheard Siva's original tantric teachings. This island has never ceased to tantalize scholars who have as the culmination of breath control), together with a significant number of other references to the subtle body and hathayogic practice found in the attempted, without success, to localize the toponym Candra-dvīpa some-

KJnN, would make him the founding guru of this discipline, or at least of where on the India subcontinent60 and to explain its implicit identification,

one of its major canons.56 Abhinavagupta may also be singling Macchanda in a number of tantric sources, with Candra-giri ("Moon Hill") or Candra-

out for praise on these grounds;57 it is also possible, however, that the parvata ("Moon Mountain"). Here again, it may be argued, on the basis of data found in the tantric image of the fish belly also had a sexual connotation in Kaula erotico- texts themselves, that Moon Mountain and Moon Island (for what is an mystical practice. island if not a mountain with its feet in the water?)6t are, like the Fish Belly, Such is certainly the case in the religious belief and practice of the Bauls, a Bengal-based syncretic sect whose ties to the Näth Siddhas are numerous locations that appear within the subtle body when a certain body of prac- tice is carried to its conclusion. That this is an inner landscape of the subtle

230 231 Chapter Eight Homologous Structures of the Alchemical Body

body is already borne out by the KJnN's statement that, by using the net by a passage from the Cincinīmatasarasamucchaya, a Western Transmission of yogic energy (sakti-jalam), Matsyendra the fisherman pulled his fish text which identifies the founding Siddhas Minanatha with the moon, (containing the kula teachings) out of the seven oceans (i.e., the seven Mesapada with the sun, and Kurmanatha with fire.69 cakras),62 an impossible act in the "real world." In the later works of Gorakhnäth and other synthesizers of the hatha- In this case, the mountain or island in question is located either in the yogic gnosis, the identification of the cranial vault as the place of the mi- cranial vault or the abdominal cavity, but in either case on the left side of crocosmic moon becomes a commonplace. It is moreover in a work attrib- the body, i.e., that half of the body through which the lunar idā nādī passes. uted to Gorakhnāth, the Amaraughasasana, that the left side of the head, or The K7nN itself, which maintains in its chapter colophons63 that its teach- the cavity of the left nostril, is explicitly identified with the circle of the ings were brought down (avatārita) at Moon Island, offers a certain body moon.70 We may imagine that the city of Benares-identified with the ājña of evidence to this effect when it describes the oozing of nectar, via the cakra of the cranial vault and theoretically situated atop three hilltops on brahmarandhra (the "cleft of bráhman," i.e., the fontanelle), as "gladdening the northern bank of the Ganges River, between the Varana and Asi riv- the moon" (candrabladakara) and associates the lunar posture (candrāsana) ers-could have been identified with both a mountain and an island, after with the hathayogic practice of khecari [mudra], also effected within the the fashion of Candragiri and Candradvīpa, by medieval cartographers of cranial vault.64 In addition, both this source and a number of other hatha- the subtle body. When, through the practice of diaphragmatic retention, yogic works of the Nath Siddhas and the Western Transmission place great the central susumna channel opened to serve as an upward conduit for the emphasis on the "western" or left side of the body as the locus of the most yogic life force, seed, and energy, a lunar site in the left side of the cranial critical transformations occurring within the subtle body.65 vault called Candragiri, "Moon Hill," would have flooded to become Can- The teachings of the KM were brought down, as well, at Moon Island; dradvīpa, "Moon Island," as in the case of Benares during the fish-belly however, this text places equal or greater emphasis on Moon Mountain or conjunction.71 Moon Hill, which, as indicated, likely refers to the same locus within the subtle body. This appears to be the meaning of a statement found in the first chapter of this work, which states that Moon Mountain is located to c. The Mare's Mouth the west of Meru, i.e., the subtle spinal column, which culminates in the This is not the sole possible gloss of this phenomenon, however. In addi- brahmarandbra at the summit of the cranial vault.66 tion to the microcosmic Moon Island/Moon Mountain located in the left This lunar location is found in other Western Transmission sources, side of the cranial vault, there is another site at which a similar phenome- which declare that the textual canon was "brought down"-by a figure var- non appears to occur: this is the inner linga covered by the mouth of the iously called Śrīnātha, Śrīkaņtha, Ādinātha, or Siddhanātha-at a lunar sleeping kundalini: located, according to certain sources, in the left side of location called Candragiri, Candradvīpa, Candrapurī, or Candrapītha.67 the abdominal cavity, this is referred to as the pascima-linga, the "western Once again, however, this location appears to be, at least to a certain ex- linga."72 In his fourteenth-century Chos-bbyung (History of Buddhism), the tent, an internal one. This is borne out by a passage from an A.D. 1395 Tibetan historian Bu-ston, citing an Indian source entitled the Nāthābhyu- manuscript of the Kubjikānityabnikatilaka, which maintains that Śrīnātha- daya Tantra, gives the following account: "A certain fisherman, having been with the aid of three Siddhas named Sun, Moon, and Fire-founded that swallowed by a fish, will die and be reborn as the yogin called Dārika. This tantric kula, "at the beginning of the Kali Yuga," at a site called Candrapurī one will cast wine into the river Ganges and in an hour an island called ("Moon City"), located in the western region of Konkana.68 In spite of Bdava will appear."73 The fisherman in the belly of a fish once again this purportedly geographical reference to Konkana, the most satisfactory evokes Matsyendra; moreover, in a Tibetan list of the eighty-four Mahā- interpretation of this account remains an allegorical one: Moon City is the siddhas, Dārika is listed as the disciple of Luī-pā who is, as noted, a pos- site, in the left (western) side of the subtle body, in which the hathayogic sible equivalent of Matsyendra.74 Dārika may therefore be symbolically praxis culminates, through the interaction of the three principle nādis fig- identified with Matsyendra himself or with his double, his "son" Minanath ured here as the Siddhas Sun, Moon, and Fire. This is further supported (especially since the Tibetan source says the fisherman dies and is reborn

232 233

Chapter Eight Homologous Structures of the Alchemical Body

as Darika). The mention of an island suddenly appearing out of the Ganges retraction (sambara) will occur when the fire of the kalagnirudra, the De- would also be an apparent reference to matsyodarī-yoga, during which times stroyer of the Fire of Time-a fire he identifies with the mouth of the Benares became an island. What, then, of the name Badava? submarine mare (badavāmukha)-flares upward. This cryptic utterance, The term badava, "that which arises from a mare (badava)," has a partic- made at the beginning of general discussion of "yogic retraction," precisely ular resonance in Hindu mythology or, more properly speaking, Hindu concerns the awakening of the fire of yoga in the base of the subtle body, eschatology. Badava is the name of the submarine fire-or, more precisely, an awakening that leads to total yogic resorption in the cranial vault, the of the fiery fluid composed of both fire and soma-which originated from microcosmic equivalent of the pralaya, the cosmic dissolution. The bādava Siva's third eye when he incinerated Kama, the Hindu Eros. When that is that fire, in potentia, that slumbers in every subtle body: when mastered fire, once released, threatened to engulf the entire universe, it was placed and channeled upwards, it affords liberation; when allowed to burn nor- by Śiva inside the mouth of a "submarine mare" at the bottom of a south- mally, it is the fire of time that cooks all beings to death.79 ern sea, whence it continues to belch flame down to the present day. One It will also be recalled that matsyodarī-yoga is at once a description of the day, when it will have burned away, evaporated, all the water in that ocean, yogic practice of diaphragmatic retention (kumbhaka),80 by which the nādīs it will incinerate the entire universe, reducing all to ashes.75 Bādava is peripheral to the medial susumnā, the channel of the rising kundalinī, are therefore the fire of the cosmic dissolution, in potentia, as well as a fire in emptied. This practice not only suddenly fills the susumnā with breath; it the mouth of a female creature. also fills this channel with semen, and it is here that Darika's island in the Which brings us back to the sleeping kundalini, who sleeps (like the Ganges suddenly comes into view. Earlier, I cited a passage from the Kāśī Goddess in the myths of Moon Island) in the left (western) side of the Khanda of the Skanda Purana that described the Matsyodarī as taking the subtle body. (The K7nN itself, while it mentions a goddess named Kundalī, "path of retraction" (samhāra-mārga),81 i.e., as draining the Ganges into a makes no explicit mention of the kundalini in this role.76 One may, how- channel that flowed in a direction opposite to its normal flow. Such is also ever, see, in its imagery of the badava-the fire of universal dissolution or the case with the subtle body: when the kundalini rises, she also siphons retraction (sambära) in the mouth of a submarine mare (badavā)-a pre- upwards the semen that had previously remained inert and subject to loss figuration of this more common, but later, image of this female serpent in the yogin's abdomen. In this light, we must evoke another yogic com- power.) Let us recall here that the kundalini, when she sleeps, is identified monplace: kalāgni, the fire of time, is identified with adhoretas, that is with with the kalagni, the fire of time that cooks all creatures to death, through "downward-tending semen," while kälägnirudra is identified with ürddh- the aging process. Let us also bear in mind that the fire of yoga (yogāgni) varetas, "upward-tending semen."82 that destroys the fire of time, is identified as kālāgnirudra, the "Rudra of What happens when the kundalini rises? A "column" of ambrosial se- the Fire of Time," that is, a fire which is greater than, which consumes, the men is raised, via the susumņā nādī, to flood the cranial vault. This is the fire of time.77 When she sleeps, the kundalinī is associated with the fire of meaning of the island called "Mare's Mouth" that suddenly appears in the time, a time whose passage is marked by the movements of sun and moon Ganges when Darika throws wine into its waters: when matsyodarī-yoga in the subtle body; when she awakens, sun and moon (here the ida and takes place and the normal downward current of bodily entropy is reversed, pingalā nādis) are immobilized, and the kundalinī, doubling as the susumnā the upward channeling of seed causes the theretofore submerged bādava nādī, is said to "consume time."78 fire to flare up from the base of the subtle body. This reading is supported Therefore, while the kundalini's sleep is associated with the fire of time, by a poem of the fifteenth-century mystic poet Kabīr, who instructs the the loss of the yogic fuel that is semen, and being-towards-death, her po- yogin to "reverse the [flow of the] Ganges, and dry up the ocean."83 When tential for reversing such a process likens her more to the mythic bādava applied to the sacred geography of Benares, the "path of retraction" taken fire at the bottom of the southern sea, a fire that, once it has consumed the by the Ganges during a fish-belly conjunction transforms the city into an waters of the ocean, will become the kalagnirudra, the fire that consumes island. When applied to the sacred geography of the KJnN and KM, the time. Once again, it is Matsyendra's KJnN that offers a primal and funda- landform of the subtle body that appears through this dynamic is known mental reading of this yogic process. Here, it is said that the universal as Moon Island or Moon Mountain.

234 235

Chapter Eight Homologous Structures of the Alchemical Body

Lastly, there is the matter of the wine Darika throws into the Ganges to titioner who can master himself in its presence. Whence the most seminal trigger this process. Here, we should note that the tantric Buddhist tradi- of all of the banīs of Gorakhnath, who adds an alchemical nuance to this tion of which Darika was an exemplar was a form of tantrism which em- sexual image: "Penis in the vulva's mouth, mercury in the mouth of fire, he phasized the symbolic use of the five makaras, the "antisacraments" of who can retain these [volatile substances of semen and mercury], him I call wine, fish, flesh, parched grain, and sexual intercourse. It now suffices to my guru."88 return to the MBbT passage cited above to understand the meaning of There is a certain irony to this verse, if we recall that in myth, Gorakh's Därika's gesture. The sacramental wine that the tantric practitioner im- guru Matsyendra was a teacher who was quite incapable of retaining his bibes becomes an offering into the mouth of the kundalini who, sharing seed in proximity to beautiful women, to yoginīs. This weakness of Mat- his enjoyment (bhoga), awakens to rise and thereby transform him into a syendranatha for the fair sex is the subject of a body of legend set in the second Śiva.84 land of Kämarūpa, which is itself identified, precisely, with the center of The badava, the fire in the mouth of the submarine mare, has a number the adbovaktra, the lower mouth or vulva of the yogini, but also with the of other usages in tantric and alchemical geography. A natural gas vent at center of a mystic triangle located at the base of the subtle body, in the the important sākta pītha of Kāmākhyā is called bādava kuņda, the "pool of same region as the kundalini.89 Once again, the locus of these myths is not the submarine[-mare] fire."85 This evokes another such vent, also identified only a geographical site-the region identified with Assam, in which the with a goddess; this is Jvalamukhī, "She of the Flaming Mouth," in Hima- temple of the goddess Kāmākhya is found-but also a locus within the chal Pradesh. In a generic sense, jvālamukbī can also have the sense of subtle body. Here, the geographic symbolism plays itself out on a double "(mouth of a) volcano," and it is here that we are brought back to the geol- register: Kamarupa is both that region of eastern India from which the ogy of alchemy, according to which mercury and sulfur naturally occur in Śaktism identified with the Yoginī Kaula is said to have originated90 and a areas of volcanic and geothermic activity. In fact, the origin myths of mer- portion of the body, subtle or concrete, identified with feminine sexuality. cury and sulfur evoke, for certain modern Indian interpreters, volcanic The mythic perils of Matsyendranāth are very likely a reflection of a cleav- eruptions. In the latter case, the sulfur that arises from the churning of the age within the tantric tradition. While a number of sects, including the ocean of milk would be the mythologization of a submarine volcanic Matsyendra's Yoginī Kaula, incorporated sexual intercourse into their eruption.86 practice,91 others, including the Näth Siddhas whose main doctrinal expo- If this is in fact the case, then the origin of sulfur-the uterine blood of nent was Gorakh, were overtly misogynous, treating sexuality (epitomized the Goddess associated with the lower half of the body in which the kun- by the vulva) as a trap into which the yogin could fall and thereby lose all dalinī sleeps-is to be identified with a flaring up of a female submarine the benefits of his prior efforts (in the form of his precious semen).92 As fire, a bādava. Lillian Silburn explains, this cleavage may be reflected in two understand- ings of the term fish belly, the one sexual and the other hathayogic:

d. The "Plantain Forest" in the Midst of "Love's Body" the yogin resides at the source of the movements of emanation and Gas vents and volcanos that are so many "mouths" of fire or flame in these resorption of the universe, a state praised as matsyodarīmata because mythic and geographical contexts evoke other types of mouths from tan- it is comparable to the stomach of the fish, which continuously con- tric imagery. Apart from the oral orifice in the head, the most celebrated tracts and expands automatically. This realm of bliss has some con- mouth in tantrism is the "lower mouth" (adhovaktra) i.e., the vulva (yoni) nection with sexual experience for, like the susumņā, the organs are of the yogini. And indeed, the ca. twelfth-century SSS (2.14), a foundational subjected to a similar contraction and expansion conductive to an text of the Western Transmission, identifies the mūlādbara cakra with the intimate union which, in a yogin, involves a perfect coincidence of submarine fire (vādavānala), which its commentator locates in the upper Siva and the energy, of subject and object, of seed and womb. It is part of the yoni.87 Like the kundalinī herself, the vulva is an organ fraught from this coincidence that supreme Beatitude and Consciousness with danger but also the promise of great power and pleasure for the prac- originate.93

236 237 Chapter Eight Homologous Structures of the Alchemical Body

The mythic cycle of the perils of Matsyendra, most fully developed in Ultimately, however, Gorakh must shock Matsyendra back into yogic two ca. seventeenth-century Bengali-language dramas, the Min Cetan (The consciousness. Matsyendra's fall into mundane consciousness is best sym- Awakening of Mīna) and Goraksa (or Gorakha) Vijaya (Gorakh's Victory),94 bolized by a son, named Binduknäth ("he who is formed of the drop [of are a garden of yogic verse on human weakness and yogic mastery in the yogically transformed semen]"), whom he has fathered on Kamala while face of female sexuality. These works offer two mythic explanations for her prisoner in Kadalī. Gorakh kills and skins the boy, scrubs his skin like Matsyendra's fall. The first is connected to later versions of the myth of a washerman to remove all its bodily impurities, and hangs this skin on the the revelations made by Siva on Moon Island: the Goddess, angry that roof to dry, like the hide of some skinned beast-all of these metaphors Matsyendra has overheard a revelation reserved for her, curses him that he for yogic purification. Matsyendra and Kamala are horrified at this act, and too will one day "fall asleep" and forget all he has overheard.95 A second when they bewail the murder of their son, Gorakh revives 108 Binduk- account casts the Goddess as the "stepmother" of the four original Naths, naths! Seeing at last the illusion of the world to which he has become whom she tests by tempting them with her sexual charms. When Matsyen- attached through his sexuality, Matsyendra now truly "awakens," and Gor- dra becomes sexually aroused, she curses him to be debauched in the King- akh leads him out of the Kingdom of Kadali (not before, however, the dom of Women.96 The YSA97 explains Matsyendra's fall as a direct upshot women of the kingdom attempt to take Gorakh's life-Gorakh avenges of his yogic powers. Wishing to experience the sensual life of a king, Mat- himself by transforming them into bats!).107 syendra performs the act of parakāyapraveśa, entering into the body of an- Let us dwell for a moment on the name of the kingdom into which other person.98 In this case, however, Matsyendra, debauched by Queen Matsyendra has fallen into a sexual lethargy as the plaything of some six- Kamalā, into whose husband's body (King Trivikrama) he has entered, for- teen hundred women: Kadalī Rajya is the Kingdom of the Plantain Forest. gets his true yogic self. At this point, the western YSA account dovetails Why plantains? Our first clue comes from classical Indian poetics, in into the Bengali account of Gorakh's rescue of his guru (called Mina in the which the thighs of a beautiful woman are compared to the smooth, firm Bengali song cycle) from the Kingdom of Women or, more properly, the trunks of the plantain tree. In this perspective, Matsyendra has been lan- Kadalīrājya, the "Kingdom of the Plantain Forest,"100 which is generally guishing in a forest of beautiful thighs, a fair statement arrived at through located in Kāmarūpa (Assam) or Simhala (Sri Lanka).101 simple arithmetic. Moreover, the plantain is that one of nine sacred trees The story opens when Gorakh learns, from Kaneri (or Känha), another representing the Goddess most closely associated with vegetative power. Näth Siddha, that his guru is three days away from his death in the King- So it is that plantain fruit, a small banana with seeds in its interior, is used dom of Women, where he has been the prisoner of sixteen hundred in fertility rites in Bengal, in which men swinging on hooks drop plantain women.102 The kingdom of Kadali, as Gorakh quickly learns, is prohibited fruits on women below. The imagery here needs no explanation.108 Else- to all males, save for the imprisoned "king" Matsyendra, and so it is neces- where, "plantain house" (kadalīgrha) is a term employed in Sanskrit drama sary for Gorakh to find a subterfuge. This he does by taking on the appear- to designate a pleasure garden in which the king assigned trysts with ance of a woman and engaging himself in a female musical troupe. It is women to whom he was not yet wedded.109 in this disguise that Gorakh comes to the court of Queen Kamala and This metaphor is greatly expanded in the south Indian Vīrasaiva poetry her sixteen hundred female subjects, where he finds his guru fallen into a of Allama Prabhu or Prabhudeva, who time and again compares the human swoon, at death's door.103 body, subject to passion and sorrow in a world of death and rebirth, to a The music begins, and when Gorakh plays the first beat on his two- plantain forest:110 headed drum, the drum sings out "Awaken, Matsyendra, Gorakh has come!" Upon hearing this drumbeat (nada),104 Matsyendra awakens from I see no one who can conquer the body's vast plantain grove: the his stupor. But Gorakh and Matsyendra are not yet out of the woods. Gor- seven seas of the world encircle it. In life's vast wood the poisonous akh must now return his guru's strength, his yogic powers to him.105 This rain of the five senses pours. Anger's huge tiger roars and roars ... he does by urging him to perfect his body anew: kāyā sādha, kāyā sādha The child of sin called greed is eating and eating ... The well of guru mochandar:106 passions cannot be used.111

238 239 Chapter Eight Homologous Structures of the Alchemical Body

The Plantain Grove, Kadali Vana, to which Prabhudeva is referring, is kaj[ j]alī is both black in color and an elixir of immortality), would sound first and foremost a geographical location: this is a grove at Srisailam, a far-fetched, were it not for the fact that the same term is found, in a similar cultic center of Virasaivism, but also an important center for such sects as context, in another source. This is the Punjabi legend of the Nath Siddha the Kāpālikas, Nāth Siddhas, and Rasa Siddhas.112 Quite unlike the nega- named Püran Bhagat. This yogin has been dismembered (whence his alter- tive portrayal Prabhudeva gives of his metaphorical Plantain Grove, the nate name of Cauranginath, "Four-[dismembered]-limbs Nath"), thrown Kadalī Vana at Srisailam was a source of healing herbs and magically reju- down a well, and left for dead by his evil stepmother, whose sexual advances venating waters. This tradition, however, of such a grove enclosing a pool he had refused. He prays to Gorakhnāth, who restores him to wholeness of healing and life-giving waters is not original to this site or this period: and initiates him into the Näth order, following which Püran Bhagat re- in the third book of the Mababbarata, Bhima finds such a pool in the midst mains in his well for twelve years, at the end of which he attains immortal- of a plantain forest located on the flanks of the Himalayas!113 ity. Püran Bhagat's well-also a fountain of life of sorts-is located in Ka- There are, in fact, a number of medieval literary references which sup- jalī Van, near Sialkot (Jhelum district), in the Punjab state of Pakistan.117 port this reading of the Kadali Vana toponym. One of these is Malik Mu- The eighth -to twelfth-century Yoga Vasistha generates a polar opposition hammad Jayasī's sixteenth-century Padmāvat, a beautiful yogic allegory of between the well and the forest that surrounds it: here, the human mind the wooing of Padmavati, princess of Simhala, by a prince-turned-Näth (which is trapped in an existence of its own making) is allegorized as a man Siddha named Ratansen. This source makes more than one reference to a who, having inflicted all manner of wounds upon his own body, plunges Kadalī Van which, accessible to Siddhas alone, contains a "fountain of life." into a well (the torments of hell); he then leaves the well to enter into the In this source, variant readings of Kajarī and Kajalī are also given, variants Plantain Forest (the pleasures of heaven).118 which may be read in three ways. The term is either: (1) kadali van, "plan- There are a number of levels upon which one may interpret these juxta- tain forest," a place identified with the sensual life (as in the legends of positions, either of a well or fountain of life at the heart of a Plantain Forest Matsyendranäth), but also with a grove of yogic realization and immortal- identified with bodily proclivities, a Forest of Mercuric Sulfide, identified ity (in the Padmävat); (2) kajalī van, "forest of black mercuric sulfide," of with bodily immortality, a land of death and darkness, identified with the mineral hierophany of the sexual essences of Siva and the Goddess, a failure to attain superhuman goals-or a combination of all three. Kaj- which does in fact constitute an elixir of immortality; or (3) kajari van, jalī, which can be black poison for the uninitiated, is an elixir of immortal identified with Zulmät, the name of the land of death and darkness (kaj- life for the consummate alchemist. Matsyendra's Kadalī kingdom was un- [j]alī also means "lampblack") to which Iskandar (Alexander the Great) doubtedly a forest of women's thighs, in the midst of which he nearly lost traveled, according to Muslim legend.114 his life-albeit through a form of tantric practice-but where he in the I have already discussed the first reading at length, The third reading is end realized yogic immortality through Gorakh's intervention. Püran Bha- treated in the Padmavat with specific reference to Alexander, as well as to gat too should have died in his empty well in the midst of the Kajali Forest, the prince-turned-Näth Siddha named Gopicand, who, realizing the tran- but instead realized bodily immortality through his initiation by Gorakh. sience of this world, left his kingdom and went to Kajarī Van, where he In the final analysis, the common theme to these evocations of an immor- realized yogic perfection. Like Prabhudeva, Kabīr identifies the Kajarī Van tality that one may grasp even in the midst of bodily sufferings is once with the human body that is subjected to the trammels of carnal passions: again a tantric one. A woman's thighs can lead to the death of a yogin, but "This body is a kajarī ban and the mind is an elephant gone mad, the jewel they can also constitute a "boat to immortality." Just as the kundalinī can of wisdom is the goad but few are the Saints who can apply it!"115 Another both drain a yogin of his semen (in a process called called "eating poison" source, the oral epic entitled the Alha Khand, has its eponymous hero in one source) and transmute that essence into the draft of immortality, schooled by Gorakhnäth in the transitory nature of mundane existence and so too the space between a woman's thighs-the Kamarupa in the midst of the virtues of the renunciant life. He follows Gorakh into Kajarī Forest, "a her yoni-can arbor a fountain of life for the tantric practitioner who mysterious land from which it is said that someday he will return."116 knows his way around.119 The second reading, which combines the first and third (the alchemical The tantric hero (vira) of the Kaula traditions was precisely that excep-

240 241

Chapter Eight Homologous Structures of the Alchemical Body

tional individual who was capable of experiencing, to its very limit, the life bination evokes the internal unity of these bipolar systems. The head and of the body in ways that would utterly destroy the beastlike noninitiate torso of the human body, the communicating chambers of a bicameral al- (paśu) for whom every venture into the prohibited world of feminine sexu- chemical apparatus, the vault of heaven reflected in the waters at the bot- ality was a deadly enterprise. Whereas the latter, bound like a sacrificial tom of a well:121 all of these are so many images of the unity in difference victim (paśu) to religious and social convention, could only come to know (bbedabheda) that is the hallmark of nondualist tantric thought. It is within pain and death through the mystico-erotic practices of the tantric sects, such a symbol system that the locus of ether (kha, ākāśa, vyoman, šūnya) in the former could use the same to transcend the human condition and expe- the human body is identified with the cranial vault, which is characterized, rience the bliss of Sivahood. We illustrate the double register, upon which in a number of sources, as a downturned well. the landscapes of the subtle body have been interpreted, with a chart: The terms kha and ākāśa at once signify, from the Upanisads onwards,

overcoming death "heaven," "cave," "hole," and "[empty] space." According to the Mabā- being-towards-death bbārata, ākāśa has the following properties: it is the elemental substrate

normal respiration swallowing into the fish-belly of sound (sabda), it is all-permeating (vyāpitva), and it has the nature of a

downward-flowing Ganges Fish-Belly Conjunction hollow or cavity (chidratva).122 In Ayurvedic physiology as well, the ele-

sleeping kundalini rising kundalinī ment ether is present in the body in the form of bodily cavities or orifices.

kālāgni kālāgnirudra/yogāgni In its quality as void or empty space, ether is vital to tantric worship: it is

underwater, simmering bādava exposed, upward flaring bādava in a clearing within one's own being that one meditatively constructs the

samsāra sambāra divine object of ritual devotion.123 As I demonstrate, the realization of

adboretas ürddbvaretas an analogous void is necessary to the hathayogin, in his quest for the ab-

Kāmarūpa/adhovaktra Candradvīpa/yoginīvaktra120 solute.

deadly Plantain Forest rejuvenating Plantain Forest Since the time of the Upanisads, the seat of the soul has been located in the heart.124 There, the soul is suspended in the midst of a void that extends outwards for ten finger breadths from the core of the subtle body. The void or ether (kha, akasa) of the heart remains a key notion in Hinduism, down 3. Lunar, Empty, and Hg Wells in Hindu Tantrism into the Tantras.125 However, the concept of another "inner void" also emerges in the Upanisads to continue down into the tantric period. This In the last chapter, we spoke of the creative opposition between a bird is the void of the medial channel of the subtle breath, which bisects the in the vault of the "cranial" heavens and a serpent in the hollow of an "ab- subtle body vertically.126 This channel, running upward from the mūlādbāra dominal" cave. Apart from their symbolization as bird and snake, these two cakra through the heart (which bisects the body horizontally), ends in the poles of the Samkhyan hierarchy of elements, the alchemical hierarchy of "cleft of brahman" (brahmarandbra), the fontanelle, in the top of the cranial metals, and the yogic hierarchy of the cakras may also be represented as a vault. Whence the statement in the Kāthaka Upanisad (6.16): "There are a set of two wells. The image is a powerful one, in any tradition, for what is hundred and one channels of the heart. One of these passes up to the crown a well if not a channel between our sunlit world and the dark subterranean of the head. Going up by it, one goes to immortality." regions? Chthonic dragons guarding enchanted wells, wishing wells- It is this channel, but more especially the upper endpoint of its trajec- common fare in western folklore, they attest to the universality of this tory, that the Nath Siddhas and a number of other tantric schools identify theme. In the yogic and alchemical traditions, the first of these two wells with the ether or void, which they also call kha or sunya. For the Nath is conventional, in the sense that it is upturned and set into the "base" of Siddhas, sunya is at once the highest metaphysical principle (param tattva), these parallel structures. The second well, corresponding to the upper end the cranial vault or fontanelle, and the abode of Siva.127 Here, the impor- of these parallel hierarchies, is downturned and placed above the second. tance of the ether lies not in what it is but rather in that which it contains When the two are brought together into a double-well image, their com- in its empty space-and here, once again, yogic and alchemical elixirs be-

242 243 Chapter Eight Homologous Structures of the Alchemical Body

come identified with one another, this time through their association with metaphorical wells in the city, the well set into the earth and that which the "wells" in which both are found. takes the form of water pots on the women's heads. In no other medieval Indian tradition do we find greater importance On the level of the bodily microcosm, the interpretation is the follow- attached to this image, of the body as a set of wells, than in that of the Nath ing: the well set into the earth here is the muladhara, the lowest of the Siddhas. Gorakhnāth, who uses the terms gagan [mandal], śūnya, and kha128 cakras from which the yogin's semen, the raw material of his bodily trans- to speak of the cranial vault, also states, in perhaps his most renowned bāni: formation, is raised. As it rises along the length of the medial susumņā nādī, "In the circle of ether is an inverted well that is the place of nectar. He this semen is transformed into nectar. This process is consummated in the who has a guru drinks his fill; he who has no guru goes thirsty."129 Here, uppermost sabasrāra cakra-the thousand-petaled lotus located in the cra- Gorakhnäth's well of nectar in the ether is at once the seventh and upper- nial vault, the ethereal sphere of the void-sunya or gagan mandal-which most cakra, the thousand-petaled sabasrāra located in the cranial vault, and is figured in Gorakh's poem by the water pot atop the female water bearers' the moon, repository of immortal soma, shining in the heavens. The moon, heads. Just as these female water carriers (who may be further identified identified with the god Candra/Soma, has been viewed as a primal source with the kundalint) slake the thirst of the animals in Gorakh's metaphorical of ambrosia since the time of the Rg Veda.130 In Gorakhnath's bani, the dow- city, so the yogin's refined semen, now transformed into nectar, fills the nturned well that quenches mortal man's thirst for the nectar of immortal- "void city" (sunya puri) of the cranial vault. When allowed to flow back ity is identified with the soma that is the stuff of the full moon. downwards, it floods the yogin's body with its fluid of immortality.135 Another of Gorakhnath's poems, contained in the Gyan Calisa (Forty This poem schematically portrays the yogin's body as two vessels, a Gnostic Stanzas),131 combines the image of the well with the attainment of lower well and an upper water pot, of which the lower is a source of energy yogic and alchemical siddhis. Here, the yogin who fills the ākāa (the ether and raw materials and the higher the container of the subtle immortalizing of the cranial vault) with water from a subterranean well gains bodily im- essence of those raw materials. In a Nath Siddha legend as well, a similar mortality and the power of transmutation. The same symbolism is found identification is made between the body and two superposed water pots in a number of poems of Kabīr, the language and imagery of whose mystic carried on a woman's head. In the tale of Guru Gūgā, or Gūgā Pīr, a low- poetry was greatly influenced by the verses of Gorakhnäth. In one of these, caste woman speaks of her body in terms of these, protesting to Gūgā that Kabīr says: "The well of heaven has its opening below; its bucket is in she cannot give him water to drink from her earthen pot (because the vessel the underworld (patale); the hamsa drinks of its water, but few know of its of her body is impure). Güga lets fly an arrow, smashing the two pots si- source."132 In another, he speaks of a land with "an inverted well, with an multaneously, to which the woman reacts with the words "Look at the state opening, as narrow as a thread, through which the married soul draws of my body; thou hast broken my two pitchers."136 Once again, we find an water ... From the well in the lotus above, the devotee drinks the echo of this imagery in Kabir (even if he is mocking the Siddhas in the nectar."133 same verse): "Between the sky and the netherworld, in the space between These images are similar to those found in a longer poem by Gorakh- the two gourds [is the Path]: the six darsanas tripped and tumbled down näth, who once again uses the dynamics of drawing water from a well to along with the eighty-four Siddhas!"137 describe the workings of the subtle body.134 Here, he compares the human In the raising of vital breath, seed, and energy, it is at the level of the body to a city filled with all manner of hungry and thirsty animals, which navel that a preponderance of raw semen and thermal yogic energy yield are so many allegorical representations of the human condition. So the to a preponderance of cool, refined nectar; therefore, it is here that the cows and buffalo of the city, tethered to the stake of illusion, represent the mouths of the upper and lower "wells" of the yogic body are joined. So it absence of discrimination (aviveka), while its dogs are the mind that steals is that we find a metaphorical symmetry between these two poles or "wells" away and conceals true knowledge. In this city, however, is a well whose of yogic imagery. Above, male semen, moon, soma, nectar, Siva, fluidity, water slakes these animals' thirst (for liberation). From this well, the and coolness are identified with an upper well; and below, female uterine women of the city draw water which they carry in pitchers on their heads- blood, sun, fire, energy, the Goddess, dessication, and heat are localized in and these pitchers thus constitute portable wells. In this way, there are two a lower well.

244 245 Chapter Eight Homologous Structures of the Alchemical Body

This image of the body as a well into which the yogin enters in order to semen are simultaneously raised into the cranial vault, the abode of Siva. succeed in his sadbanas carries over into Nath Siddha myth and legend as well. The fullest account to bring such imagery into play is the Punjabi It also evokes a regressus ad uterum from which the yogin, like the Vedic

legend of Püran Bhagat, already mentioned above, who is dismembered sacrificer in his initiation hut and the Ayurvedic patient in his womb hut, emerges, reborn, as well as the ultimate alchemical operation described in and thrown into an empty, broken-down well (jīrņāndhakūpa). Gorakhnāth the RA, in which the alchemist's own body becomes the corpus alchymicum comes to rescue and restore him, after which he returns to the well to per- when he plunges into a cauldron of boiling oil.145 We will return to this form yogic austerities for twelve years: when he emerges, he has been trans- formed into a Siddha.138 Down to the present day, Punjabi women come, on image, of a Siddha entering a well that is a mesocosmic double of his own body, in chapter ten. the new moon, to partake of the cool healing waters of Puran Bhagat's well Wells are generally considered to possess healing properties, the well of (which is no longer empty), which are said to cure barrenness.139 Vaidyanath in Benares-as well as Puran's Well and wells of Gorakh in In a number of versions of the Gopicand legend cycle, this Nath Siddha seals his guru Jālandharanāth (or Hādi-pā) into a well; upon his release Kacch (Gujarat) and other Nath Siddha holy sites throughout India146- being prime examples.1+7 Many tantras contain instructions for the dedica- from the well, again after twelve years, Jalandharanāth reduces metal effig- tion of wells to the Goddess; and while the sakta pithas are not wells them- ies of Gopīcand to ashes.140 A nineteenth-century hagiography has Mast- selves, they are generally located in the proximity of a well or spring.148 nāth, another Nath Siddha, creating a well to save the besieged city of his The HYP (1.13) instructs yogins to live alone in a small monastery in princely protégé Man Singh.14 Gorakhnāth dries up wells in anger, but whose courtyard a well should be located. The alchemist's laboratory also fills wells that are empty.142 Elsewhere, he throws gold bricks or coins should also be equipped with a well in its northeast corner, which is that into a well near Kāmākhyā, thereby turning its waters to gold. When his part of the laboratory in which are also housed his all-important transmut- teacher, Matsyendranäth, insists that such will upset the order of the world, Gorakhnäth turns the well water into crystal; and this is the origin of crys- ing elements.149 Yet another case of such structural homologies between body, alchemi- tal in the world.143 Näth Siddha accounts of yogins who create and enter into wells in order cal apparatus, and universe is to be found in the tantric geography of the Indian subcontinent. In the last chapter, we briefly passed in review a num- to meditate seem to parallel Taoist myths of "old men of the gourd" who ber of alchemical hierophanies found at such sākta pīthas as Hinglāj in Ba- suspend a gourd from the roofbeam of their huts and enter into the uni- luchistan and Kāmākhyā in Assam.150 In addition to these, such important verse of that gourd in order to attain the Absolute, the Tao. The Taoist Saiva sites as Girnar in Gujarat151 and Srisailam in Andhra Pradesh152 are alchemical sources dating from the first half of the first millennium A.D. also possessed of alchemical mythologies. The most systematic and strik- are particularly rich in legends on this subject, such as the following: ing example of the geographical and geological macrocosm reproducing Che Ts'ouen, who came from Lu, was a student of the Great Alchem- the alchemical and yogic macrocosm is that found, however, at Kedārnāth, ical Path. After meeting Chang Chen, he became the administrator an important and ancient Siva temple built over a jyotirlinga located in the of Yun-t'ai (i.e. the mountain called Chiang-su). It was his habit to Chamoli district of northwestern Uttar Pradesh. suspend a bu[-lu] vessel of about ten liter's volume [from the roof The final stage of this pilgrimage, the fifteen kilometers that separate of his house]. This hu[-lu] vase transformed itself into Heaven and the village of Gauri Kund and the temple of Kedarnath itself, has been Earth; it contained sun and moon. Che Ts'ouen passed his nights marked, at least as far back as the twelfth century and down to the present there, and called himself "Heaven in a [Double-]Gourd" (hu-t'ien). day, by a series of natural springs or pools (kunds). Down in the village of The people called him the "Old Man of the Gourd." Following this Gaurī Kund (this being the name of a natural hot-water spring in which he realized the Tao [and became an immortal].144 pilgrims bathe), a reddish, sulfurous pool close by the side of the Gaurī Temple is known as Rtu Kund ("the menses pool," where the Goddess takes As in the Taoist case, entering into a well is, for the Näth Siddhas, tanta- her menstrual bath), while at the upper end of this final stage and some mount to entering into yogic practice, in which consciousness, breath, and thousand meters higher in elevation, behind the temple of Siva Kedārnath

246 247 Chapter Eight Homologous Structures of the Alchemical Body

and at the foot of the peak of the same name, is a clear pool called Ret the place. Here, the seven pools of Ataka, crowned by a door, a sword, Kund ("the semen pool"), which is said, in both Puranic sources and mod- a river, and a philosopher's stone, appear to be so many representations ern pilgrim's guides, to contain the mercury that arose from Siva's seed, of the yogic body and the raising of the kundalini up to the seventh cakra dropped from the mouth of Agni when he transported it to the shore of the which, when pierced, yields the nectar of immortality.155 Ganges. Slightly higher, just behind the temple is the Amrt Kund, "pool of Another parallel may be found in the Indo-Tibetan Buddhist tradition, nectar"; and still higher, some three kilometers to the north of Kedārnāth, mentioned in the previous chapter, of the cave known as Ca-ri in south- is the Brahmaguphā, the "cave of bráhman." eastern Tibet. This cave, entered through a narrow cleft identified with Once again, but this time on a quite massive scale, the alchemical body the Goddess's yoni, opens into a series of caves and lakes (inhabited by gods, is reproduced, with Siva, semen, and mercury located at the upper end of goddesses, and dākinīs), which are explicitly identified with the nādīs and a geographic hierophany and the Goddess, uterine blood, and sulfur at the cakras of the Buddhist subtle body. The pilgrim who follows the subterra- lower end of the same self-enclosed system. The relative locations of these nean passages of this "womb-cave" to their very end is in fact effecting the two pools on the Kedarnath pilgrimage route also clue us into the fact that ascent of the goddess Vajravarāhī's medial channel, from her yoni up to her the imagery here is hathayogic as well: the pool of blood below and to the brabmarandhra. The pilgrim, borne upward as it were by streams of divine south is to be identified with the yogin's lower cakras, the place of the sun blood, semen, and sindūra (mercuric oxide), concretely effects the union of and uterine blood; the pool of semen above and to the north is the place Prajñā and Upāya (the Buddhist homologues of Siva and Sakti) and su- of the moon and of semen that has been transmuted into the nectar that preme realization.156 rejuvenates the yogin's body and so affords him jīvanmukti. Still higher, the In addition to the important role they play in alchemical geography and pool of nectar and cave of Brahman are further indications that the upper the poetry of yogic experience, wells also are vital, as we saw in the last reaches of the Kedarnath site correspond to the ambrosial climes of the chapter, to myths of the origin of mercury.157 Also in the last chapter, we cranial vault and the fontanelle, the cleft of brahman.153 reproduced an account, from the RRS and other sources, of the extrac- It will be recalled that, in the schema of the sakta pitbas, it is the shrine tion of mercury from one such well, in Darada-desa, by knowers of tantra of Hinglaj Devī, located on a mountaintop in southeastern Baluchistan, (tantrajñās) and semidivine Wizards (vidyādharas). This is the technique by that is identified as the Goddess's brahmarandhra. Like Kedārnāth, the which a maiden on horseback entices mercury to rise up out of its well and route to Hinglãj, too, is spangled with images of the yogic body. The most pursue her over hill and dale. Once it has been caught in troughs dug to important of these is the Moon Well, candrakupa, which, at the upper end this end, these sources tell us that this mercury is purified in a sublimation of the pilgrim's long journey, is a place of great geothermic activity, of fire apparatus (pātana yantra).158 and water that well up out of the depths of the earth. Both sun and moon The medieval alchemists in fact knew of three types of sublimating ap- are also present here, in the form of massive carvings hewn into the living paratus (yantras)159 for the extraction (akrsti) of mercury from cinnabar, rock: this may indeed be the earliest such image on the Indian subconti- i.e., the naturally occuring mercuric sulfide (darada) in which it is most nent, already appearing as it does in the writings of the fourth-century commonly found: these are the apparatus of upward sublimation (ürddhva- Greek geographer Ctesias.154 patana), downward sublimation (adbopātana), and transverse sublimation Yet another yogic schematization of a natural landscape may be found (tiryakpātana). In all three apparatus, mercury is made to sublimate and in the description of a Nath Siddha site called Ataka, located in the eastern thereby leave behind its residual irpurities in the ores in which it naturally Iranian region of Khorasan. Here, the intrepid pilgrim passes by a series occurs (or with which it is amalgamated), and thereafter to recondense of six pools until he comes to a cave upon whose door is figured a "fear- through the interposition of cold water.160 some cakra." Over the door hangs a sword. Whoever touches this cakra The upward-sublimation apparatus is composed of two superimposed "will explode into a thousand pieces," but he who succeeds in passing vessels, whose mouths are sealed together with several layers of mud- through this door gains access to the all-transmuting philosopher's stone smeared cloth stretched across their interface. A slow fire beneath the (sparsamani) locked inside and to the Ganges River, which flows through lower vessel heats a mixture of herbs and powdered mercury ore, mercuric

248 249 Chapter Eight Homologous Structures of the Alchemical Body

sulfide, most often in the form of cinnabar. The mercury that evaporates (drsti) below, the cleft (bheda) above and the channels (sirab) below, one upwards condenses on the inner surface of the downturned base of the becomes liberated in the body by using the [vidya]dbara-yantra." 165 upper chamber, which is cooled from above by a cloth soaked in cold water Here, the yogin whose "gaze" and "channels" are below is clearly stand- (or by a superimposed cold-water recipient). The mercury that has con- ing on his head: the sole description of a hathayogic technique in which densed on this downturned inner surface has a smoky luster to it: when the term sirab is employed in the plural is that of the "chin lock," or jaland- rubbed with a cloth, it immediately takes on the properties of fluidity, lus- bara bandha, so called in honor of the great Näth Siddha, but also in refer- ter, etc. one associates with pure quicksilver. In the bottom of the lower ence to the "net[work]" or "ganglia" (jalam) of channels which meet in the vessel there remains the dross of this reaction: free sulfur together with region of the throat. This practice effects a lock (bandha) on these channels other mineral and plant matter, oxidized and devoid of its original mer- and thus seals off the head as the recipient (dhara) of the nectar held in the cury content.161 cranial vault and not allowed to fall downward into the fire of the sun in Because of the function it serves, this apparatus is known as the ürddhva- the yogin's lower abdomen: jalam-dhara bandha. This upside-down version patana. This is not, however, the sole name by which it is known: a number of that technique is the structural parallel to the vidya-dhara yantra: the two of medieval alchemical sources also call it the damaru yantra (Siva's two- chambers of the alchemical apparatus are the head and torso of the yogin, headed drum apparatus),162 the sāmbhavī mudrā (the hermetic seal of Sam- the mud-smeared layers of cloth stretched across their mouths are the net- bhu, Siva),163 and the vidyadhara yantra, the "wizard apparatus." 164 This first work of channels, and the nectar that is held in the head is the mercury term clearly refers to the shape of this apparatus, the second to a technique that condenses on the incurved inner surface of the downturned upper of hatha yoga, and the third to both such a technique and to the mythic chamber.166 A number of other interpretations of this verse are possible, Wizards who were the innovators of a body of techniques that, precisely, including a sexual one: the Yoni Tantra employs the term dhāraka to signify fused alchemical and yogic practice into a single Saiva system. both "vulva" and "vessel." 167 The structure and dynamics of this yantra appear to replicate the dy- Most alchemical sources enjoin the practitioner to follow his use of the namics of yogic reversal-known as ürddhvaretas from the time of the Ma- ürddhvapatana yantra with that of the adbopātana yantra, the "apparatus of habharata-as such is described by Gorakhnath in his poems about wells downward sublimation": after having sublimated mercury three times and water jars. In this case, the upper chamber of the apparatus is the dow- in the former, one is to do so seven times in the latter.168 The adhopatana nturned celestial well in the yogin's head, and the lower chamber the matrix yantra is identical to the ürddhvapatana with the difference that in this case, of matter and energy-in the "pot" formed by the diaphragm through the the "cool" chamber is placed rightside-up in a hollow dug into the ground. hathayogic technique called kumbhaka-from which the gross elements Its mouth is bonded to that of the "hot" chamber, here placed facedown, are caused to rise through heating. The fire beneath and cool water atop with the cooking fire burning atop its upturned base. In this case, one the two chambers complete the parallel. These are the equivalent of the smears the inner surface of this downturned "hot" chamber with an amal- visceral sun and cranial moon, through the balancing of which the yogin gam of mercury and copper. When this is heated, the mercury sublimates comes to enjoy the nectar (mercury, the refined semen of Siva) that "con- and falls down into the "cool" chamber, which is this time filled with water. denses" to ooze downward from the top of the cranial vault, the lunar There, it recondenses into its natural, but purified and stabilized, form.169 circle. The adhopatana yantra is in fact a closer homologue to the upside-down These parallels are made explicit in a number of medieval texts. In the yogic posture Gorakhnath describes in his Amanaskayoga than is the vid- fifteenth-century alchemical Rasendracintāmani of Dhundhukanātha, this yādbara (or ürddhvapātana) itself. In both the alchemical adhopātana yantra bicameral alchemical apparatus is termed sāmbhavi mudrā. Now, this same and the dhara yantra described by Gorakhnäth, the "seminal" fluid, trans- term is used, in the Amanaskayoga of Goraksanātha, to introduce a yogic formed by a heat source located above, drips downward to remain (held technique that that text seems to identify with the vidyādbara yantra: "Here there by gravity) in the reversed upper chamber of the apparatus or, in the is the description of the sambhavi mudra: The fist [?] above and the gaze yogic case, of the cranial vault. This posture also evokes the viparītakarana,

251 250 Chapter Eight Homologous Structures of the Alchemical Body Nepal, as well as by Tibetan Buddhist monks. This is an instrument made or "reverse practice" of yoga, which Gorakhnäth praises in his Goraksa Sa- from two conical or hemispherical drums, joined together at their tapered taka (133-35) as a foolproof method for retaining in one's head the pre- ends, with their heads facing outwards. Held at its tapered middle, the cious nectar one has so carefully distilled from "raw" semen. Here, the description of the posture itself is clearly the inspiration for that of the damaru is played by turning the wrist back and forth such that a bead or thong, attached by a cord to the tapered portion, strikes the two heads in dhārayantra, in everything but name: "The navel above, the palate below; alternation. The two chambers of the drum-like those of the vidyadhara the sun above and the moon below."170 Gorakh refers to this posture in yantra, the upper and lower halves of the human body, and the upturned one of his bänis, in which he says to hold the yantra upside-down (ulati and downturned wells of the macrocosm-represent the bipolar relation- yantr dhare) by standing on one's head. These structures and dynamics of ship between feminine and masculine principles common to so many mys- the alchemical body become dramatized in a nineteenth-century hagiogra- tic Hindu traditions. These are the states of involvement in and renun- phy of Lallā, or Lal Ded, the fourteenth-century Kashmiri mystic poetess, ciation of the manifest world through which the human yogin alternately who demonstrated "true penance" to her guru Srīkantha. "She placed an passes, in imitation of the divine yogin, whose yogic withdrawal from and earthen pot on her head and another under her feet; and with the waning of the moon, her body waned till, on the fifteenth night of the dark fortnight return to mundane consciousness mark the successive resorptions (pra- layas) and emissions (srstis) of the cosmic eons. (amävas), nothing was left of her except a little quantity of trembling quick- The bead that successively strikes the two ends of the damaru to emit silver. Then, with the waxing moon her body waxed and, on the full-moon the drum's sound (nada) is also identifiable with the mercury that evapo- night, she was herself again."171 This body of Hindu practices, which incorporates alchemical and ha- rates out of one chamber of these yantras to sublimate in the other. It also represents the yogin's seed (called bija, which is often closely identified, in thayogic techniques, bears striking similarities to an earlier Taoist tradi- tantrism, with nda)174 which rises (ürddhvaretas)175 with the kundalin and tion, which constitutes a still more graphic projection of alchemical imag- returns to the lower body when the yogin returns to mundane conscious- ery upon the subtle body. This is the practice termed "feeding the vital ness and activities at the conclusion of his austerities. principle." This technique, described in a fourth-century Chinese source, It is another sort of two-headed drum (called a mrdanga) that Gorakh- the Taishang suling dayou miaojing, represents the human body as a set of nath carries in the legend already discussed, in which he "awakens" his three "cinnabar fields," of which the uppermost, located in the head, con- teacher Matsyendranäth from a life of debauchery in the Kingdom of tains an ordered hierarchy of nine "palaces." The third and innermost of Women with the first beat (nada) that he strikes on it. Tibetan Buddhism, these palaces, the Palace of the Cinnabar Field, is the culminating point of whose lists of Mahasiddhas include the names of Gorakh, Mina, and other breath absorption, the Taoist precursor of the Indian prānāyāma. Immedi- illustrious figures who came to be identified with the Nath Siddhas, also ately behind this is the fourth palace, called the "Palace of Moving Beads," uses the damaru in its ritual practice. A number of damarus, used by prac- i.e., of quicksilver, fluid mercury. Once again, the question of whether titioners of tantric Buddhism in Mongolia and Tibet,176 are fashioned from these striking parallels-between this feature of the archaic Taoist body two human skulls, joined together near the fontanelle, with a rectangular and the "alchemical body" of Hinduism-are reflective of cultural and sci- hole177 drilled between them which, establishing communication between entific exchanges between China and India in this period, a period follow- the two hemispheres, "allows the instrument to breathe and thereby have ing the export of Indian Buddhism into China, must remain outside the a better sound." This hole is called the nāda.178 The damaru player gener- scope of the present study.172 ally holds a bell (ghanta) in his left hand, which thereby complements the The basic structure of the Hindu alchemical yantras, with their down- damaru held in the right. Taken together, the two are said, in the Vajra- turned upper and upturned lower chambers, may be taken as a model for yāna context, to symbolize sun and moon, the right and left channels, and both the universe and the human body. Another somewhat simpler homol- skill in means (upāya) and wisdom (prajña),179 the Buddhist homologues to ogy to the vidyadhara yantra is the two-chambered damaru (indeed, there Siva and Śakti. More than this, the two skulls which make up the drum exists an alchemical damaru yantra as well),173 the "shaman's drum" carried by the god Siva and his sectarian followers throughout north India and are ideally those of a sixteen-year-old young man and a sixteen-year-old

252 253 Chapter Eight Homologous Structures of the Alchemical Body

maiden. Of course, sixteen is a figure signifying wholeness or plenitude in narīśvara comprises the right half of Siva's body and the left half of that of these traditions, but it will be recalled that the origin of this identification Sakti, the Goddess. With this, we are brought back to our discussion of a goes back to the lunar symbolism of hathayogic practice: sixteen is the subtle body whose left or western (paścima) side is female (in spite of the number of digits in the full moon that brims with nectar in the fully real- fact that the lunar, male ida channel issues from the left nostril and the ized yogin's head.180 solar pingala from the right). As I have argued, the subtle landform known Another source from the same Vajrayana tradition reproduces this as Moon Island is located either in the left side of the lower abdomen or structural homology of the two chambers as the two peripheral channels. cranial vault. In a Caryā song (no. 3), Virūva-pā analogizes the fermentation of wine with It is in the cranial vault that the complementary processes of breath, the dynamics of yogic breath control:181 seed, and mind control culminate in the production of nectar, a new yogic body, and the state of total yogic integration that is samadhi.184 The prac- The female wine-dealer [avadhūtī, the Buddhist kundalinī] is alone, titioner cannot, however, merely fill his head with transmuted semen and but she seals together the two chambers [the two channels, sun and expect instant immortality. In the last chapter, I evoked the yogic and moon] and introduces the powder for fermenting the wine [bodhicitta, alchemical notions of khecara, "moving in the ether," and the erotico- semen]. The wine [varuni]182 now flows [borne upward on the subtle mystical and hathayogic technique of vajroli mudra, while a moment ago I breaths] straight to the Sahaja [sahasrāra] and that brings immortality. discussed the image of the cranial vault as an empty space (kha, šūnya) or On seeing the sign [of the wine] at the tenth door [the brabmadvāra, well that comes to be filled with nectar through a variety of yogic tech- at the upper endpoint of the central channel],183 the customer [the niques. At least one hathayogic source designates the practice of khecari yogin] comes of himself and drinks the nectar from the lotus of great mudra, "the seal of moving in the ether," as the culmination of vajroli mudrā bliss [mabāsukba, the Buddhist sahasrāra]. and samarasa.185 The KhV, which treats of every imaginable variation upon With these astonishing structural homologies, we are brought back to the theme of khecara, includes discussions of the khecari mudra, the khecari the cranial vault in which we began this section and to the fusion of all the mantra, and the khecari cakra.186 We now turn to a closer examination of polarities we have been discussing. this body of techniques. Although a passage in the Maitri Upanisad appears to allude to this prac- tice, the classic description of khecarī mudrā as a hathayogic practice is 4. Mouths and Doors: The Khecari Mudra in Tantric Practice found in the Goraksa Sataka: "The consciousness wanders in the ether (khe) because the tongue, having entered into the hollow (khe) above the throat The bipolar locations, on a vertical axis, of male mercury and semen above moves about. For this reason, the khecari mudra is revered by all accom- and female sulfur and uterine blood below are but one means to charting plished yogins." In practical terms, kbecarī mudrā is a technique that enables the basic structure of a hierarchized tantric universe. However, in a tantric the yogin to retain in his cranial vault the nectar that he has accumulated metaphysics that stresses unity-in-difference (bhedābbeda), or nonduality- there through his practice. Here, rather than allowing that nectar to fall in-duality (dvaitādvaita), there has as well to be a stress on equipose, equiv- into the fiery sun at the base of his spine, even when "in the embrace of a alence, and equanimity, on the union or coincidence of polar opposites. In beautiful woman,"187 the yogin internally drinks that transmuted semen practical terms, the factoring of the two into the one has been perennially and is thereby rendered immortal. enacted, in tantra, through sexual union between practitioner and consort. The key to the technique of kbecari mudra lies in the left/right division However, every subtle body is, in the traditions under study, intrinsically of the human body and the hathayogic notion of mudra itself. In literal androgynous, being divided along the vertical axis between male and fe- terms, the various mudras are so many "seals," by which the yogin is able male halves after the fashion of Ardhanarīsvara, the "half-female" form of to close off or isolate portions of the subtle body, or the body in its entirety, Siva. Following an Indian (if not human) commonplace that a woman's in order to realize the "pneumatic" or "hydraulic"188 feats of his craft. As place is on the distaff, the left side of her husband, the image of Siva Ardha- does the alchemist with his various apparatus, the yogin must begin his

254 255 Chapter Eight Homologous Structures of the Alchemical Body

practice by hermetically sealing his body off from the outside environ- by the two-mouthed kundalini, who can also at times be referred to as san- ment. This he does by closing the "nine doors," the bodily orifices.189 Fol- khinī.198 Whence Gorakh's succinct statement: "The tongue and penis are lowing this, through the practice of breath control, etc., he gradually closes [joined by] a single channel. He who retains [his rasa within it] tricks off all of the seventy-two thousand nadis, the breath and energy channels Time."199 With this, all the yogin needs to know is which valves to open in his body, save the medial susumnā, through which he will raise his breath, and which to close. seed, and mind up into his cranial vault. The identification of the two-mouthed sankhini with a channel originat- According to the yogic physiology of the subtle body, the susumņā actu- ing in the left side of the cranial vault evokes one of the esoteric classifica- ally branches into two conduits, at the level of the medulla oblongata (mas- tion systems of the postscriptural systematization of the Hindu tantras. taka granthi). Here, one branch veers slightly to the left,190 towards the This is the paścimamnāya or Western Transmission, a tradition marked by "lunar" side of the body, at which point it connects with the tenth channel, a certain emphasis on batha yoga and by its cult of the goddess Kubjikā, the called the sankhini, the "conch."191 It is this turn, as well as an inner aper- "crooked one,"200 whose teachings were brought into the world on a lunar ture known as the tenth door (dasama dvara), that affords the yogin the western mountain (Candragiri) or island (Candradvīpa), associated with possibility of preserving and using to his advantage the nectar he has chan- the western face (Sadyojata) of Siva. As already argued, these "western" neled upward to this point. venues correspond less to the topography and toponymy of the Indian sub- In order to effect khecarī mudrā, the yogi forces the tip of his tongue192 continent than to the physiology of the subtle body. On the one hand, the back and upwards into the opening at the back of the soft palate. Then, by western direction, when transposed upon the bodily microcosm, becomes concentrating his gaze inwardly to a point between the eyebrows, he causes the left side of the body; and indeed, one Näth Siddha poem refers to a his "capitalized" nectar to flow, via the sankbini, into the throat, where he "western door" (paśchima dvāra). This corresponds to an internal "western drinks it. This process was explained to me by a young yogin of the Jūna linga," of which there are two, another case of doubling in the subtle Ākhāda193 (a suborder of the "militant" Naga division of the Dasnāmi or- body.201 In a universe identified with the Goddess's subtle body, microcosm der, a division that has long been closely related to the Näth Siddhas), and macrocosm can at times become confused. So, for example, the MBbT whom I met on the Kedarnath pilgrimage trail in May 1984. As he de- states that Srīnātha is situated on the left side (vāmabbāgasthā); and in fact, scribed it, all depended on forcing the Goddess (kundalini) up into the he is the consort of the Goddess at the westernmost pitha of the subcon- head, from whence the rain of nectar (amrtavarsa) would flood down into tinent, at Hinglãj, the site identified as the Goddess's fontanelle (brahma- the body through the mrtyūnjayā nādī, a synonym for the sankhinī.194 randhra).202 It is here that yet another sort of doubling occurs in the subtle physiol- Two other readings of paścima are possible here. Mark Dyczkowski ar- ogy. First, the Amaraugbasasana explicitly states that the tenth door is two- gues that the term is to be taken temporally: the paścimāmnāya is the "sub- fold (dvividha), composed of "the seminal path of immortality, and the path sequent" or "latter" transmission, as opposed to the "former" transmission of death."195 The former of these is the "upper mouth" of the sankhini, of the pūrvāmnāya.203 Elsewhere, Gorakhnāth identifies the medial susumņā located in the cranial vault, within the subtle rajadanta, the "royal tooth," channel with the pascimamarga or paścimapatha in his Yogabīja.204 Now, one which resembles an ivory tusk. It is this tenth door that one causes to open could read this latter compound as "western path," and perhaps even as through the practice of khecari mudra.196 The latter, the lower mouth of the "subsequent [to the emptying of the left and right channels] path." One sankinī, is the place where the medial channel issues into the cranial vault: could also, however, read paścima here as "back" as opposed to pūrva, by closing off this aperture, the yogin ensures that the nectar in his cranial "front." In this case, the paścimadvāra, paścimamārga, and paścimalinga, vault will not drain back down this channel and fall into the fire of the sun would be the "back (i.e., secret, difficult of access) door," the "back way," at the base of his subtle body.197 Not only are there two mouths to the and the "linga at the [entrance to the] back [way]." sankbinī, but there appear to be two śankhinīs as well. On the one hand, the We now return to the mouths at the ends of the channels we have been Goraksa Śataka (v. 31) states that this duct is located at the base (mūlasthāne) discussing. In Abhinavagupta's discussion of the "clan sacrifice" (kulayāga), of the body; on the other, this "conch," spiral or curved in form, is doubled we are told that "knowledge is to be placed in the mouth of a woman and

256 257 Chapter Eight Homologous Structures of the Alchemical Body

then taken from the mouth of [that same] woman." The woman in question is the tantric practitioner's consort (dūti) in erotic Kaula ritual practice, and This container is placed in the midst of a fire pit filled with cow dung cakes,

her mouth is her sexual organ.205 Such is also stated clearly in the Vama- which provide constant and controlled heat, for the roasting of mercury in

mārga, a twentieth-century guide to ritualized tantric sex, in which the combination with various metallic and herbal elements.210

male adept is enjoined to assimilate himself to Siva and his partner to the In fact, when any two halves of a dual, bipolar system (a man and a

Goddess and offer his father-mouth into her mother-mouth.26 woman for example) are brought together in tantric sources, it is generally

There are, in fact, two mouths at play here as well, or rather a single the term samputa ("interlocked," "encased") or samputīkaraņa ("interlock-

aperture which, located at the apex of a mystic triangle in the mūladbāra ing," "encasement") that is employed.211 Applications of this term extend

cakra, is called the "lower mouth" (adhovaktra) when it is turned down- well beyond the realm of alchemical apparatus to a number of yogic and

wards, at which time it feeds into the (female) sexual organ. When through tantric configurations as well. In tantric initiation, for example, the guru is

yogic or erotico-mystical practice the same is turned upwards, however, it said at one point to encase himself, or his mouth, in the body or mouth of

is called the "chief mouth" or the "mouth of the yogini" and is identified his initiate.212 While there can be no doubt that an exchange of substance

with the womb of consciousness, the "circle of bliss from which the energy takes place between the guru and his disciple in tantric initiation, and that

of emission (visargasakti) flows forth as kundalini ... the matrix of creation, the guru symbolically enters into the body of his initiate, the precise nature

and the essence of Kaula doctrine."207 Similarly, the Manthanabbairava Tan- and location of this mouth-to-mouth transfer is difficult to pinpoint.

tra of the Western Transmission speaks of an upper and a lower yoni, lo- While phrases like "give me [your] mouth in [my] mouth"213 would lead

cated at the base and the upper end of the medial channel.208 In the midst one to imagine that the guru actually locks his mouth over that of his initi-

of so many mouths, let us recall here that each tantric system had its own ate, at least one other reading is possible. This latter reading would have

variations on the subtleties of the subtle body, making it impossible to gen- the guru entering his disciple's subtle body to place his mouth at the mouth

erate a single, unified account of the two upper branches of the split of the sankhini nadī and thereby open that "valve" to liberation in the

susumnā channel, the two "conch ducts," the two "tenth doors," the two body.2!4 Alternatively, the guru, having filled his mouth with the nectar

"western lingas," the two "yonis," the two "mouths of the yogini," etc. falling from his own sankhinī nädī, then transfers that nectar directly into

As in erotico-mystical practice, so too in alchemy, the practitioner must the mouth of his disciple. In many forms of tantric initiation, the guru is

reckon with the body's doors and mouths. So it is that the MBbT (9.19) described as coursing through the body of his initiate, in which he enters

states that the alchemist, once he has ingested edible mercury (together into union with the initiate's sakti and opens his channels to liberation.215 I

with meat and a woman's milk), must close the nine doors. More often than discuss important parallels between the language of initiation of this sort

actually swallowing said mercury, the alchemist will hold it in his mouth, and that of alchemical transmutation in the penultimate chapter of this

under his tongue or in a hollow he has cut into his palate. But this mercury, book.

held in the alchemist's mouth, is said to possess, at a certain stage of its In the physiology of the subtle body, the two nostrils, as well as the

preparation, a mouth or mouths of its own, with which it swallows, or more lunar and solar channels that issue in them, are termed putas. In hathayogic

literally, "has its mouthful (grāsa)" of the elements it causes to dissolve practice, these two putas are said to "swoon" or fall "lifeless" when the seal

within itself. In these intermediate phases of its preparation, mercury may that joins them together is pierced, by the opening of the medial su-

be either "mouthed" (sa-mukha) or "mouthless" (nir-mukba).209 sumna.216 Yet another yogic use of this notion may be found in the term

As we saw earlier in this chapter, hermetically sealing two mouths of a "interlocking crow's beak" (kākacancu-puta), which is a synonym for khecarī mudrā.217 two-chambered apparatus is also vital to the alchemist's craft. Here, while the term mudra may be employed, it is the term [sam]puta that more often As we know, the khecari mudra of hatha yoga is a technique by which the

comes into play. A puta is a capsule or envelope, generally having the form practitioner ensures himself the benefits of the fluid of immortality he has

of a closed crucible formed by placing two earthen plates face to face and culled through long yogic practice.218 While the alchemical sources first

sealing the joint between them with seven layers of mud-smeared cloth. treat kbecari as the power of flight attained by mercury when it reaches a perfected, stable, solid state, they also reserve a yogic application for the

258 259 Chapter Eight Homologous Structures of the Alchemical Body

term.219 Mercury is rendered kbecari by refining, stabilizing, and fortifying plicit in descriptions of the layout of the laboratory proper: radiating out- it, in the urddbvapatana and other apparatus, in red "female" sulfur. A pill ward from this divine center, the cardinal directions are identified with a of khecari mercury, held in the yogin's mouth during his sādhanas, catalyzes mandala of divinities as well as with the practical "rites of passage" (sam- the transmutation of his semen into nectar and greatly multiplies his pow- skäras) of the alchemist's art. Thus, the eastern wall of the structure is re- ers and longevity. This practice, called gutika bandha, is in fact the ultimate served for the storage of herbs and other plant matter: in the Hindu organ- end, the final samskara of the alchemist's art: the application of perfected ization of space, the east is the direction whose regent is Indra, the god mercury to his own body (śarīra yoga). Treated at length in hathayogic renowned for the theft of Soma, divine herb of immortality. To the south- and alchemical sources alike, this is the prime alchemical adjunct to yogic practice.220 We will return to these final ends of yoga and alchemy in east, the direction of Agni, Fire, are located the alchemist's distilling instru- ments. To the south, the direction of the dead and Yama, lord of the dead, chapter ten. the alchemist stores his oxidizing, "metal-killing" (lobamāra) chemicals. To the southwest, the direction of Nirrti, the dread goddess of disorder, are

5. Two Architectural Mesocosms arranged mortar, pestle, and other grinding and pulverizing instruments. Arrayed along the western wall, the region of Varuna, god of the waters,

Earlier in this chapter, I showed that the wells of poem and legend into are the alchemist's liquifying apparatus. To the northwest, the direction of Vãyu, Wind, are the bellows of the alchemist's furnace. The north is the which Nath Siddhas entered in their meditative practices were the mystic homologues of their own cranial vaults. In Ayurveda, the practice of "en- quarter of the god Kubera, lord of the Vidyädharas and possessor of fabu- lous riches and wealth. Here are located coloring (rañjana) agents, the rea- tering into the hut" (kutī praveśa) is cast as a return to the womb. Another son for this being that half the alchemist's art of aurifaction (or aurifiction, mesocosmic world of the same order, which the tantric practitioner may counterfeiting gold) lay in the tinting of base metals with amalgams of gold create and inhabit as a means to identifying body and universe, self and absolute, is the alchemical laboratory, appropriately called rasamandapa, and mercury.229 Lastly, the northeast, the direction identified with that form of Siva known as Iśāna, is consecrated to the use of mercury as agent the temple or pavilion of mercury.221 of transmutation (rasavedha). The alchemist who inaugurates or reconsecrates such a laboratory be- The layout of the alchemist's temple-laboratory not only follows the gins by tracing, in red sindūra (mercuric oxide, a mineral representation of logic of the Hindu organization of space, consecrating as it does each quar- the Goddess), upon a raised altar in the center of this structure,222 an intri- ter of the structure to an activity that corresponds to the regent of that cate mandala upon which all of the alchemical raw materials, processes, direction; it also follows the serial logic of the alchemical operations, the etc., of his craft are symbolically represented.223 At the heart of this mandala he establishes a phallic emblem of Siva composed of mercury (rasalinga) in samskäras, which he effects in transforming "raw" unrefined mercury into a "perfected" agent of transmutation and elixir of immortality. All things amalgam with either gold or sulfur,224 which he must propitiate daily if he Hindu begin in the east, the direction of the sunrise, where the day begins. is to hope to meet with success in his practice.225 This linga is set into a So too, the enumeration of the quadrants of the laboratory begins with this semicircular silver chasing (pitha), which may be taken as yet another min- direction, with the herbs that are the raw materials of the alchemist's work eral emblem of the Goddess's sexual organ.226 Also at the heart of this struc- and the lowest elements on the hierarchy of "metals,"230 and at that point ture, the alchemist installs a fire pit or basin (kunda) called the yonicakra at which the alchemical "life-cycle rites" begin. Following the east, this that will serve as a support for his ritual propitiation of the tantric god- orientational arrangement of the laboratory also reproduces the order of desses of his craft.227 Lastly the alchemist uses his rasalinga as a support the alchemical samskras, which begin with the distillation, grinding, and for his meditation (dbyana), projecting upon it an image of Siva Raseśvara killing of mercury and continue with its transformation of other metals (or Rasabhairava) in sexual union with the goddess Rasabhairavī (or through liquification, heating, and coloring, to culminate in their transmu- Rasāńkuśī).228 tation. That the union of these divine principles is a creative one is made ex- The principal role in this alchemical ordering of space and time is still

260 261 Chapter Eight Homologous Structures of the Alchemical Body

played by the alchemist who, holding the center stage-in which he, as a galdīp itself is in order, appearing as it does in a great number of Näth tantric practitioner, identifies himself with Siva in union with the Goddess Siddha legends. Like Ratansen in the Padmavat, Pūran Bhagat (Caurangi- in the rasalinga located at the center of his laboratory-orchestrates the näth) is also required to go to this kingdom. However, rather than going alchemical samskäras that will have, for their final result, the transubstanti- to conquer a woman as does Ratansen when he wins Padmavati, he is sent ation of the body and being of the alchemist himself. there by his guru Gorakhnäth to test his powers of yogic resistance against Just as we find a reasoned ordering of time and space in the alchemical the amorous advances of that country's queen, Sundran ("Beautiful").236 mesocosm that is the laboratory, so too do we encounter a wide array of Gopīcand, sent, in a similar way, to the "south country" by his guru Jālan- yogic representations of the subtle body, in which that body is projected dharnāth, falls into the clutches of a sorceress named Hīra ("Diamond"): upon a given mesocosm, or vice versa. A most graphic example of such a when he rejects her advances, she transforms him into a ram, and Jaland- projection is the early nineteenth-century Hamseśvarī (Our Lady of the harnāth has to come and rescue him. According to one source, Gopīcand's Cosmic Goose) temple (Hooghly district, West Bengal), whose six stories, own sister Candravali, prior to deciding to follow in her brother's footsteps labyrinthine floor plan, and maze of ladders reproduce, in explicit fashion, and become a yogini, is married the king of Simhala. We therefore appear the nadīs and cakras of the subtle body.231 A similar projection, on a far to be in the presence of a common motif which is, however, in the Gopi- more massive scale, is that of the pithas, the "footstool" pilgrimage sites cand legends, distributed between the Näth Siddha himself and his sis- of the Goddess which, since the earliest Buddhist Tantras, have at once ter.237 In one version of his legend, Püran Bhagat is said to be the disciple constituted geographic locations on the Indian subcontinent and centers of Matsyendranäth whose own Plantain Forest, it will be recalled, was lo- within the subtle body. Thus the four "original" pitbas of Buddhism-Kā- cated in one version of his legend, in Simhala! Matsyendra fares less well marūpa, Jālandhara, Pūrņagiri, and Uddiyāna-are identified with the four than do any of the junior Nath Siddhas of these accounts, but this is likely cakras of the Buddhist subtle body. This number comes to be expanded to due to his connections with the Yogini Kaula.238 In every case, we may 51 (the Sanskrit phonemes) in the KM and the Pithanirnaya, 68 in Nath conclude that this southern island-fortress is a mythic multiform of the sources, 134 in the Satsabasra Samhita, and so on.232 At least five sacred Plantain Forest already described above, at once a locus of lethal sensuality Šākta-Saiva sites-Kāmākhyā, Srisailam, Hinglāj Devī, Kedārnāth, and and rejuvenating yogic liberation. Here, then, is the allegorical ascent of Girnar-have alchemical traditions connected to them; to these may be the fortress of Simhala. In it, we will recognize a great number of land- added Siva's Mount Kailash, which is described as an alchemical wonder- land in the opening lines of the Rasārnava.233 scapes within the subtle body, as have been described throughout this

We have noted that the Nine Näths, rather than being historical figures, chapter.239

are in fact symbolizations of various elements of the universal macrocosm [Siva speaks] "I will tell you of the fortress of Simhala: the ascent is -the nine directions, including the zenith or center, and the subtle body of seven storeys.240 No one who has set his foot on the heavenly way (nine of the "ten doors," the principal nādīs, etc.).234 This is appropriate has returned as a living being. The fortress is as bent (bānk) as is your inasmuch as there was no medieval sect in India that was as fascinated, body: man, as you may see, is an image of it. It is not to be attained even obsessed, with the array of correspondences it discovered between the by fighting, through the use of force (hath);241 those who do attain it universal macrocosm and the microcosm of the subtle body. While the are those who have attained the marks [of yoga].24 Nath Siddhas mapped out these "static" homologies, in all their many- There are nine gates in that fortress and five constables patrol splendored detail, in such works as the Siddha Siddhanta Paddhati, Gorakb Bodh, and Prān Sankali, 235 nowhere are they presented more dynamically therein. It has a hidden tenth door:243 inaccessible is the ascent to this and the way is exceedingly crooked (banki).244 Only he who has and dramatically than in Muhammad Jayasi's Padmavat, in which the obtained the secret (bhed) and climbs like an ant can penetrate (bhedi) storming of the fort of Simhala is made into an explicit allegory for trans- that pass.245 At the base of the fortress is a deep pool (kund): in the formations within the hathayogic body. midst of this there is a path, as I declare to you ... 246 As the diver Before we enter into the heart of the matter, a word on Simhala or San- plunges into the ocean and only then does the pearl come to his

262 Chapter Eight

hand; so he who seeks out the door of heaven will make the ascent of NINE Simhala-dvīpa. The tenth gate is [as narrow and lofty as] the silhouette of a palm tree: he who has reversed his gaze (ulati disiti)247 can see it. He who The Dynamics of Transformation in Siddha Alchemy advances while restraining (bandi) his breath and his mind ... can go there ... 248 Then king [Ratansen] received the siddhi gutika.249 Thereupon he realized the siddhi[s] ... When Siva gave him the siddhi [guțikā], a great clamor arose as the yogins assailed the fortress. All the lotus maidens climbed up (on the roofs) to observe. In laying siege to the Simhala fortress, [it was as if] the yogins had raised a grove of medita- I. The "Work in Two Parts" tion huts [in its place].250 This dramatic depiction of the subtle body plunges us into the dynamics In an earlier chapter, I showed that the tantric universe was a unified sys- of yogic and alchemical transformation, which are to be the subject of the tem that oscillated between withdrawal (nivrtti) and return (pravrtti) on

following chapter. the part of a cosmic yogin, between effulgence (prakāśa) and reflection (vi- marsa) on the part of supreme consciousness, between emission and re- sorption, etc. In such a system, the yogic body becomes the stage for the return of the absolute from existence to essence through the descent and ascent of the kundalinī. As has been indicated, the absolute emanates into the manifest universe and human bodies as a means to enjoying its bound- less potential. The return, however, to unity and wholeness is, for those human manifestations of this emanatory dynamic, anything but natural, requiring as it does a forceful (batha) reversal (ulata) of what are, in mortal creatures, irreversible tendencies (entropy, aging, disease, death). Thus, while it is the case that the process of return is, from a divine or absolute perspective, internal to the process of emanation, it is nevertheless an ardu- ous task for the individual who would attempt to realize such through his own subtle body.' So it is that long before the much-glorified stage of yogic or alchemical reintegration, in which an emanated or differentiated mani- fest world implodes on itself, into a single essence-a drop, vibration, pho- neme, wave, photon, quantum, etc .- a great deal of unglorious prepara- tory work must be carried out. That work is the subject of this chapter. The human techniques for reversing the natural tendency (in an ema- nated universe) towards greater differentiation or entropy and for realizing for oneself the primordial unity of the absolute-with its concommitant bliss (bhoga), supernatural powers (siddhis), and bodily liberation (jivan- mukti)-are the stuff of hatha yoga. However, as the texts themselves tell us time and again, the human microcosm is intimately entwined with the alchemical and hathayogic mesocosm, with the latter two at once constitut-

263

264 265

Chapter Nine The Dynamics of Transformation

ing parallel and interpenetrating systems. Whence the classic statements it is significant that suta, "that which was generated," is one of the five of the RA: "As in metal, so in the body .... By means of the Work, a stable standard alchemical terms for mercury, Siva's seed, which lodged itself body is attained. Mercury and breath [control] are known as the Work in in the womb of the goddess earth, at the conclusion of the origin myth two parts (karmayogo dvidha)."2 This statement is echoed in Gorakhnath's of mercury.5 Amaraugha Prabodha (v. 5), which describes the twofold rājayoga as botani- In alchemy, as in other spheres of Hindu culture, the number attached cal and spiritual and the twofold hatha yoga as the "practice of breath and to the samskāras is eighteen, or more properly speaking, sixteen plus two.6 seed." The quest of the alchemist and that of the yogin are one and the The first sixteen samskäras prepare mercury and other elements for trans- same. This is the credo of Siddha alchemy. mutation, while the final two are the issue of all that has preceded: these Not only the ends, but also the means these two interpenetrating sys- are the transmutation of base metals into gold (called vedba, "penetration") tems employ for the realization of their final goal are also strikingly homol- and transubstantiation, the generation of an immortal body, called sarīra- ogous. In the last two chapters, I showed the content and structure of the yoga ("body Work"), bhaksana ("eating"), debavedha ("penetration of the two systems to be parallel and interpenetrating; in the present chapter, I body"), or simply sevana ("use").7 turn to their common dynamic. Stated in their simplest terms, hathayo- In yoga in general, and hatha yoga in particular, the sādhanas ("realiza- gic and alchemical techniques for immortality, termed sādhanas ("realiza- tions") are so many homologues of the alchemical samskāras. As outlined tions") and samskāras ("perfectionings") respectively, comprise four phases: in the introductory chapter, the term sädhana is derived, like siddhi ("real- purification, immobilization, reversal, and transformation. Quite often, ization," "supernatural power") and siddha ("realized individual," "super- the language employed to describe these sequential techniques is identical man") from the verb sadb, to "realize, accomplish, perfect." At least one in the two systems; and even when such is not the case, structural homolo- hathayogic source8 lists seven sadbanas together with their results. These gies abound, with references to digestion, consciousness-raising, flux, are purification (sodhana) through the six practices (satkarmāni); solidity phase changes (from the virtual to the actual), and a vertical dynamic per- (drdhatā) through the postures (asanas); immobility (sthairya) through the vasive in both systems. These sequences further constitute models of or yogic seals (mudras); composure (dhairya) through the retraction of the for a number of sacrificial, medical, and tantric practices and techniques, senses (pratyahara); lightness (laghava) through breath control (prāņā-

which I also outline here. yāma); direct perception (pratyaksa) through meditation (dbyāna); and im- Samskara is a Sanskrit term that has come to cover an extremely wide maculateness (nirlipta) and release (mukti) through total yogic integration semantic field in India. Derived from the same prefix + root (sam[s]-kr) as (samādbi). is the word Sanskrit itself, its literal sense is to "per-fect," to render whole or complete; more technical definitions include the "production of new qualities" and "that which makes a person or thing fit for a given func- 2. The Alchemical Samskāras tion."4 It is in this latter sense that the term samskāra plays a pivotal role in Hindu alchemy. The samskaras are those operations that render mercury If we are to demonstrate a more explicit connection between the alchemi- fit for transformation, for the production of new qualities in the mineral cal samskāras and the hathayogic sādhanas, we must go beyond these gen- world. From another perspective, the alchemical samskāras may be seen as eral observations and undertake a point-for-point comparison of the the exalted "rites of passage" of those humans who, applying the element graded practices that define the two disciplines. We begin our comparison mercury-first "generated" in the form of the "raw" semen of the god here with a summary of the alchemical samskäras, in which the sixteen stan- Siva-to metals and thence to their own bodies, transform themselves into dard samskāras are listed with Arabic numerals, supplementary operations equals of the gods (who, out of jealousy, asked Siva to adulterate mercury with roman numerals, and the effects of such operations with roman let- with impurities). As in the case of the human rites of passage, the principal ters. Following this summary, the balance of this chapter will consist of a aim of the alchemical samskāras is to purify a (mercurial) body that has commentary in which a series of yogic (and, where applicable, sacrificial, been tainted through the process of birth, of coming into existence. Here medical, and tantric) practices will be juxtaposed with homologous al-

266 267

Chapter Nine The Dynamics of Transformation

chemical operations-by which it will be shown that Siddha alchemy did twenty-six alchemical bandhas. Each of these bandhas has a specific medi- in fact constitute a "Work in two parts." cal application.13 I. Śodhana is the "(preliminary) purification" of mercury, its physical c. Mārana, the "killing" of mercury (or any metal) reduces it to a fine cleansing through washing, melting, marinating, and roasting it in vari- ash or oxide (bhasma), such that the human body is able to absorb it when ous preparations.9 it is taken in medical preparations. Metals other than mercury are generally II. Kşetrīkarana, "making (oneself master of) the field," is the purifi- killed, in preparation for internal use, by heating them together with iron cation of the human body, through special diets, emetics, etc., such that it pyrites and mercuric sulfide. When mercury is killed, it loses its fluidity, becomes capable of absorbing mercurial preparations. Although the inges- density, luster, and brilliance. Tantric alchemy attributes fantastic powers tion of such preparations constitutes the eighteenth and final samskāra, the of transmutation to said mercury, which it identifies as "killed ash" (mrta- preparation of the body, analogous in certain ways to the preliminary puri- bhasma) or "killed mercury" (mrtasūtaka).14 fication of mercury, must be undertaken well in advance of its ingestion. It 4. Utthāpana is the "resurrection" of swooned mercury, effected by is for this reason that I introduce it here. steaming it with alkalis, salts, and plant matter and by rubbing it in the I. Svedana is the "sweating," "steaming," or "fomentation" of mercury open air. Through "resurrection," mercury recovers the brilliance, etc. it in a water bath together with plant and mineral substances. This is the first had lost through mūrcchana. samskara, i.e., the first step in the chemical transformation of unrefined 5. Patana, the "sublimation" or "distillation" of mercury refers to the mercury into an agent of transmutation. As such, it differs from sodhana, three processes by which mercury is distilled upwards, downwards, or in which the mercury ore is first cleansed physically or mechanically, rather transversally.15 than purified chemically.10 6. Bodhana, "awakening," or rodhana, "countering, coagulation," are al- 2. Mardana is the "rubbing," "grinding," or "trituration" of steamed ternative terms for the sixth samskāra, by which mercury, which has be- mercury in a mortar, together with plant and acidic substances. come purged of its toxic content but also its strength through the preced- 3. Murcchana is the "fainting" or "swooning" of mercury, by which it ing operations, has its "virility" (virya) restored to it through irrigation in is ground in a mortar together with vegetable matter, until it loses its natu- a salt bath. This operation gives mercury a "mouth" (mukha) with which ral characteristics (luster, density, volatility, etc.) and form and becomes to absorb other elements in the samskāras that follow.16 purged of certain naturally occurring toxins, impurities, and defects.1 This 7. Niyamana, "regulation" or "restraint," reduces the motility of mer- is to be distinguished from another process of the same name. cury (which it has recovered through the previous samskāra), raising its III. Mürcchana as a supplementary operation takes mercury that has al- temperature of evaporation and rendering it lustrous in appearance. This ready been processed, either through the first eight samskāras or by other is done by soaking mercury in a bath of alkaline and herbal substances and means, and combines it with sulfur to produce a number of medical prepa- then steaming it.17 rations for Ayurvedic rasāyana or vajīkarana therapy.12 For reasons that will 8. Dīpana, "kindling" or "enflaming," further enhances mercury's po- be made clear later in this chapter, we pass directly from our discussion of tency and luster through steaming in an alkaline bath. This operation is "swooning" to that of two other vital transformations of mercury, known said to kindle mercury's desire to "consume" other metals.18 as "binding" and "killing." d. Taken as a group, the first eight samskāras serve to purify and detoxify a. As a result of the process of swooning, mercury becomes "swooned," mercury such that it may be used internally in the treatment of diseases.19 mūrcchita. It is here, however, that the great divide-between the medical use of mer- b. Bandhana is the "binding" or "fixation" of mercury, which, like curials (rogavada) and the use of the same for the ends of transmutational swooning, leaves mercury stable and thereby manipulable, in a state in (lobavāda) and elixir alchemy (debavāda)-is drawn. The realization of which it is not subject to evaporation, even when heated over fire. Left these higher, tantric aims requires eight additional samskāras, by means of unbound, mercury remains volatile whenever it is exposed to heat or sun- which mercury truly begins to behave like a living being, hungering after light. Once purified, it can be "fixed" via any one of the twenty-five or metals to consume, penetrate, and thereby transmute. The transubstantia-

268 269 Chapter Nine The Dynamics of Transformation

tion of the alchemist's own body, similarly penetrated by mercury, then virile and, by extension, virility) which takes over the body into which it follows automatically. enters, transforming human tissue into alchemical diamond or gold-gold Of the remaining samskāras, (9) [gagana]grāsa, (10) cārana, (11) gar- which, in the immortal words of the Brahmanas, "is immortality."26 In the bhadruti, and (12) bābyadruti are but phases in a single continuous process, following chapter, I provide further discussion of these final samskāras. which culminates in (13) jarana, "digestion" or "assimilation." These begin with the measuring out (mana) of a given quantity (a "seed," consisting of the calcinated ash, or the "essence," sattva) of mica or a metal for its 3. Purification consumption (grasa) by mercury. This is followed by the "chewing" (cār- ana, literally "coursing") of the mica or metal by said mercury, followed It is in the ancient sacrificial tradition that we find the earliest statements by either its internal (to the mass of mercury) or external processing and concerning the alternation between manifestation and nonmanifestation liquification (garbha-druti and babya-druti, respectively),21 with the end re- on the part of the absolute. This is described in terms of an alternation of sult being in that metal's total digestion or assimilation (jārana) into the divine generations and of the transformative, even reanimating proper- mercury. ties of fire. The primordial creator (Purusa-Prajäpati), having poured him- IV. Because it issues from a series of operations, jārana may be viewed self-indeed, poured his rasa-into the manifest world, lies broken and as the culmination of these alchemical samskāras as much as an operation dying. He can be restored to his prior unmanifest wholeness only through in itself. From this point onwards, it becomes quite artificial to distinguish the intervention of his son, Agni, who is the sacrificial fire. By piling a mercury from the metals to be transmuted. Once mercury has digested a fire altar (agnicayana) Agni reconstitutes the body of his father and thereby given metal, that metal no longer exists per se; rather, it is alchemically becomes known as "his father's father." It is through this alternation of activated mercury itself that will become transmuted into silver or gold.22 generations that it is possible for the divine to be at once one and many, 14. Rañjana, "tinting" or "coloration," involves the heating of mercury transcendent and manifest; and it is only through the offering, on the part with "seeds" of gold, silver, copper, sulfur, mica, and salt, such that mer- of a human sacrificer, of a major sacrifice that this alternation, this resur- cury takes on the natural colors of the minerals it has absorbed or swal- rection of the father, can be effected. Indeed, it is in this very same sacri- lowed.23 ficial context that one finds the earliest use of the verb samskr, to refer the 15. Sārana (literally, "flowing"), the "potentialization" of mercury in restoration to original wholeness of a god whose body has been identified preparation for transmutation, is effected by heating it in oil into which with what Jan Heesterman has termed the "broken world of sacrifice."27 molten "seeds" of metals, diamond, etc. are poured.24 Whenever such a sacrifice was to be offered in ancient India, the piece 16. In krāmana ("taking hold, "progression"), mercury is smeared with of ground upon which that sacrifice was to take place had to be prepared a mineral and herbal paste and heated in a puta such that it becomes ca- long in advance of the actual ritual offering. Preparation essentially in- pable, as a transmuting agent, of penetrating both metals and bodily volved sealing off a theretofore mundane parcel of land through a series of tissues.25 purificatory acts. Here, purification involved plowing and cultivating the V. The final two samskāras, (17) vedha ("transmutation") and (18) śarī- land, allowing it to lie fallow, and finally covering it with a layer of sand, rayoga ("transubstantiation"), together constitute the final end of the alche- which symbolized the semen of the "emptied" Prajapati. The language mist's work. As such, they are as much the issue of the sixteen prior sam- here was sexual: the raised altar was the womb (yoni) upon which the semen skāras as alchemical operations in themselves. (retas) of Prajāpati was spread, thus filling the void that was left when Praja- With these operations, mercury truly confounds itself with the metallic pati's body had first become dispersed.28 Only after the sacrificial ground or flesh-and-blood "bodies" in question, ultimately replacing them with a had been renewed in this way could it serve as the mesocosm for a sacrali- mercurial or alchemical body. If life is, as the Buddha said, to be regarded zation of cosmos, society, and the sacrificer, through the offering of the as a disease or a wound, then it is appropriate to liken mercury to a bealing sacrifice itself. virus (from vir; the Indo-European root denoting masculinity-as in semen It was not sufficient, however, that the sacrificial ground alone be puri-

270 271 Chapter Nine The Dynamics of Transformation

fied in preparation for the sacrifice. A similar sort of preparation had also of these five sequences-each a fortnight in duration and interspersed with to be effected upon the person of the sacrificer, the human pivot of the five-day rest periods-the rejuvenation therapy proper may begin. ritual act. Here, it was necessary that the human sacrificer, no doubt in Interfacing as it did with the therapeutic alchemy of Ayurveda, tantric imitation of the primal sacrificer, Purusa-Prajapati, renew his body in order alchemy makes a nearly identical use of the term ksetrīkarana, with its place that it be rendered fit to perform the sacrificial act.29 This act of preparing in the order of operations homologous to that found in Ayurveda. Simulta- the sacrificer's body, of transforming it from a mundane human body into neous to his preparation of the mercury he will eventually ingest, the alche- a body worthy of communing with the divine through sacrifice, is dīksā mist must also prime his body, in order that it be capable of absorbing that (literally "habilitation"), a term generally translated as "initiation." Here, potent elixir. Here, ksetrīkarana means observing, over a fifteen-day pe- the mechanics of initiation entail the same dynamic as the preparation of riod, a strict purificatory diet which effects, through the same five treat- the sacrificial ground: the purification of a mundane body transforms it ments as those employed in rejuvenation therapy, the evacuation of every into a sacrificial body. As with the preparation of the sacrificial ground, the bodily impurity with which the mercury might abreact.35 As a result of this preparation of the sacrificer himself is described in terms of sexual repro- operation, the body becomes a field in which a mercurial seed (rasa-bīja), duction and animal gestation. Sequestered within the initiation hut, the when ingested, will germinate. The fruit this seed will produce is an im- sacrificer, "cooked" through the inner heat of his austerities (tapas) and mortal, alchemical body. While apparently agricultural, the language here, the external heat of burning fires placed in the hut, symbolically sheds his too, is sexual. The first three samskäras of the human rites of passage, which mundane body. An embryo of his "new" sacrificial body takes form, incu- effect the purification of the womb, are called the "perfectionings of the bates, and is born out of the "womb" of the hut, three days later.30 field," the ksetra-samskāras, by which the female womb is prepared to re- The symbolism of this homologous preparation through purification, ceive the seed (bīja) that is male semen.36 of both sacrificial ground and sacrificial body, is directly appropriated by Here, the planting of a mercurial seed in the body also reminds us of the three elements of Siddha practice: hatha yoga, alchemy, and tantric rit- tantric initiation (diksa), which may be performed either with or without ual. This first groundbreaking phase is called ksetrīkarana, "mastering, pre- "seed" (sabīja or nirbīja).37 As the MBhT describes it, the "seed" is the con- paring the field" in the first two systems, and bhūtasuddhi, "purification of densed mantra-identified with the semen of Siva, who is himself identi- the elements" in the third. This metaphor of the tilled field is altogether fied with the guru here-that the guru plants in the initiate's body. Indeed, natural within an agrarian society and is not original to these three tradi- no mantra can be effective without a "seed," and no tantric deity exists tions. Classical Samkhya calls the body-mind complex the field (ksetra) and independent of his or her seed mantra (bija).38 Like Prajāpati and Agni in the soul the farmer (ksetrajna, literally, "knower of the field") who tends Vedic sacrifice, the relationship between the guru who plants the seed and said field without himself being affected by it.31 In Hinayāna Buddhism, the disciple in whose body he plants it is one of an alternation of genera- the notion of the transfer of merit is explained through the image of the tions. So too, the alchemist whose corporeal field has been properly pre- assembled saints, who constitute an excellent field (khettupama) in which pared gives birth to a new, immortal self, out of the old, through the mer- meritorious acts may be sown and thereafter bear fruit for the benefit of curial seed he has planted there.39 Even if the body so produced is most others.32 often called a golden, adamantine, or realized body (svarna-deha, vajra- It is in the medical tradition that the theories and techniques pertaining deba, siddba-deha), it is in fact a mercurial body, an alchemical body. to the preparation of the body as field are most fully developed. In its re- Like the Ayurvedic physician and the alchemist, the hatha yogin must juvenation therapy, the term ksetrīkarana refers to the preparation of the also effect a preliminary cleansing (sodhana) of his body. This he does body for the medicines that will be absorbed in the treatment per se.33 This through a series of operations known as bhūtasuddhi, the "purification of class of treatments-called samsodhana cikitsā, "purificatory medicine" or the elements," or the satkarmani, the "six practices," in which the inner pañcakarmāņi, "the five treatments"-combines such clinical treatments as contours of his body are purified with air, water, and fire.40 Like the five emetics, purgatives, sudation, etc. as means to purifying the body by void- treatments of Äyurveda, the majority of these practices ranging from nose ing it of the elements responsible for its humoral imbalance.34 At the end wash (neti) to enema (vasti), flush out the body's physical impurities.4 Such

272 Chapter Nine 273 The Dynamics of Transformation practices, which cleanse the channels of the subtle body, channels through carried out, the element mercury must be primed, purified, and conse- which the five vital breaths must freely circulate, are already anticipated in crated before further operations may be undertaken. the preparation of the fire altar mentioned above: the furrows plowed into It is in this context that the term employed for the alchemical trans- the site create the channels through which the vital airs circulate in the formations, samskara, takes on its greatest fullness of meaning: "Samskāra: body of Prajāpati, the sacrifical Man." Given the task he has set himself- those acts and rites that impart fitness. Fitness is of two kinds. It arises to reintegrate bodily microcosm with divine macrocosm-these prelimi- from the removal of taints or by the generation of fresh qualities."48 naries are vital to the yogin's success in all that follows. From the outset, Of the classic enumeration of the eighteen alchemical samskāras, the the intrinsically gross human body cannot be charged with the slightest first eight concern nothing more or less than the preliminary prepara- impurity as such would, like a speck of dust on a mirror, skew the perfect tion of the element mercury for its eventual application to other minerals. interface, reflection, and realization of one world in another.13 Of these, the first two-svedana and mardana, "sweating" and "rubbing"- The term bhūtasuddhi has a much more extended application in the deal most directly with mercury's physical or mechanical purification.19 realm of tantrism, where it also forms an important element of worship The "sweating" of mercury immediately reminds us of the initiation of the and initiation." In tantric ritual, bhūtasuddhi refers to the preliminary puri- Vedic sacrificer and the kuti-pravesa treatment of Ayurvedic rejuvenation fication of the divinities residing in each of the five elements (bhūtas) that therapy. It is by sweating off, purging oneself of one's old body that it be- make up the body.45 Here, bhūtasuddbi is the purification of both a meso- comes possible to take on a new one. The same is the case in hatha yoga. It cosmic worship site-once again referred to as a "field," ksetra-and the is only after his effecting of the "six acts" that the practitioner begins to microcosmic body of the worshipper himself (when the two are not iden- truly transform his body-by sweating, which is viewed as the first stage tified),46 a situation that mirrors that of the twofold Vedic preparation of in the process of yogic transformation. Here, he induces perspiration sacrificial ground and the sacrificer's person. Both are dessicated, "blown through a number of techniques of breath control (prānāyāma), techniques out," and burned up before being cleansed with water and flooded with which culminate in diaphragmatic retention (kumbhaka). The sweat that is "nectar," processes which, identified with the dissolution of the mundane voided through the pores (literally, "hair wells," roma-küpa) has in fact been self, constitute the first step towards the creation of a new divinized self. forced out through the seventy-two thousand nadis, thereby purging them Here, the lower elements of earth, water, fire, and air, are successively im- of all impurity.50 ploded into their higher emanates, until there only remains the most sub- The parallel with the alchemical samskāras appears to be deliberate here, lime element on the pentadic hierarchy. This is ether, the empty space left since immediately following this, the yogin is instructed to rub (mardana) by this dissolution, within which the tantric practitioner will establish, his body with the perspiration that has been so produced. The combination through visualization techniques and the planting of seed mantras, a new of these two processes renders the body-not unlike mercury at the con- world at the center of which he will construct the body of that divinity with clusion of the first two alchemical samskaras-firm and light.51 whom he will come to identify himself.+7 The tantric worshipper or initiate who has transformed his own being through these processes becomes capable, in turn, of transforming other 4. Immobilization: Murcchana, Bandhana, and Mārana beings, indeed, the entire universe, through his limitless powers. Such is also the case with mercury in the alchemical context. In its mundane form, The second verse of Patañjali's Yogasūtras defines yoga as "preventing as it is found in ores and its various compound forms, mercury is incapable thought from going around in circles." One may indeed consider the entire of transforming other mineral substances in any way whatsoever and is history of yogic theory and practice to be a footnote on this formula.52 capable only of poisoning the foolish alchemist who would attempt to in- Reduced to its simplest terms, yoga ("yoking") is concerned with impeding gest it. Like the bodies of the Ayurvedic patient, batha yogin, and tantric movement, with the immobilization of all that is mobile within the body. practitioner, and like the sites upon which Vedic and tantric rituals are This is a point that the HYP makes clear in its opening chapter: "The pos-

274 275 Chapter Nine The Dynamics of Transformation

tures, the various [methods of breath control leading to] kumbhaka[s], the HYP promises the same results, but takes rasa to mean semen, the immobi- practices called mudra, then the practices concentrating on [hearing] the lization and retention of which are paramount to the hathayogic discipline, nāda: this is the sequence [to be observed] in hatha [yoga]."53 One first im- as witnessed in a poem attributed to Gopicand:56 "Steady goes the breath, mobilizes the body through the postures; next, one immobilizes the breaths and the mind is steady, steady goes the mind, the semen. Steady the semen, through diaphragmatic retention; one then immobilizes the seed through and the body is steady, that's what Gopicand is sayin'." the "seals"; and finally one immobilizes the mind through concentration A number of other references further elaborate on the interchangeabil- on the subtle inner reverberation of the phonemes. ity of mind, breath, and rasa in the techniques evoked in the hathayogic and The theory here is simple: stop this, that stops. But the practice is any- alchemical sources. The HYP equates the immobilized mind with bound or thing but simple as anyone who has attempted to maintain a yogic posture, fixed mercury and states that the former results in the immobilization of sit still, or simply stop thinking for any length of time knows all too well. both breath and body and thereby bodily immortality. In like fashion, the What a difficult, even heroic undertaking the immobilization of the body RA emphasizes the necessity of unwavering mental concentration for suc- constitutes, yet what fantastic results it yields! For immobilization leads to cess in the performance of alchemical operations.57 Gorakhnāth, with the reversal, reversal to transformation, and transformation is tantamount to directness we have come to expect of his poetry of yogic experience, states bodily immortality and, precisely, to the supernatural ability to transform, the matter clearly and succinctly: "Penis in the vulva's mouth, mercury in reverse, or immobilize whatever one desires in the physical world (siddhi). the mouth of fire; he who can retain these [semen and mercury], him I call Immobilization or stabilization, in the form of swooning, binding (or fix- my guru."58 ation), and killing, constitute the "hump" that the Siddha alchemist must Here, it is the same image, of immobilized rasa-that is at once cal- get over in order for his goals to be realized. All that follows does so more cinated mercury and semen-that comes to be employed, in a wide array or less automatically, as if a critical mass has been reached and a chain of applications, to describe the parallel yet interpenetrating processes of reaction triggered-whence the most frequently recurring passage, in the Siddha alchemy. Although swooning is the sole member of the triad-of entire alchemical tradition. murcchana, bandhana, and marana-to actually constitute an alchemical A single summary passage, concerning the matters of immobilization, samskara per se (the latter two being the effects of a number of combined stability, death, resurrection, and the power of flight, appears, in nearly operations), all three are nevertheless central to the alchemist's craft. Ac- identical form, in all of the alchemical classics-the Rasārnava, Rasabrdaya cording to a list provided in the RA, all seven of the standard forms of Tantra, Rasendra Mangala, Bhūtiprakaraņa, Rasendracūdāmani, and Rasarat- alchemical mercury produced through the samskaras are either swooned nasamucchaya.54 More remarkable is the fact that this exact same verse, at (mürcchita), bound (baddha), or killed (mrta).59 We now turn to a closer least in the form in which it is found in the RA, also occurs in the fifteenth- examination of this triad, within the Work in two parts. century HYP of Svatmarāma. In this latter case, however, Svātmarāma is able to employ "alchemical" terminology to make a yogic point. That is, the same verse may be read on two different registers, the one alchemi- a. Mūrcchana cal and the other yogic. In the context of the "Work in two parts" and the It will be recalled that alchemical samskara known as "swooning," following vital matter of immobilization, however, the two perspectives ultimately directly upon the "sweating" and "rubbing" of mercury, causes the liquid become fused into one. metal to lose its luster, lightness, volatility, etc., which are immediately re- The verse in question reads, in the RA and the HYP, as follows: stored to it in the course of a fourth operation, called "resurrection." Re- "Swooned, rasa, like the breath, drives away diseases, killed it revives itself, calling the intimate connection between immobilizing the mind, breath, bound it affords the power of flight."55 In the alchemical RA, the term rasa and rasa in the Work in two parts, we should expect to find hathayogic refers to swooned, bound, and killed mercury, the ingestion of which, murcchana playing an analogous role. Hathayogic "swooning" is in fact a in tandem with the practice of breath control, renders the practitioner direct result of kumbhaka, of "potlike" diaphragmatic breath retention, the healthy, immortal, and possessed of the power of flight. The hathayogic methods for the realization of which it is now time to describe.

276 277 Chapter Nine The Dynamics of Transformation

The yogin first draws the subtle breath (prana) in through the left nostril are the: (1) mūla bandba, ("root lock"), an inner contraction of the anus and thereby into the lunar channel. Having retained it for as long as pos- which draws the downward-tending apana breath upward through the me- sible in the abdomen, he exhales it, via the solar channel and through the dial channel; (2) uddiyāna bandha ("the lock of the upward-flying [bird]"), right nostril. He then inhales through the right nostril, retains the breath a contraction of the abdomen which, by emptying the lungs, drives them in the abdomen as before, and releases it, via the lunar channel, out the and the diaphragm up into the upper thorax and causes the prāna to "fly left nostril. By continuing this process, pumping the outer nādīs "like a up" through the medial channel into the cranial vault;63 and (3) jālamdbara blacksmith's bellows," the yogin's diaphragm will, at a certain point, remain bandba ("the lock of the net bearer"), a contraction of the throat by means filled with air, the pressure of which will force open the orifice (known of which the yogin seals off his head from his torso and constricts the net- as the "door of brahman," the brabmadvara) of the medial channel.60 The work (jala) of subtle channels and supports, thereby arresting the down- subsequent inrush of air into the medial channel causes the two peripheral ward flow of nectar that has accumulated in the cranial vault.64 The con- channels to empty. Deflated in this way, they are called "swooned" (mūrc- joined aim of the three bandbas is to gradually restrict the field in which cha), and indeed, one of the eight types of kumbhaka is itself called mūrcchā, the volatile breath, seed, and mind may move. First forcing them up out of "because it causes the [normally volatile] mind to swoon," i.e., to become the abdomen, they "lock" them into the torso; they next "contract" them one-pointed in its concentration.61 inside the neck and head; and lastly, they "bind" them there. It is not the swooning of the two peripheral channels that itself "drives The alchemical parallel to these hathayogic techniques is introduced by away diseases," as HYP 4.27 states. It is rather the effect of their empty- a verse from the HYP: "Instability is a natural characteristic of mercury ing-i.e., the opening and filling of the medial channel-that is essential, and of the mind. When mercury is bound, when the mind is bound, who as this becomes the path by which the kundalini surges upwards, carrying in this world does not become realized (siddbyati)?"65 In alchemy per se, with her the life force, seed, and "swooned" mind into the cranial vault. there are twenty-six bandhas which effect the binding (bandhana), fixation, Just as the swooning of mercury is, in the alchemical context, a necessary or immobilization of mercury by altering its physical and chemical compo- intermediate step to the transformation of metals and the human body, so sition and behavior. Through the bandhas, mercury takes on the consis- too in hatha yoga, swooning affords the practitioner mastery over disease tency of a gel, paste, or solid powder, in which states both its temperature and death. The parallel is explained in these very terms by Brahmānanda, of evaporation and its powers as a transmuting agent are augmented. Yok- in his commentary to this verse: "Through the use of certain herbs, mer- ing the alchemical discipline to that of hatha yoga is an evocative descrip- cury, which is inclined to be volatile, becomes swooned (mürcchita) in the tion of a general principle for the fixation of mercury: "In the midst of the same way that the breath becomes swooned in the culmination of kum- Ganges and Yamuna rivers there is a rāksasa named Application (prayoga). bhaka, when one has exhaled and stopped inhaling."62 In his presence mercury is immediately bound." Here, Ganges and Yamuna stand for the bright and dark lunar fortnights, respectively, mediated by the full moon night. The nocturnal "application" in question consists of b. Bandhana culling a number of powerful herbs which, on the full moon night, stabilize The hathayogic texts recommend a combination of postures, together with mercury.66 Here again, we are reminded of the Work in two parts, as well a number of respiratory and "hydraulic" techniques, for the immobili- as of the yogic immobilization of the breaths in the lunar and solar (Ganges zation of the breaths and the diaphragmatic retention that trigger the rise and Yamuna) channels, which fill out the subtle moon located in the cra- of the kundalini and all that follows. This body of hydraulic techniques is nial vault. generally subdivided into "hermetic seals" (mudrās such as the vajrolī and The same cranial vault is the locus for another sort of bonding of yogic khecari) and contractions or "locks" (bandbas). These sources describe three and alchemical method. Here we are referring to the technique known as principal locks which, effected at the levels of the abdomen, thorax, and gutikā bandha, in which the yogin fixes a mercurial pill (gutikā) in a recess head, work hydraulically to effect internal changes in pressure, such that of the palate as a means to catalyzing the effects of his yogic practices. This breath and seed become immobilized or begin to be drawn upward. These technique will be described at greater length in the next chapter.67

278 279 Chapter Nine The Dynamics of Transformation

Apart from these specialized contexts, both the yogic and alchemical about its true cosmic nature, falls into the state of an individual sub- traditions employ metaphorical images of binding or snaring to describe ject with limited knowledge. This means that the nature of Māyā is the rigors of their respective and complementary tasks. So, for example, twofold: limitative and dispersive.72 the RA (2.90) praises the mantra called rasānkuśa, the "elephant goad," in the following terms. "One who attempts to obtain mercury without knowl- Ropes and knots; binding, loosing, and cutting; it is in these terms that edge of the elephant goad [mantra] is like one who attempts to mount an India has perennially portrayed the conundrum of existence. Creatures enraged elephant in the jungle without a goad. He is beset [with tribula- (paśus) are bound to a phenomenal world that is a tightly stitched net of tions]." Similarly, in one of his bānīs, Gorakhnāth states that by immobiliz- magic or a veil of mayā; and when this life ends, it is the noose (pāsa) carried ing the breaths, one may bind (bandhilai) the elephant of the mind and by Death himself that ensnares them. In such a world of bondage, what bring it into its pen.68 As we have already seen, the fixing of mercury is also term could better describe liberation and salvation than moksa-which referred to as the clipping of its wings (paksaccheda), by which it is pre- precisely means the loosing or releasing of the knots and webs and snares vented from volatizing and "flying upwards" (evaporating). Similarly, in that fetter all of creation?73 yogic practice, it is crucial that the "cosmic goose" (hamsa) of the breaths Maya need not, however, be viewed as a net of illusion spread to obscure and vital energy be tethered so as not to leave the body.69 the numinous absolute. It can also be seen as the self-actualization of the In India, all of these images-of binding, ensnaring, tethering, and the divine creative impulse, as the "measuring out" (mā) or manifestation of like-are as old as sympathetic magic and sorcery itself. As far back as the pure consciousness, which is free to bind itself if it so chooses. In this, the Rg Veda, we meet gods like Indra who, through the use of magical, invisible tantric perspective, the phenomenal world, rather than being a straitjacket cords, defeat their demonic enemies in battle. Indra's greatest victory is to the soul, becomes a field of play for the realized (siddha) individual. By won when he "envelops the enveloper," the archdemon Vrtra (the "Enve- extension, it is no longer moksa or release from conditioned existence that loper"), who has trapped the world in his coils of drought and darkness. is the Siddha's goal, but rather liberation in the body (jivanmukti), in which Once the binder has been bound, the entire universe is set free, released the individual experiences the world, for himself, in the same way as does from its bondage.70 Indra's feat is reproduced, with variations, by Gorakh- the divine absolute.74 Once one enters into the universe of the Siddhas, the nāth, who, according to a relatively late Nepali myth, binds nine "serpent veil of maya becomes as if turned in on itself. How does one escape the clouds." This provokes a drought which only ends when his guru Matsyen- trammels of existence? By binding the bondsman, cutting the cords, burn- dranāth comes to Nepal from Assam, and Gorakh rises to salute him.71 ing the burner-and even consuming Death, the great Eater. Once bound, In the case of the Vedic and post-Vedic Indra, this god seemingly saves normally volatile mercury and breath afford normally earthbound humans the world in order to ensnare it in his own magical web, for Indra's net of the power of flight. The medial channel, when opened, becomes "the eater magic (indrajāla) is also a veil of māyā for those lacking the insight to dis- of Death," and the upward surge of energy that courses through it cuts cern, behind a world of appearances, what is really real. All of the gods, in through the three knots (called granthis) which are the sole remaining ob- fact, impose their maya upon the world, if only to differentiate themselves stacles to the yogin's immortality and freedom.75 When the flighty gazelle from humans who, were they to penetrate the ultimate reality behind of the mind is at last snared in net of the unstruck sound, individual con- the phenomenal worlds, might otherwise stop offering them sacrifices. In sciousness becomes dissolved in the pure universal consciousness of the more philosophical terms, divine.76 The Siddha not only transcends the laws of nature, he also breaks out the chief cause of [the individual soul's] bondage in the objective of the bonds of the human condition: "the yogin is unbindable (abādhya) world is ... Maya, usually considered to be an aspect of the God- by any incarnate creature. He can neither be bound by [the noose of] (dess)'s eternal Sakti. It finds itself ... in the traditional series of cate- Death, nor by karma." "The great binding (mahabandha) [affords] release gories. The Maya is that power by which the Supreme Being veils (vimocana) from the noose of death (kālapāśa)."77 Like the god Indra of yore, itself, so that the jīva [the individual soul], enveloped by ignorance the yogin or alchemist binds in order to be released from bondage; now,

180 281 Chapter Nine The Dynamics of Transformation

however, rather than doing battle with enemies from without, he is able tion; the mudras are great hoop earrings which, placed in the thick of the to domesticate Death, time, illusion-nay, human finitude-without ever ears at initiation, open a subtle channel vital to yogic practice; and the having to leave the inner landscape of his subtle body. kantba is the patchwork tunic traditionally worn by the Nath Siddhas.84 Herein lies the critical importance of the yogic and alchemical bandhas: they bind the mechanism of bondage itself, hobbling, binding with its own noose a Time (Kala) which is identified with death (kala) before actually c. Mārana turning it back on itself. None of the trammels of existence can fetter the Since at least the time of the epics, the final of the four idealized stages of Siddha at play in the world, least of all the ultimate sorrow, the sorrow of life observed by males in India has been termed sannyāsa, because it is death. For whenever Death tries to catch a Siddha in his noose, the latter marked by the act of "laying together" the sacrificial fires that had defined either slips away, thus cheating Death ("skewing time": kālavañcana)78 or one's social existence. Abandoning one's sacrificial fires is tantamount to actually beats him up. This is a theme found in the legend of Gopīcand, in abandoning the world: henceforth, although one remains in the world, one which Hadipa (Jalandharanāth), upon learning that his young disciple has is no longer a part of it. been taken by Yama's minions to the world of the dead, thrashes them and Of vital importance to the yogic tradition is the fact that the sacrificial their master so thoroughly that they promise to never meddle with the fires in question are gathered together within one's body. There, they serve Nath Siddhas again.79 Mayana, Gopīcand's sorceress mother, following her both as a cremation pyre-by which the now-obsolete mundane, social initiation by Gorakhnath, holds similar powers over Yama: "If she should body is shown to have truly died to the world-and, in the postcrematory die in the day-time, he [Gorakh] would not let the sun go, but would existence of the sannyasin (the "renouncer"), as the seat of sacrifice, which bind him down,-if she should die at home, he would not let Yama go, but has now been internalized. It is here, in the inner fires of tapas, which fuel bind him down,-if she should die of a cut from a flat sword (kha'a = a the offerings of one's vital breaths in the inner sacrifice known as the pra- large sacrificial knife), he would bind the goddess Candi [who bears nāgnibotra, that the practice of yoga very likely had its theoretical origins.85 such a weapon] down-Mayana would [out-]survive even the sun and the In the experience of the renouncer, the internal processes of yoga do moon."80 nothing less than sustain him in a life beyond death. It is to this notion In a later episode of the same legend cycle, Gopīcand has Hādipa bound that the HYP passage we evoked at the beginning of this section alludes: in chains and buried beneath a stable (or at the bottom of a well), under "killed, rasa, like breath, revives itself."86 It is through this paradoxical pro- several feet of horse manure. These Hādipa tears away by means of a single cess that the hatha yogin and alchemist ultimately realize the bodily immor- mantra. The chain on his hand becomes transformed into a rosary of beads; tality that is the final end of their practice. The language both traditions the heavy stone placed upon his chest becomes his yogic garb; and the rope employ to describe this process, through which the practitioner, once with which he was bound becomes his loincloth.81 'slain" (mrta), is able to revive himself by tricking, even killing Death, car- The regalia and sectarian markings worn by the Näth Siddhas are, in ries us to the very heart of Siddha mysticism: you have to first die to be- fact, all explicitly symbolic of those various elements of their yogic practice come immortal. This is the import of one of Gorakhnath's most celebrated which afford them their fabled mastery over the processes of aging and banīs: "Die yogi die, dying is sweet [when you] die that death by which death. As such, they constitute a mesocosm, a model of and for the micro- Gorakh, in dying, gained his vision [of the Absolute, immortality]."87 An cosm of the subtle body and the macrocosm of the universe. In the words alchemical echo is provided in the Padmavat: "The Siddha's immortal body of Bhartrhari, "the form of the Jog [ yoga] is the ear-ring, the patched quilt, is like mercury. You can break it down, you can kill it, but you can't make the wallet, the staff and the horn, the sound of which is emitted in the it die."88 Universe."82 Here, the horn is the singnad, a piece of gazelle horn into What is it that dies? It is the gross body, a husk that is to be cast off like which the Nath Siddhas blow to produce the nada, the silent sound that the slough of a snake. How is this body made to die, in order that the serves to bind and control the wavering mind.83 In the legend of Hādipā, golden, adamantine, or siddha body may emerge? As in the processes dis- the japamālā is a rosary of rudrāksa beads used as an aid in mantric recita- cussed above, breath control plays an important role here. So too, however,

28z 283 Chapter Nine The Dynamics of Transformation

does the generation of yogic heat, of the fire of yoga (yogāgni) that burns three primal sacrificial elements of fire, fluid, and air and of their homo- up the fire of time or death (kalagni). In the former case, it is once again logues in the bodily microcosm and the alchemical mesocosm. But what the two peripheral nadīs whose emptying is likened to a death that gener- happens when these essences, these dosas, these gunas, actually combine? ates new life (the filling of the susumna). In the seal called the mabavedha What is the composite or compound that is produced? Since the time of (the "great penetration"), "the breath, overflowing the two putas ... the Vedas, the ringing answer to this question has been ashes.94 quickly bursts [into the medial channel]. The union (sambandha) of moon, Here, we should note a certain symmetry between the rasa of Vedic sun, and fire which is to occur surely results in immortality. When the death- sacrifice and the rasa of alchemical practice. In the former case, the raw like state (mrtāvastha) arises, how can there be fear of death?"89 material of the sacrificial oblation has no true ontic being until it has been Passing through a state of death to bodily immortality is most especially cooked, refined, transformed, through exposure to fire. In this context, ev- effected through the mediation of fire-here, the inner fire in which the ery natural and cultural process becomes a matter of cooking: in the aging renouncer has immolated his mundane body once he has "laid together" process, one is said to be "cooked by time."95 Elsewhere, the advance into his sacrificial fires. This image is carried over into the hathayogic system, the Indian subcontinent by Vedic culture was led by Agni, the sacrificial in which the fires of yoga (also called the fire of the absolute, the brab- fire that was carried in a firepot, at the vanguard of "civilization"; it is this mägni), kindled at the base of the subtle body, renders the medial channel same sacrificial fire that restores and perfects (samskr) the rasa of Prajapati a "cremation ground" (śmasana).90 The hallmark of any cremation ground that flowed from him to create the world. So too, initiation, cremation, is, of course, its ashes, and it is with ashes that every Saiva renouncer, from and the passage into the renunciant stage of life are all so many cases of time immemorial, has smeared his body. Internal fires and external ashes cooking "to a turn" that which would otherwise remain unaltered, in a are brought together, in the hathayogic context, in two variations on the raw, untamed, undomesticated, even uncivilized form.96 Thus the prince- vajrolī mudrā, called the sabajolī and amarolī mudrās. In the former, the part- turned-yogi Gopīcand can state, "I used to be an unfired pot, thrown ners smear their bodies with a mixture of cow-dung ash and water follow- whichever way [the wheel] turned. When I was made a Jogi my guru did ing vajroli; in the latter it is a mixture of "the nectar that flows from the the firing."97 moon" and ash that one smears over one's body.'1 What are the implica- In the exemplary cooking process that is the sacrifice, the most subtle tions of this externalization of an internal process, this wearing of one's remainder of the process of combustion, beyond the cooked "leftovers" yoga on one's sleeve? The answer to this question is quite nearly as old as consumed by humans and the smoke and aroma of the cooked offering the Veda itself. enjoyed by the gods, are ashes, bhasma[n] (from bhas, "consume," but also "shine"). Ashes are the shining remains of what has been consumed; they are the solid essence of the combustion of fluid oblation by purificatory 5. Ashes to Nectar fire. As such, ash is rasa in its optimal form-which is exactly what the alchemical tradition tells us in its myth of the origin of mercury (rasa). Its I have already mentioned the primacy given by the RA to the eating of original cosmic matter being concealed by the "accidents" (kañcukas) of its killed or oxidized mercury (mrtasūtaka, also called "dead ash," mrtabbasma) outward form, native mercury cannot, in and of itself, afford immortality. for realizing immortality.92 I have further alluded to the process of diges- It is only after these superfluous accretions have been dispersed, by reduc- tion (jarana) by which such is obtained and to the fabulous properties pos- ing mercury to ashes through the alchemical samskaras, that its inherent sessed by said mercury. "Killed" mercury, itself "unresurrectable" (nir- basic properties can be actualized. Ash, bhasma, is the supreme manifesta- uttha)-that is, incapable of returning to its prior, native state-is now tion of primal matter. It therefore follows that the reduction of mercury, capable of reviving other "killed" metals.93 What is it that makes rasa in the rasa of the absolute, to ash is tantamount to the recovery of a primor- its ashen (bhasmībbūta) state the most optimum form of mercury, both for dial perfection, of the absolute before its fall into nature, into manifesta- transmutation and transubstantiation? tion. Preternatural (prior to nature) its powers are therefore supernatural Throughout this book, we have emphasized the complementarity of the (transcending nature). At the same time they are, in a cyclic universe, a

284 285

Chapter Nine The Dynamics of Transformation

reminder of the ultimate destiny of the universe when all things will be termed a "theology of ashes," at the conclusion of the important myth of reduced to dross (the serpent Sesa, "Remains," whose body is composed of Siva in the Pine Forest. Here, after the phallic god has brought a group of ash) and only the pure gold remain.98 Herein lies the root of the Hindu uncomprehending Vedic sages to heel, he explains to them the importance calcinatory gnosis. of ashes. It is in the context of early Saivism that a true "cult" of ashes first be- comes incorporated into the Hindu tradition. Ashes, called either bhasma I am Agni joined with Soma ... The supreme purification of the uni-

or vibhüti ("omnipresent," a term synonymous with siddhi, or the eight sid- verse is to be accomplished with ashes; I fortify my seed with ashes

dbis taken as an aggregate), have since at least the time of the Pāsupatas and sprinkle creatures with it ... By means of ashes, my seed, one is

been integral to Saiva initiation and worship. Initiation involves bathing in released from all sin ... Ashes are known as my seed, and I bear my

ashes (bhasma-snāna), while the worship of Bhairava entails smearing cow- own seed upon my body ... Let a man smear his body until it is

dung ash on the forehead-in place of the normal blood-red sindūra-as pale with ashes ... Then he attains the status of Lord of the Host

well as consuming a pinch of said ash.99 (gāņapatyam) ... and grasps the supreme ambrosia.102

In so doing, humans are, of course, merely imitating the supreme god Siva's subsequent initiation of the Vedic sages culminates in an ashen himself, whose own dark body is made luminous by the patina of iridescent bath, a bhasma-snāna, after the model of Pāsupata initiation rites. The role ashes with which he besmears himself and the stuff of whose body is in fact of ashes in initiation has a much more ancient pedigree, however, as at- ashen. Tradition in fact holds that Siva first took to wearing ashes follow- tested in an important myth from the Mahabbarata. This is the epic myth ing his incineration of Kāma, the Hindu Eros, in an irrefutable proof that of Kävya Uśanas and his disciple Kaca, a myth which, while it bears a num- the fires of yoga (yogagni) were greater than those of burning passion (kā- ber of common themes with an Indo-Iranian myth, is nonetheless Hindu magni). Unmoved after having been pierced by Kama's arrows of lust, Siva in the detail that concerns us.103 Here, Kaca, the son of Brhaspati, the chap- opened his third eye-located at the level of the ajña cakra-and, with the lain of the gods (and grandson of Angiras, "Ember"), has been sent to Kā- supernatural accumulation of thermal energy that his yoga had afforded vya Usanas, the chaplain of the Asuras, the antigods, to wrest the secret of him, reduced Love's body to ashes. The fire that burned Kama was at once immortality from him. Through a number of plot twists, Kaca not only the sublimate of Siva's rasa (for Siva, as a yogin, is ürddbvaretas), his yogic becomes the disciple (for five hundred years) of Kāvya Usanas, but also the breaths, and the heat of his yogic austerities. The refined essence of these beloved of the latter's daughter, Devayänī. He is hated and feared, however, three elements-ashes-were taken by Siva and smeared over his body by the antigods, who see him, quite rightly, as an enemy agent. They there- together (as we are told in one version of the myth) with mercury, said fore kill and feed the boy to wolves, only to see him revived when Kāvya to be the concrete form taken by his agitation at the sight of Pārvati in Usanas pronounces his immortality formula over the boy's scattered re- this myth.100 mains. The formula once pronounced, the boy bursts out of the bodies of In another myth, Siva shows his superiority over a human ascetic named the wolves, killing them as he himself is reconstituted. Mankanaka when the latter, having pierced his skin, "bleeds" pure vege- The antigods then kill the boy again, but this time they burn his body table sap (sākarasa). He then dances for joy, proclaiming that his austerities and place the ashes in Kavya Usanas's evening cordial. When the guru calls have been successful (siddha). When Siva does likewise, and bleeds pure, out to his disciple, the boy answers him from within his belly. Now the snowy-white bhasma, the chastened ascetic is obliged to acknowledge the only way to save Kaca is to reveal to him the secret of immortality, such great god's supremacy.101 Elsewhere, a myth from the RA (11.158-61) re- that Kaca, once he has been revived by bursting out of Kāvya Usanas's lates how the demon Ruru, having eaten a pala of twelve-times-calcinated body, might revive his guru in turn. When Devayānī says, "Do it, Daddy," mercurial ash (bhasmasutaka), uproots the earth and pulverizes Siva's Kävya Usanas has to give in and so initiates his disciple into the secret of Mount Kailash. Siva "kills" mercury in combination with lead and red arse- immortality. When the formula is spoken, the boy bursts "like a full moon" nic, smears this amalgam on the tips of his trident, and laminates Ruru. out of his guru's body; he then revives his guru. This parallelism, between ash and rasa, is explained, in what may be This myth, which brings together the consumption of ashes, the trans-

286 287 Chapter Nine The Dynamics of Transformation

mission of a secret of immortality, and initiation, also seems to be the pro- pindas, "balls," which is the same term as is employed for the rice balls totype for myths of tantric transmission, in which the guru becomes the offered to the deceased in Hindu funerary rites, for aniconic stone monu- disciple of his disciple, and the disciple the guru of his guru. It is, as well, ments to the dead which go back to the dawn of human civilization, for the a humanization of the myths of the great gods Agni and Prajäpati and an embryo in the womb (all signified by the Marathi pind), as well as the Siva alloform of the legend of Gorakh and his guru Matsyendra. As such, it may linga (Marathi pindi), the head as sacrificial offering or, conversely, the hu- well be, as Georges Dumézil maintains, that "within the corporate body of man torso or body as a whole.107 It is this final usage of the term pind that sorcerers, the disciple is just as important as his master as concerns the is most common among the Nath Siddhas. The human body is a pind, a continued transmission of the ... supernatural knowledge that is its com- "ball," which can be transformed, through initiation, yoga, and alchemy, monwealth and its justification. Each needs the other." 104 into an immortal body. In the last chapter, I noted another usage of this In this particular myth, Kaca's initiation further serves to render him minimalist definition of the body as a sphere or globe: this was the suffix Kavya Usanas's "biological" son. As such, he is, in a sense, Siva's grandson, -oli found in such terms as garbholi (the round fetus in the womb), a term given that Kavya Usanas himself once entered into the body of Siva, from that may bear some connection with the Hindi term for hailstone, oli. Else- which he exited in the form of a figure named Sukra ("Semen") as the result where, the insignia of the Nāgā akbādās (whose close relationship to the of that god's yogic powers of digestion.105 Kavya Usanas's legacy lives on Näth Siddha orders has been noted) are nothing other than globes (golās) in medieval alchemical-as well as certain modern medical-traditions, in of ash;108 and the pills of mercurial ash which, held in the mouth, cause which the perfected alchemist is called a kavi (an alchemical wizard) and a wholly subtle and immortal alchemical body to emerge out of the husk the Bengali Ayurvedic physician a kavirāj. The successful alchemist is a of the gross, biological body are called gulikā (or guțikās), "globules."109 kavi because he, like the asura Kāvya Uśanas (Uśanas, son of a kavi), is According to Siddha logic, these are the wombs from which new life will capable, through his poetic incantations and mystic knowledge, to bend emerge: ashen globules engender immortal globes. nature to his will and thereby realize wealth, invincibility, and immortality This leitmotif of Siddha theory and practice is epitomized in the legend for himself.106 of the birth of Kayanath ("Body-Nath"), a figure who likely lived in the The symbolic use of ashes by Saiva sectarians has become greatly ex- seventeenth century, in the Jhelum district of present-day Pakistan, histori- panded since the age of the initiatory practices we have outlined to this cally one of the most important centers of the Nāth sampradāya.110 point. So it is that the cremation ground, upon which human corpses are summarily reduced to ashes, was long the preferred haunt of tantrikas who, One day, Gorakhnäth held a great feast for his fellow yogins. When in their drug- or austerity-induced trances, saw and danced with the wild the food had been dished out on leaf plates, the place of honor was and fulminating Kali and Bhairava. The infernal dance of this divine pair given to Gorakhnth's venerable disciple Ratannath. As fate would is purificatory, serving as it does to burn away the decaying matter of a have it, there were two plates of food lying before Ratannath when dying cosmos-both within themselves, in the bodies consumed there, and he sat down. After eating the food from the first plate, Ratannath in the universal conflagration for which the burning ground is a meso- then stood before the second plate. Having pronounced a mantra, cosm-for it is only when an ash-smeared Siva incinerates the universe he then caused ashes to flow from his body, after the fashion of Siva. with his ecstatic dance that universal liberation becomes possible. These he fashioned into a ball, which he placed before the second A body of devotional practices offered, in coastal Andhra Pradesh, to plate of food. He then announced that that ball of ashes would eat the terrible Saiva divinity Virabhadra, constitutes yet another practical ex- the food sitting in front of it. The other yogins, who had begun to tension of the Saiva theology of ashes. Here, the "growing" bodies of "ash laugh, were quickly silenced when they saw that Ratannath's yogic fruits," formed of a mixture of cow-dung ash and acacia gum, are at once energy was causing the ball of ashes to expand. The ball then split ritual reminders of the postmortem life of the dead, the "fruits" of karma, open and a laughing, fully-formed boy emerged from it and set about and manifestations of the expansive power of vibhuti, which is homologized eating the food on the plate before him. It was then decided that this with semen, blood, and bone in this tradition. These "ash fruits" are called boy should be given a name and initiation into the Näth order. Gor-

288 289 Chapter Nine The Dynamics of Transformation akh named the boy Kāyanāth, son of Ratannāth; and Vicārnth initi- that is borne out in a great number of Nth legends, which we now pass ated him. Kayānath later became the abbot of the Bhera monastery, in review. performed many miracles, and lived to the age of 101 years before Just as his divine guru Siva had done to him upon his own initiation, quitting his body.111 Matsyendra, after he has exhumed Gorakh, initiates him with a bath of ashes. Gorakh does the same to his disciples, first "creating" them from While the account of Kayanath's birth constitutes the richest example ashes and then initiating them with more of the same. The Nath Siddhas, of the symbolic use of ashes by the Nath Siddhas, it is by no means the and Gorakh in particular, are great yogic progenitors, fecundating women earliest such legend. No less a figure than Gorakhnath himself is the prod- with their yogic "seed," which they carry in their wallets (jboli) in the form uct of a similar manipulation of ashes. In a legend known throughout of rice grains, barleycorns, ashes, or water in which their loincloth has been northern India and Nepal, Gorakhnäth is himself the product of ashes and washed. A number of Gorakh's illustrious disciples, including Gūgā Pir cow dung-whence his name Go-rakh ("Cow-Ash"). As the story goes, a and Carpati, are conceived and born in this way.115 brahmin woman who desires a son is given a pinch of ash by Matsyendra, Like calcinated mercury, the ashes of a Nath Siddha are capable of who instructs her to eat it together with milk, following her purificatory transforming matter in a myriad of other ways. In the legend of Pūran bath after her next menses. Instead of eating the ash, she throws it onto a Bhagat, Gorakh turns bullocks into men and women into asses with ashes; heap of cow dung behind her hut. Twelve years later, Matsyendra returns with ashes he dries up wells and causes a garden to burst into bloom. In and asks for news of his son. When the woman avows that she had dis- the legend of Gūga Pir, Gorakh creates gemstones and caters a wedding carded the ash, Matsyendra scoops away twelve years of accumulated cow with ashes. Pavannäth enriches an old brahmin couple with a gift of ashes dung to reveal a perfect twelve-year-old child yogin-for the boy has been that transforms their poverty into wealth. With ashes from his dhūni, Man- practicing his sadbanas there since birth-whom he names Gorakh and iknath flattens a wall.116 Another Nath Siddha tradition relates the creation makes his disciple.112 of the earth from ashes from the dhuni of the Goddess Sakti. When Siva Gorakh's ashen dunghill, both a womb (like the wells in which Nath promises to marry her, Sakti gives Siva two handfuls of ash from her dhūni: Siddhas of legend meditated for twelve years) and a tumulus (like the samā- when these ashes are strewn upon the waters, the earth is created. In a dhis under which Nath Siddhas are buried), has its most significant struc- similar account, Visnu creates the earth with ashes from Gorakh's dhūnī. tural parallel in what is perhaps that sect's most important external at- Yet, it is also ashes to which the universe will return, and these are the tribute. This is the dhūni (from dhu, "waft") a conical pile of wood ash and ashes that form the body of the cosmic serpent Sesa. But Sesa is also a cow-dung ash which the wandering yogin heaps up wherever he alights. venomous serpent, whose fiery breath itself reduces all matter to ashes.117 Its fire warms him, its coals serve to light his chillum pipe, and its ash is As such, he is like Siva who, possessed of an ashen body and wearing ashes both the present or grace (prasāda) he bestows upon all who come to visit on his yogic body, also reduces all matter to ashes in the universal dissolu- him and the substance with which he smears his own body in imitation of tion-a dissolution dramatized in his incineration of Käma with the yogic Śiva."13 More than a rough and ready hearth, the dhūnī is, quite literally, a fire he emitted from his third eye. Siva's creative semen becomes destruc- double of the yogin's subtle body, a body that has already been cooked and tive when it has become transformed, yogically, through the raising of the transformed through his yogic austerities. The ashes of the dhūni represent internal kundalinī serpent, into fire; yet it is also by virtue of his yogic prac- the continuity of his unending sadhanas and thereby of his immortal subtle tice that he is immune to the calcinating poison spat by the great serpent body. Long after a yogin has quit his mundane body, his dhūnī (maintained Vasuki, which, when he drinks it in the myth of the churning of the Ocean and kept burning, in some cases, for centuries), like his samādhi, remains of Milk, turns his throat blue. as a memorial and testimony to his continuing presence in the world. India The symmetry that emerges out of this data parallels that of the dis- is a country dotted with the samadhis and dhunis of its great yogins, whose course of binding and piercing: just as the Siddha, by virtue of has having eternal essence is epitomized through ashes."4 This is an identification bound and pierced the volatile elements of his alchemical body, himself

290 291

Chapter Nine The Dynamics of Transformation

becomes unbindable and impenetrable and all-binding and all-penetrating; activated through combination with female sulfur. The same alternation so the same Siddha, by virtue of having calcinated and reduced his alchem- is found in the Kaulavalinirnaya, a sixteenth-century ritual compilation, ical body to ashes, himself becomes uncalcinable and all-calcinating, an which terms the bindu of the subtle body, located above the sixth, the ajña ashen, ash-smeared, ash-producing, alchemical touchstone. cakra, as either Bodhinī ("she who awakens") or Rodhinī ("she who ob- structs"). In order to understand this terminology, we must recall that the emana- 6. Reversal and Transformation tion of the Kaula universe is, on the microcosmic level, phonematic as well as material. All that exists is ultimately acoustic, vibratory in nature, an Our discussion of the transformative powers of the ashes of death has car- emanate or devolute of the primeval sonic vibration that gave rise to all ried us somewhat far afield of our survey of the serial progression of the creation, the mantra Om. When projected upon the subtle body, this alchemical operations in their relationship to the theory and practice of acoustic emanation takes the form of the fifty-one phonemes (mātrkās, batha yoga. In order that we might return to this progression, let us recall "matrices" of subtle sound) of the Sanskrit alphabet, which are situated on here that while swooning (mrcchana), the third alchemical samskāra, is a the petals of the cakras (styled as lotuses) of the subtle body. When the prelude to the powerful and transformative operations of binding (ban- yogin raises the inner kundalini, he is in effect telescoping these phonemes dhana) and killing (mārana), it also precedes samskāras four through eigh- back into their higher emanates, causing the last and "lowest" phonemes, teen, as those operations are described in the alchemical sources. the final letters of the alphabet, to be absorbed into ever more subtle pho- With samskäras four through eight, we find ourselves in the presence of nemes, culminating in the vowels and the phonemes ha and ksa. These two what appears to be a series of priming techniques not unlike those of yogic ultimate phonemes are located on the two petals of the sixth cakra, the prānayāma or the alternating ascent and descent of the kundalini. Following ajñā, which is situated behind the meeting point of the two eyebrows. The its swooning, mercury is first resurrected (utthapana), after which it is made hierarchy of sound does not end here, however, since manifest sound is but to fall (pātana). Next, it is awakened (bodbana) or suppressed (rodhana), then the emanate of a subtler, nonmanifest vibration. Beyond the ajña, there- restrained (niyamana), and lastly kindled (dīpana). fore, exist a number of levels of increasingly subtle substrates of sound, As in alchemy, utthapana is a term employed in hathayogic discourse to through which manifest sound is made to shade into nonmanifest sound. designate an important "phase change." The kundalini, when she is awak- These shadings, located, in the subtle body, between the ajña and the ened from her sleep, rises up (ut-tha) along the length of the medial chan- crown of the head (or even beyond, i.e., outside the contours of the body, nel. Because the raising of the kundalinī is a repeated practice, she quite in a number of systems), generally include nāda (resonance), bindu (drop), naturally falls down again (pat) to the base of the subtle body after the and bīja (seed).120 yogin has reemerged from his state of yogic withdrawal. As shown in the This brings us back to the bodhini/rodbini alternation, which presumably last chapter, the kundalini can be made to fall or fly upwards as well, after occurs because the mundane mind (manas) becomes obstructed, held back the fashion of the alchemical patana apparatus.118 The experienced yogin at this point in tantric meditation: but this suppression (rodhana) of the may, of course, raise and lower his vital breath, energy, seed, and con- fickle mind is nothing other than the first awakening (bodhana) of divine sciousness at will, which translates into so many withdrawals from and re- consciousness.121 The hathayogic use of the latter term has been described: turns to mundane consciousness. When he finally chooses to opt out of after a period of sleep, the kundalini awakens to rise along the medial chan- mundane existence altogether, he definitively halts the downward flow and nel and thereby transform the yogin's body and being. The former is also enters into samādbi. an important term in the yogic practice of the Krama Kaula, in which ro- More common than the term utthapana in hathayogic parlance is the dbana designates "coagulation." When the yogin penetrates the void, he term bodhana, the awakening of the kundalini.119 Now, it happens that al- falls prey to the residues of prior thoughts, acts, events, and impressions. chemical sources alternate between the terms bodhana and rodhana to desig- The yogin coagulates (rodhana) these residues into a single mass which he nate this sixth operation, in which wholly stabilized mercury is naturally then melts down or liquidates (dravana) through meditation on the circle

292 293 Chapter Nine The Dynamics of Transformation of Kālīs. Through successive coagulations and liquifications these are ulti- the reabsorption, on a microcosmic level, of the five sense-capacities, mately annihilated, permitting the mystic to penetrate the ineffable void.122 action-capacities, and subtle and gross elements back into their ethereal With this, there comes the dawning of the light of pure consciousness, a essence. So it is that the element earth, predominant in the lowest cakra, possible parallel to the eighth alchemical operation, dīpana. the mūladbara, becomes absorbed into the element water at the level of the While most of the parallels between these four preparatory alchemical second cakra, the svadhisthana. Water is absorbed back into fire at the third samskaras on the one hand and yogic and tantric precept and practice on cakra, the manipura; fire into air at the fourth, or anābata cakra; and air into the other are nonexplicit and tenuous at best, we find ourselves on firmer ether at the fifth cakra, the visuddhi. All are telescoped, swallowed back, ground in the case of jarana, the thirteenth samskāra. Jarana, it will be re- into the mind (manas) which, identified with the sixth cakra, the ājña, will, called, consists of the stadial consumption or digestion of ever-increasing in the final phases of this process, itself be absorbed into its source and quantities of mica or sulfur by mercury, until said mercury becomes bound essence, the pure Siva-consciousness located in the thousand-petaled cakra (baddha) or killed (mrta).123 This is a progressive operation, in which mer- of the cranial vault. cury, by taking increasingly large mouthfuls (grāsa) of mica, in six succes- This process is said, in a banī of Gorakhnāth, to effect the absorption or sive operations, becomes calcinated. At each stage in this process, the mer- digestion (jarana) of the celibate yogin's seed back into its ambrosial es- cury in question becomes physically altered: in the first stage, in which it sence, by which he himself is rendered immortal. He who is unable to consumes one sixty-fourth of its mass of mica, mercury becomes rodlike retain his seed, however, is condemned to death.129 As we have already (danda[vat]). It next takes on the consistency of a leech, then that of crow shown, it is not semen alone that is raised and transformed through the six- droppings, whey, and butter. With its sixth and final "mouthful," in which stage piercing of the cakras. The rise of the kundalinī is also a telescoping mercury swallows one-half its mass of mica, it becomes a spherical solid.124 of sound back into its primal unmanifest substrate. It is this dynamic that This six-step process, by which mercury is bound, is followed by an- grounds the practice of mantra yoga, the acoustic complement to the other six-step process, in which the proportions of mica or sulfur swal- hydraulic techniques of hatha yoga. This process, characterized by ever- lowed by mercury greatly increase. It is this latter process that constitutes deepening states of yogic absorption, is qualitatively measured in terms of jarana proper. After praying to Lord Siva that he "swallow my mouth- the audition of increasingly subtle sounds (nāda). ful,"125 the alchemist causes mercury to absorb a mass of mica equal to its An important turning point in this process, known as "reversing the own. Next, mercury is made to swallow twice its mass of mica, and so on nada,"130 occurs when the external voicing of mantras yields to the audition until the proportions ultimately reach 1:6, with mercury absorbing six of unvoiced mantras called the "unstruck sound" (anābata nāda).131 So it is times its mass of mica. In this final and optimal phase mercury, said to be that as the yogin's consciousness is raised through the final phases of yogic "six-times killed," is possessed of fantastic powers of transmutation.126 practice, the anabata nāda, which begins to reverberate at the level of the More superior yet is a sequence called kbecari jārana, in which mercury is heart, undergoes a series of qualitative transformations, successively re- made to absorb vast quantities of powdered gemstones, the densest sub- sounding like the roar of the ocean, of clouds, the sound of the kettledrum, stances known to man. At the conclusion of this process, mercury takes the conch, bell, horn, flute, lute, and finally, the buzzing of bees.132 These shape of a linga. The alchemist who ingests said mercury is immediately transformations, to ever subtler levels of sound, effect the total absorption transported to the realms of the gods, Siddhas, and Vidyādharas.127 Mod- of the yogin's mind. Becoming one with its object of knowledge-i.e., the ern researchers in rasa sästra lament the fact that a certain step in the tech- Siva-consciousness that is the substrate of all sound and being-it becomes nique of jarana has been lost, and with it the ability to transmute base fused with this object of knowledge and telescoped, absorbed, dissolved metals into gold and afford immortality and eternal youth.128 into it.133 These homologies, between batha yoga, mantra yoga, and alchemy, We may adduce three hathayogic parallels to alchemical jārana, i.e., ab- are seemingly taken for granted in a passage from the HYP (4.96), which sorption or digestion. The first and most obvious of these is the six-stage states that "the mind is like mercury which, bound (baddham) and freed process of the piercing of the cakras. When the yogin raises his seed, from its transitory nature through its assimilation (jārana) of the sulfur of breath, and energy through the six cakras, he does nothing less than effect mantric vibration (nāda), wanders about in the ether."134 A similar use of

294 295 Chapter Nine The Dynamics of Transformation yogic practice, to produce the same result.139 Elsewhere, we have already terminology may be found in the alchemical RRS, which designates the seen how Dharamnath was able to incinerate the (since then) lost city of ultimate phase of alchemical jarana, by which a mercurial compound is Pattan and a good part of the Sind by simply remaining motionless, stand- roasted in a puta with six times its mass of sulfur, as sabija-bandha ("binding ing on his head atop a betel-nut or sopari stone for twelve years.140 with seed").135 There exist, however, a number of legends of the Nath Siddhas in which In addition to such qualitative measures of absorption as the inaudible yogic immobility is raised to a higher level than the mere retention of some sounds heard by the yogin in his practice, there also exists a set of quantita- fixed posture. These are the many accounts of dismemberment, inhuma- tive measures of the same, which moreover appear to bear a direct relation- tion, or exile that, in a sense, set the Nath Siddhas apart from other yogic ship to an alchemical progression of a similar order. These are related to traditions. Shashibhushan Dasgupta refers to these in his discussion of the the asanas, the yogic postures, through which the yogin not only immobi- 'general air of supernaturalism" proper to Nath Siddha traditions. Others, lizes his body, but also his breath, seed, and mental activities.136 Main- such as Eliade and Van Kooij, find in these accounts traces of the shamanic taining a yogic posture alone requires tremendous powers of concentra- origins of Näth Siddha practices. tion, and so we should not be surprised to find a quantitative measure of While many of these feats-like those of the Taoist magicians and im- immobility combined with breath retention to be a yardstick for yogic in- mortals in China-may indeed be typologized as shamanic, following the tegration. Here it is the four-second matra ("measure of time") or pala (the "morphology" of this phenomenon so beautifully developed by Mircea equivalent of six mātras, i.e., 24 seconds) that is of critical interest. As one Eliade,141 they also betray an alchemical stamp. Mercury and its com- retains a yogic posture and one's breath for an increasing number of palas, pounds in flux are never anthropomorphized in the Indian alchemical texts ever-dilating durations of time, one acquires ever more fantastic siddhis. (where they are either deified or zoomorphized) as they are in western al- After twenty-four years-i.e., 60 X 24 × 365 X 24 palas, the highest mul- chemical traditions, in which the corpus alchymicum of the "Ethiopian" or tiple in this progression-of total yogic absorption, one outsurvives the "Son of God" is dismembered, crushed, cooked, etc. to be resuscitated in universal dissolution, in which even the gods Brahmā, Visnu, and Siva a new divine form. It is rather in the legends of the Näth Siddhas that such perish.137 allegorical dismemberments and tortures occur; and while the origins of This notion of a "geometric progression" of a durational order has its these accounts may be as old as religion itself, the "gloss" they receive in alchemical parallel, in which the consumption of an increasing number of Nath Siddha mythology appears to be both alchemical and yogic. palas ("straws," here a unit of weight equivalent to 82.624 grams) of mer- Of all the legend cycles of the Näth Siddhas, the song cycles concerning cury that has undergone the process of jārana with mica yields a dilating Gopīcand are the among the richest in this sort of imagery.142 The outline life span, such that the consumption of ten palas renders the alchemist a of this legend is as follows: Mayana[mati], the mother of Gopicand and a second Siva.138 When rasa and breath control constitute the Work in two "sorceress" disciple of Gorakh, loses her husband Manikcand to Yama (i.e., parts, a time-mass continuum of this order is quite natural. The impacts of he dies). When she is unable to wrest her husband back from Death, she is these two allied practices on the human body are also identical: one can "given" a son to replace him, as it were, by Gorakh. She learns, however, neither be consumed by old age (ajara) nor by death (amara). We will dis- that her son will not survive his adolescence if he does not become a yogin. cuss other important progressions of the same order in the next chapter. The intrigue of the cycle lies in the ends to which Mayana goes to bring her son under the tutelage of a guru-who in this case is Hadi-pa, or Jālandharanāth.143 7. The Legends of Gopicand and Cauranginath: Alchemical Allegories? The origins of Jalandhara go back to the original cosmogony, as related in the Bengali Goraksa Vijay and Min Cetan. Out of a void (sūnya) there appears, like a bubble, the Egg of Brahman. The unmanifest Nirañjana According to legend, countless Nath Siddhas have closed themselves into then manifests itself as Ādinātha who, from his tapas, creates the goddess wells or caves for twelve-year periods to emerge with transformed bodies; Ketakī,144 out of whom arise, from her mouth, the god Brahma; from her six months of yogic or alchemical austerities are said, in guides to hatha-

296 297 Chapter Nine The Dynamics of Transformation

forehead, the god Visnu; and from her yoni, the god Siva. To test these (As we shall see, the RA, in its concluding verses, promises just such an three, Adinatha takes the form of a decayed corpse, rotting in the wash of end for the Rasa Siddha who has perfected his craft: he will live on, above a river. Only Siva is willing to perform Ādinātha's funeral rites: in so doing, the terrible ocean of existence and beyond the cosmic dissolution, in the he uses Visnu as his firewood and Brahma as his fire. From the burning world of the divine Siddhas.)148 corpse of Adinatha arise the five original Nath Siddhas: Mīna, from his Gopīcand is eventually persuaded to become Hādi's disciple. In order navel, Gorakh from his matted hair, Jālandhara (Hādi-pā) from his bones to test his sincerity, however, Hadi sells him-for twelve cowries worth of (hāda), Kānha-pā (Kanerīnāth) from his ears (kāna), and Caurangī from his cannabis-into bondage in the house of a sorceress named Hīrā (Dia- limbs (anga). Siva then takes Ādinātha's Ketakī as his consort: she becomes mond), in the "southern land," for twelve years. During this time, Hādi sits the goddess Gauri. Gaurī then decides to test these four sons' (Caurangī is at the bottom of the sea, beneath fourteen fathoms of water. Hīrā trans- left out of this episode) yogic constancy.145 All fail except Gorakh, who forms Gopicand into a ram and later sends him to hell, which Hadi har- looks upon her as his mother-and as a reward is "reborn," as his own son, rows to save him. At the end of this period, Hadi completes Gopicand's from water wrung from his karpati (loincloth) and drunk by the Goddess: initiation; then, judging that he has made improper use of his yogic powers, this son is Carpati. Matsyendra is cursed to be debauched by sixteen hun- nullifies them. A furious Gopicand has Hadi sealed, with horse dung, into dred women in the forest or kingdom of Kadalī; Kānha-pā is exiled to a the bottom of a well. country called Dahuka; and Hadi-pa is turned into a low-caste sweeper Twelve years later, Kānha-pā, also a disciple of Hādi, is informed by (Hadi). Such is the mythic origin of Gopicand's guru. Gorakh that his guru has been buried in a well by Gopīcand. Kānha-pā, Mayanā learns that it is only by taking yogic initiation from a Hādi that through a subterfuge, manages to exhume Hadi. Knowing that Hadi's ac- Gopīcand can be saved. But the young prince appears to be more troubled cumulated yogic energy, to say nothing of his wrath at being buried in by his sorceress mother and the prospect of exchanging his kingship for such a way by his disciple, might well destroy Gopīcand entirely, Kānha- pã makes three effigies of Gopīcand: one of iron, one of silver, and one of servitude to a sweeper than he is by the hearsay that he should otherwise die young. He therefore puts his mother through a series of terrible or- gold.149 When Hadi is called up out of the well, his angry gaze reduces the deals, which are so many tests of her yogic power: three effigies to ashes. His fury sufficiently diffused at this point, Hādi is presented with the real Gopīcand, whose yogic powers he then restores.150 She was thrown into fire, but even her garment was not stained with These themes, of dismemberment and restoration, and of twelve-year smoke; she was drowned in water bound within a bag, but mother periods of forced meditation, are quite common to Nath Siddha legend. Gangā herself came forward to welcome her into her [Gangā's] bo- Gorakhnäth explains in a poem that he was born, after twelve years passed som; she walked on a bridge made of hair; she walked on the edge of inside the womb of a barren woman, with both arms and legs cut off: the a razor; she was shut up for seven full days and nights within a boiler commentator to this passage glosses this as birth, through yoga, of a new containing boiling oil, which was being heated from below con- body from which the net (arms and legs) of maya has been cut away.151 In stantly;146 she crossed all the rivers in the boat made of the husk of a another account, Gorakh awakens after twelve years of "death" (mūva).152 corn, but nothing could bring about her death, neither was any part In still another legend, Gorakh hangs himself upside down from a tree of her body damaged in any way. Mayana herself declared to her son over a raging fire, until the odor of his burning flesh attracts the attention Gopīcand, "By the practice of mystic knowledge one becomes im- of the god Brahmā, who grants him a boon.153 While still a novice, Gorakh plucks out the pupil of his eye, which he mortal ... just like the current of the tide-wave running backward barters for food to feed his guru Matsyendra: Matsyendra restores his ... When the creation will sink below and finally dissolve, and the earth will be not and there will remain only all-pervading water, the eye, through the use of a mantra.154 This is, of course, the same Matsyendra sun and the moon will set forever and the whole universe will be whose own recumbent and lifeless body is diced up by Queen Kamalā, destroyed-I shall float on for ever-I shall have no death." 147 when he has entered into the body of her husband. Siva has these pieces

298 299 Chapter Nine The Dynamics of Transformation

brought to his abode on Mount Kailash, where he safeguards them until Lūnā lights the fire and puts on the cauldron, and Puran Bhagat is plunged Gorakh's return, twelve years later. Matsyendra's soul is then restored to into the boiling oil for four hours, from which he emerges unscathed.166 his reconstituted body.155 More often than reconstituting Matsyendra, Lūnā nevertheless has her way, and an outcaste is ordered to cut off Pūran however, Gorakh is busy dismembering Matsyendra's various sons-gen- Bhagat's hands, gouge out his eyes, and throw him down a well.167 The erally to awaken Matsyendra from his attachment to phenomenal exis- outcaste, pitying the young prince, spares him, and instead slays a fawn tence. On one occasion it is Matsyendra's son Mīna that Gorakh beats and brings its eyes and blood to Lūna; she, however, tests this blood by against a stone like a washerwoman until he is dead.156 In another legend, plunging a pearl into it and sees that it is not that of her stepson.168 The Gorakh kills and skins Matsyendra's two sons Nemi and Parsva-whose outcaste, fearing for his own life, goes to the forest where Puran Bhagat names are identical to those of two founders of the Jain religion-and later has hidden. In order to spare him, Pūran Bhagat has him do Lūnā's bid- revives them.157 Elsewhere, he transforms certain of his disciples such that ding. His arms and legs are cut off, his eyes gouged out, and his body one half of their bodies becomes gold and one half iron.158 thrown down a dry, broken-down well (jīrnāndhakūpa). Also at Pūran Bha- These supernatural, fantastic, and clearly shamanic elements, a staple gat's request, he returns to the city to tell Acchran that her son will return of Nath Siddha legend, are nowhere as evident as in the best known after twelve years.169 Acchrän, turned out of the city by the king, comes to and perhaps the oldest of such accounts, the legend of Caurangi ("Four- the well in which Puran Bhagat has been thrown. He cries up to her "Set Limbs") or Püran Bhagat. This latter name, by which he is known in the my elephant free, mother, to go to the Kajali Forest ... "170 longest recension of his legend, from the Punjab,159 may be read as "Total Twelve years pass, until one day Gorakh and his retinue of yogins come Devotee" (pūrna bhakta). Its second member may, however, be read in an- to Püran Bhagat's well on their way from their monastery at Țilla to Sial- other way: among many tribal and popular traditions bhagat is a term em- kot. There they find the young man and draw him out with a single thread ployed for a shaman, soothsayer, sorcerer, exorcist, or a person who com- of spun cotton. Gorakh then restores his eyes and, sprinkling nectar (am- munes with the dead. Elsewhere, the term is applied to oracular devotees rta) over him, his limbs.171 Twenty-four years later, Gorakh returns again of the Goddess. When one considers that the Mongol term böge/bö'e/beki to the well and finds that Puran Bhagat has remained inside, in the practice means "male shaman," this alternate reading for Pūran Bhagat's name (i.e., of yoga. When Püran Bhagat asks Gorakh to initiate him into the Nāth "Total Shaman") ought not to be rejected out of hand.160 order, Gorakh first tests his mettle by sending him to beg alms from Sun- While Caurangi is generally held to be the son of King Sālivāhana (Sāl drän ("Beautiful"), the queen of Simhala. There, after an episode evocative Vahan, Salwān, Sulivān, Sulwahan in vernacular transcriptions), a Punjabi of the experiences of Gopicand and Matsyendra in that land, but in which recension of his legend first calls him King Sankh. Finally, one source Püran Bhagat remains celibate and free (and Queen Sundrān kills herself), maintains that Cauranginath was born of Siva's seed.161 Püran Bhagat is the Gorakh and his disciple Kānīpā (Kānha-pā) initiate him with ashes and son of Sankh/Salwän's first queen Acchrän, and it is not until the age of earrings (mundrān).172 twelve that has his first audience with the junior queen, Lūnā (or Lūnān/ Now a full-fledged yogin, Püran Bhagat returns to Sialkot, his child- Nūņān/Noņān).162 This queen, who is in reality a terrible witch,163 falls in hood home, where his mere presence causes his garden to burst into love with her adolescent stepson and attempts to seduce him. When he bloom. He restores his mother Acchran's sight and forgives Salwan and refuses her advances, she, like Potiphar's wife in the biblical story of Jo- Lūņā. To the latter he gives grains of dhak and rice from his yogin's wallet, seph, denounces him to the king, claiming that he had seduced her.164 promising that by swallowing them whole, she will conceive and bear a To say that death has no fury like a woman scorned would be a great son, Rasālū.173 understatement in Lūna's case. At her urging, the king orders that Pūran Apart from their clearly "shamanic" motifs, the accounts we have passed Bhagat be bound (bandhke) hand and foot and that his head be cut off and in review may in certain cases be also be read as so many alchemical allego- kebab'ed.165 When the boy is brought before the king, he takes a vow of ries.174 Dismemberment, binding, sealing under the earth with horse dung, truth: let the king plunge him into a cauldron of boiling oil; his innocence boiling in oil, and birth from a burning corpse evoke the alchemical opera- will be proven if no part of his body, not even one of his fingers, be burnt. tions of mardana, bandhana, samputa, sāraņa, utthāpana, and mārana. Hādi-

300 301 Chapter Nine The Dynamics of Transformation

pa's incineration of metallic effigies of Gopicand, Gorakh's transformation seller who was called Ranka, i.e. the poor, because he was an utter of the stuff of his disciples' bodies into bonded gold and iron, the names pauper and evidently near bankruptcy. After the peasant had bought Hīrā ("Diamond") and Kajalī Van, and an array of other elements also ap- from him what he wanted, he returned to the golden man, and then pear to betray an alchemical matrix. found that in the place where the cut off finger had been, a new finger It is, however, the legend of Puran Bhagat that arguably may most be had grown.178 qualified as a medieval Indian alchemical allegory. Like mercury, its hero is born of Siva's seed; his stepmother Lūnā (lavana, corrosive "salt")175 In another such account, the body of an alchemist, aided by the king of causes him to be dismembered (mardana, mārana), after first boiling him Malwa, becomes transformed through a series of alchemical samskāras. in oil (särana). The well he is thrown into is jīrnandba, which evokes the This account, related above in chapter three, ends with the king withhold- closed (andha) crucible in which mercury is calcinated (jīrna, the past pas- ing the final "packet" of alchemical preparations, with the result that the sive participal form of the same root jr that generates jārana) or bound alchemist, rather than arising out of the cauldron as a perfect, invincible (baddba). This well is moreover identified with Kajali Van, the "forest of immortal, is transformed into a anthropomorphic lump of silver.179 black mercuric sulfide" (kajjali).176 After twelve years, he is drawn out of What is remarkable about this latter account, apart from its content, is his well and restored through the magical techniques of Gorakhnäth, who that it appears to correspond quite directly to the alchemical operation restores him to wholeness with amrta. The legend ends with Püran Bhagat with which the RA closes, at the end of its eighteenth chapter. This proce- offering Lūņā grains (gutikās) to eat, by which she conceives Rasālu, "the dure, related above in chapter five, culminates when the Alchemical Man mercurial one" (rasa).177 emits the mantra "Hum!" and arises out of the cauldron with a massive body that shines like the sun and is possessed of all the siddhis. If such homologous operations are only implicit in the Gopicand and 8. The Man with the Golden Finger and Gorakh's Smithy Püran Bhagat legends, they are made explicit in a poem in which Gorakh- näth compares his subtle body to goldsmithy:180 Such legends of bodily dismemberment and restoration as that of Pūran I take the gold [the void essence] in the goldsmithy [the cranial Bhagat are not wholly unique to the Nth Siddha tradition. A number of vault]-I am a goldsmith by trade. Pumping the bellows [of my similar accounts, apparently of Rasa Siddha inspiration, are recounted by breaths], and stabilizing my mercury, I have fixed it and then mixed the eleventh-century Muslim savant Alberuni. One is "a tale about ... the it together with mica. city of Valabhi": I the goldsmith am in my gold. The root cakra [mūlādbāra] is my A man of the rank of a Siddha asked a herdsman with reference to a firepot. I forge it on the anvil of vibration [nada], using my drop plant called Thohar, of the species of the Lactaria, from which milk [bindu] hammer to press out the gold. flows when its leaves are torn off, whether he had ever seen Lactaria In an ever-verdant forest my poisoned charcoal [burns, with its from which blood flows instead of milk. When the herdsman de- fire] blowing upward naturally, through the bellows' twin jet.181 clared that he had, he gave him some drink-money that he should Harmonizing [the jets of] sun and moon [idā and pingalā, upbreath show it to him, which he did. When the man now saw the plant, he and downbreath] I stop the breath [kumbhaka], and breath is merged set fire to it, and threw the dog of the herdsman into the flame. En- into breath. raged thereby, the herdsman caught the man, and did with him the With one ratti [grain] of work, you can steal away a māsa [lentil's same as he had done to his dog. Then he waited till the fire was extin- weight]182-and I am the ratti that does the stealing. Stealing the guished, and found both the man and the dog, but turned to gold. māsā, I remain within the māsā: by gathering up [gold] in this way, it He took the dog with him, but left the man on the spot. Now some is I who am gathered into myself.183 peasant happened to find it. He cut off a finger, and went to a fruit- Gold above, gold beneath, gold in the midst of gold.184 He who

302 Chapter Nine

makes the triadic void his dwelling-place has a body that is neither TEN pure nor impure.185 That which is beyond mind [unmani] is the balance beam. Mind is the weighing pan, and six lentil's-weight of breath are in the pan. Penetration, Perfection, and Immortality While Gorakhnäth was sitting here seeking after gold, gold came in [to his smithy] of its own accord." 186 ... the tiny points of light that appear all over the Himalayas when the holy men light their fires ...

Whereas Indian medical tradition defines rasayana as the regeneration or reconstitution of the body, through the restoration of the vital bodily fluids (rasas), to a youthful state of health and virility,2 the alchemical usage of the term is a more ambitious one. In alchemy, the rasa in question is mer- cury, which substitutes itself for human bodily fluids and thereby trans- forms a body of flesh and blood into a golden (svarna),' adamantine (va- jra),4 or perfected (siddba) body. As shown in chapter five, the Siddha path consists of three complementary approaches to the attainment of such a body: the erotico-mystical, the hathayogic, and the alchemical.5 These ap- proaches are complementary inasmuch as all play on the correspondence between human and divine vital fluids (rasas) on the one hand and, on the other, the dynamic of "piercing" or "penetration" (vedha), which effects the transformation of the body in question through all three of these comple- mentary techniques. This same term, vedha, is employed in (1) a form of tantric initiation involving a transmission of vital fluids from teacher to disciple, (2) the hathayogic piercing of the cakras as well as in a particular technique (called mabavedha, the "great penetration") employed to that end, and (3) the alchemical transmutation of base metals into gold. In fact, the alchemical use of this term makes little sense-for why should trans- mutation be referred to as "piercing" ?- until it is viewed in the light of these parallel and complementary practices.“ I begin this chapter by show- ing the ways in which the notion of "perfection through penetration" was meaningful in the tantric context and conclude by delineating the various registers upon which said perfection manifested itself-in the form of supernatural powers (siddhis), bodily immortality (jivanmukti), and apo- theosis in the form of accession to the higher spheres of the superhuman Siddhas.

303

304 305 Chapter Ten Penetration, Perfection, and Immortality I. Initiation in Medieval Indian Literature kingdom and family restored to him by the god Dharma-who has been disguised as Hariścandra's employer, an untouchable disposer of corpses The encounter (or high-impact collision) between representatives of reli- on the Benares cremation ground. The happy ending has the royal family gious and temporal authority and power is a favorite literary theme in In- and its entire capital city elevated, apotheosized as it were, to an atmo- dia, where kings have been running into one or another sort of holy man, spheric level-with Viśvamitra's blessing. both inside and outside of texts, for millennia. One of the best-known This is the ca. fourth-century A.D. Puranic myth of Hariscandra,10 in myths of classical and modern Hinduism is the account of Hariscandra, which an old and uneasy alliance between the sacred authority of the brah- king of Ayodhya, the Job of Hinduism, and the trials he suffers at the hands man priesthood and the temporal power of ksatriya royal sovereignty is the of Viśvamitra, who had earlier been the officiating priest at his royal conse- source of narrative tension. But India is changing. The Gupta empire, the cration. Following the ritual services he has rendered, Viśvāmitra the priest last Hindu empire of India, collapses in the seventh century. A troubled (even though he was himself a prince by birth) has the right to demand a "middle ages" follows, during which much of north India falls under the fee (daksina) from Hariścandra. According to brahmanic ideology, this fee yoke of foreign, non-Hindu rulers. The old order is crumbling, and the is Viśvāmitra's due not only because he has performed a service for his ritual order ensured by the old brahmanic system is beginning to look anti- king, but also-and most important-because temporal rule is, fundamen- quated. New forms of religious order and authority (monastic orders) as tally, a brahman prerogative that brahmans, in a mythic past, delegated to well as new religious sects appear to fill the power vacuum or to aid Hindu- their ksatriya juniors.7 This they did, we are told, in order that brahmans ism in going underground in times of Muslim rule. might concern themselves with the more elevated matters of dharma and When paradigms shift, so too do paradigmatic narratives. A tenth- moksa, leaving for kings the dirty work of governing, dispensing justice, century play by Ksemīśvara, the Candakauśika1 (Wrath of Viśvāmitra), is waging wars, fathering princes, etc. Brahmans retained, however, their au- a case in point. The core of this dramatic narrative is the same as that of thority over ksatriyas in their role as the advisors, the mentors (mantrins) the Puranic myth: Hariscandra, suffering the fury of Visvamitra, has been of their kings, by playing a sacerdotal role in rituals vital to the harmonious relegated to a position of abject slavery in Benares. What is novel in Ksem- continuation of the latter's rule and, most important, by deigning to per- Isvara's otherwise rather dreary adaptation is the sectarian guise of Hariś- form the ceremony of royal consecration, of which a crucial moment was candra's savior. Rather than having Dharma, the Hindu god of Universal the ritual "punishment" (danda), by a brahman priest, with a rod (called Order disguised as the untouchable lord of the Benares cremation ground danda in the hands of the king, but called a spbya in the sacerdotal context) (a figure nowadays called the dom-rāj), Ksemīśvara has the god parade as of the otherwise unpunishable (adandya) king.8 And, following his conse- a Kapalika.12 He bears all the markings of a tantrika: he is ash smeared, cration-the delegation by his brahman initiators of their own sovereignty decked out with human bones and skulls, "having the appearance of Siva to him-the king also found himself in their debt in the matter of the fee Bhūtanātha." He reads Hariscandra's mind, controls "vampires," and pos- that was their due. In effect, the king wrote his brahman priests a blank sesses alchemical preparations which afford him immortality (rasāyana), the check in these situations, and it is the amount of this check that sets up the power to transmute (dhātuvāda), and the power of flight (pādalepa), and dramatic tension in the myth of Hariścandra. possesses an eye unguent (añjana) that permits him to see underground to In this Puranic myth," Visvamitra, who is furious at Hariscandra for a find buried treasure.13 This Kāpālika, portrayed most sympathetically by number of reasons, exacts his revenge by ruining Hariscandra financially Ksemīśvara, having introduced himself to Hariścandra, departs and then The daksina that Visvamitra demands of Hariścandra is astronomical, an returns in a twinkling with a vampire in tow; on the vampire's shoulders is amount far greater than even a king can possibly pay. The righteous Hariś- an immense mineral treasure, discovered with the aid of the Kāpālika's can- candra attempts to acquit himself of his debt by giving his entire kingdom see-underground-with-it unguent, which Hariścandra may now use to pay to Vivāmitra, yet this is not enough. He then sells his wife and son into off his debts to the irrascible Viśvāmitra.14 slavery, but Viśvāmitra is still unsatisfied. Lastly, he sells himself into slav- Ksemīśvara's adaptation of the original Purānic account is truly remark- ery and, after a dozen years of terrible suffering and depradation, has his able when we consider the sort of people the Kāpālikas were made out to

306 307 Chapter Ten Penetration, Perfection, and Immortality

be by their orthodox critics. The Kapalikas, fornicators with menstruating stories ending in riddles which the king must solve lest his head burst into women, cremation-ground consumers of human flesh, worshippers of the a thousand pieces. At the end of the final story, the vampire informs the female sexual organ, brahman murderers, were the Hell's Angels of medi- king that Ksantasila's true intention has been to offer him, the king, as eval India. This drama is therefore a watershed of sorts in the history of a human sacrifice to the vampire storyteller, in order to win for himself, the literature on brahman-ksatriya relations in classical India. What Hariś- Ksāntaśīla, lordship over the Vidyādharas, the Wizards of medieval Indian candra cannot pay the brahmanic priest Viśvāmitra retail the Kāpālika tradition. Kşāntaśīla, who bears all the markings of a terrible Kāpālika, has wizard can, through his supernatural powers, get him wholesale. We may the tables turned on him by the king, who cuts off his head, the which he therefore take Ksemīśvara's superimposition-of a Kāpālika guise upon a then offers, together with the heart, to the vampire. Siva descends, and in god of moral order-to be an indication of just how tattered that old order a deus ex machina scenario reminiscent of the Puranic Hariscandra myth, was looking to certain elements of Hindu society. We may further see in grants Trivikramasena dominion over the earth as well as the Vidyādharas. this tenth-century adaptation an indication that tantric practitioners were This story takes up all of the major elements of the Candakauśika drama, replacing old-order brahman priests (even if Viśvāmitra was originally but with all the pieces turned around. Visvamitra, the wrathful exemplar the most heterodox of all the brahmanic sages) in the role of advisors to of the priestly function who empties Hariscandra's royal treasury in order kşatriya kings. to better abuse him is here replaced by the evil Kāpālika Kșāntaśīla who The new partnership that was being forged-between Hindu princes fills Trivikramasena's treasury in order to later abuse him. The good Kāpā- and representatives of one or another of the Saiva orders-such as the Kā- lika of the Candakausika is here replaced by the vampire who had been his pālika described by Ksemīśvara-is, through some one thousand years of treasure-bearing coolie; and the vampire of the eponymous tales in a sense Indian literature, described from a number of perspectives. At times, the effects Trivikramasena's initiation-in this case to lordship over the deni- new world order curiously resembles the old, with täntrikas asserting their zens of an atmospheric realm, the Vidyādharas. hegemony over their princely patrons with threats and violence. In nearly An interesting feature of the Vampire Tales narrative is the need, on the every case, narrative depictions of the täntrika-prince relationship sparkle part of a renouncer, of a royal hero in order to accomplish his ends. A with the serendipity of the medieval literary style, with vampires, wiz- similar theme, of a symbiotic relationship between a tantrika and a heroic ards, mysterious ascetics, hidden treasure, and the supernatural intervening prince, is found in an episode from Banabhatta's seventh-century Harsacar- time and again. Behind all the glitter, however, one important leitmotif ita which describes the meeting between king Puspabhūti and Bhairavā- emerges: this is the theme of a yogic or tantric initiation of the king by a cārya, a powerful south Indian ascetic, who is decked out in Kāpālika garb Saiva guru. and wears great crystal kundala earrings. One day, Bhairavācārya asks the We take our second medieval example from the eleventh-century Kash- king to come and assist him in the completion of a powerful rite, the goal miri classic, Somadeva's Vetālapañcavimati (Twenty-five Tales of the Vam- of which is to gain control over a vampire and become lord of the Vidyā- pire). These tales, inserted within the massive Katbāsaritsāgara (Ocean of dharas. Once again, the setting is a cremation ground under the dark of Rivers of Story), are themselves framed by a narrative that begins when a the moon. Once again, a corpse and all manner of tantric paraphernalia are king named Trivikramasena discovers that each of the fruits a mysterious brought into play. When the terrible vampire arises out of the corpse's ascetic has been bringing him every day for ten years has a precious jewel mouth, the king's swordsmanship saves the day, and Bhairavācārya be- inside of it. When he questions the ascetic as to his reasons for such lar- comes a Vidyädhara. Here, the same elements are present, but once again, gesse, the mystery man, named Ksāntasīla, replies that he needs the king in a slightly different order. Now it is the vampire who is evil and the Kā- to help him perform a secret cremation-ground rite. On the appointed palika who is good rather than the other way around; now the king aids the night (the dark of the moon), the king comes to the cremation ground, Käpālika, rather than the vampire aiding the king, to become the lord of where Ksantasīla bids that he bring him the corpse of a man hung in a sisoo the Vidyädharas. What remains constant is the ritual structure, the trans- tree a short way to the south. The king cuts down the corpse which, as it formative late-night cremation-ground rite which effects a breakthrough turns out, is possessed by a loquacious vampire who narrates twenty-five to another world and superhuman powers.

308 309 Chapter Ten Penetration, Perfection, and Immortality

Another, less obvious, literary account of power politics between tāntri- serpent is said to dance with the yogin. And, at the end of her rise, it is kas and kings is Kalhana's chronicle of the fall of the tenth-century Kash- the yogin's own sexual fluid which, carried upward through her body, is miri king Cakravarman. Whereas Kalhana wrote his Rājataranginī in the transformed into immortalizing nectar. As a conduit for the yogin's semen, twelfth century, the amount of detail he gives concerning a piece of palace the female kundalinī may be likened to the female sexual organ: indeed, she intrigue-which had occurred some two hundred years earlier-leads one is sometimes called dombi, and the Dombi, as the tantrika's favorite sexual to believe that the story was grounded in something more than Kalhana's imagination. While Sir Aurel Stein, who translated this work at the turn partner, was most prized for the transformative powers of her menstrual blood.17 of the century, stated that Kalhana's "amusing anecdote look[ed] to be au- For the same reason that there could be nothing more defiling to an thentic,"15 we might also choose to see in it an allegorization of a piece of orthodox Hindu than the menstrual blood of an outcaste woman, the het- tantric sorcery. erodox täntrikas valued the same as the most powerful fluid in the universe. In the spring of A.D. 936, Cakravarman, with the help of a group feudal For the uninitiated, however, the stuff was dynamite, as Cakravarman landholders called the Damaras, breaks the back of a military caste called the Tantrins, who had been the praetorian guard of his rival for the throne. quickly learns. Following the night of the big dance, he becomes the Dombīs' sexual plaything. He makes them his queens, and they and their Cakravarman has perhaps not seen the last of the Tantrins however; more- over, these Tantrins have much of the tantrika about them, as the principal outcaste relatives take over the royal court, polluting the entire kingdom

episode in Cakravarman's one-year reign16 would appear to indicate. One in the process. High offices are bestowed upon those who eat polluted leav-

night, the king is entertained in his court by an "outcaste" Dom (the same ings (ucchista) off the Dombis' plates,18 and Dom ministers proudly wear

Dom as the untouchable dom-raj of the Benares cremation grounds) singer the girls' menses-stained undergarments over their courtly attire. Cakra-

named Ranga and his two dancing daughters Hamsī ("Goosie") and Nāga- varman, his mind totally gone, rapes a brahman woman and treacherously

latã ("Serpentina"). These last two quickly steal the show, enchanting the puts to death a number of his Damara allies. Shortly thereafter, Damaras

king with an open display of their feminine charms, which are described set upon him one night in the royal latrine and flay his body as it lies upon the swelling breasts of his Dombi concubine. by a court rogue (a description which Stein leaves out of his translation): Now, we can read this account, with Stein, as an "amusing anecdote," O Lord! The music, as if enchanted by these two lovely dames, has or perhaps glimpse in it, once again, a piece of tantric power politics (on become as alluring as a liquor poured into a crystal glass! It seems as the part of the Tantrins?), the destruction of a king through a sort of loaded if the moon, bewildered by the shining teeth of these singing girls, is tantric initiation. Here, the closest parallel would be the Vampire Tales, in passionately kissing the gleaming columns of the ivory hall! Seeing which the täntrika is the heavy-with the difference that here there is no that we are speaking of her, one of them has shot a glance of smil- vampire to save poor king Cakravarman. ing indignation our way. The one singing with lowered gaze and A final group of accounts, all of which are structurally similar, are much trembling earrings, who looks so beautiful, is passionately acting out more straightforward, as they dramatize a king's tantric initiation, by a the reverse (viparīta) posture of love. Saiva ascetic of the Näth order or one of its forerunners, in terms reminis-

There is more going on here than dirty dancing, especially if we look at cent of the Candakausika or even of the ancient royal consecration of kings by brahman priests. The most elaborate of these accounts is that of Deoraj, this scene and what follows in the light of tantric imagery. First, the two dancing girls, daughters of the Dom Ranga, are Dombīs, the most prized of founder of the Rawal dynasty, which is found in the bardic chronicles of the Rāwal kings of Jaisalmer.19 Here, as we saw in a previous chapter, Deorāj partners in ritualized tantric sex, sex in which the reverse viparīta posture is the rule. Their names are revealing as well: as we have seen, in the subtle (b. A.D. 836), the future founder of the Rawal dynasty, is visited by a Jogi

body, it is both a coiled serpent (the kundalini) and a goose (the hamsa) that named Rita who bestows upon him the title of Sid[dha] and then goes off, leaving him his rasa kumbba or elixir vessel. Deorāj uses the gold produced rise from the level of the sexual organs up to the cranial vault to afford the by this vessel to raise an army and the walls of a fortified city. Rita returns practitioner immortality. In her rise through the yogic cakras, the kundalinī and exacts that Deoräj become his disciple and, as a token of his submis-

310 311 Chapter Ten Penetration, Perfection, and Immortality sion and fealty, adopt the external signs of his order. He places the earrings palace to bring him a present (prasad) of yogurt, from his mother. When (mudrā) in Deoraj's ears, the little horn (singnad) around his neck, and loin- cloth (langoti) about his loins. Attired in this way, Deorāj goes about ex- the boy returns, Gorakh instructs him to hold the yogurt out to him in his claiming "alakh! alakh!" Then, having exacted that these sectarian rites of cupped hands. Gorakh then draws that yogurt into his mouth, attempts to initiation should be continued to the latest posterity, Rita disappears. spit it back out into Prthivīnārayan Sāh's hand, and tells him to eat it.21 Prthivīnārāyan Sāh however opens his hands and lets the yogurt fall on his Two elements of this account are to be retained. The first is that we feet. Gorakh then explains to the boy that had he eaten his prāsād, the yo- have returned here to the ancient brahmanic ideology in which the king is gurt he had spat out, he would have become a universal emperor. But be- a junior partner in power to his priestly initiator, who grants him his royal cause he has let the yogurt fall on his feet, he will only conquer the earth power in exchange for a token of his submission in initiation. The second is that his priest, Rita, is a Jogi, a yogin, probably of the Pāsupata order (or as far as his feet will carry him. Prthivīnarayan Sāh's feet soon carry him to the Kathmandu Valley where, through a second intervention by Gorakh some other forerunner of the Nath sampradaya). He bears all the marks of these Saiva orders: his tattered cloak, great earrings (also worn by the Kā- his Gurkhas win a great battle and become the rulers of Nepal. The great pālika Bhairavācārya in the Harsacarita account), little horn, the words guru's intercession is commemorated, down to the present day, on the alakh! alakh! ("attributeless! attributeless!") are all particular to the Nath Nepali one rupee coin, upon which is written, verso, "Long live Gorakh- nāth!"22 sampradāya. Medieval and modern legend and history abound with accounts of just This spitting incident is not unique to the biography of Prthivīnārāyaņ Śāh. Alberuni's eleventh-century account of India includes an episode from this sort of alliance-sealed through initiation-between a tāntrika (some- the career of the Mahasiddha alchemist Vyādi that is quite similar. Here, times identified as a Nth Siddha) and a fledgling prince. As recently as the early nineteenth century, just such a pact was struck between a Nath Vyādi has, at long last, discovered the alchemical elixir and is flying Siddha and the Marwar prince Man Singh;20 a number of accords of the through the air over the royal capital of Ujjain. Looking below, he sees King Vikramāditya standing at the gate of his palace. "Open thy mouth for same order were made between Nath Siddhas and princes, in earlier centu- ries, well to the north of Rajasthan, in Garhwal (Uttar Pradesh) and Nepal. my saliva" he calls down to the king. The king instead steps out of the way In fact, nearly all of the exemplary Nath Siddhas of legend are born into and the saliva falls on the threshold, which is immediately transformed into princely families, before being saved-from rivals, enemies, evil stepmoth- gold.23 Similarly, Bāppā Rāwal's guru Hārīta Rāśi, having risen into the air, ers, death-by a powerful yogin, usually Gorakh, who then initiates him attempts to spit into his mouth in order to complete his initiation. Like Prthivīnārāyaņ Sāh and Vikramāditya, Bāppā Rāwal allows his guru's into that order. As we have seen, initiation involves the piercing of the ears with the great kundala or mudra earrings that are the hallmark of the Nath expectorate to fall at (or on) his feet, and thereby receives something less than the intended effects of his guru's gift.24 Yet another such account, con- sect, the placing of the little horn, the giving of the mantra, etc. cerning the site of Hinglaj Devi in Baluchistan, is all the more surprising, There is another feature of the Nath Siddhas' tantric kingmaking that because it is reported by the Muslim "Jogis" of the Sind. Jogi Maula Baksh should also be noted here. This is a facet of the initiation rite which, al- relates the story:25 though it is evoked in only a few legendary accounts, plays a much more important role in instructions for tantric initiation (diksa) as such are found One day Shah Abdul Latif Bhitai, the Sind's great mystic poet, met in a number of classic tantric texts. The most comprehensive mythologiza- Jogi Madan Vaniyo who taught him ablutions of the spirit and took tion of this ritual is found in a chronicle of the Gurkha (named after Gor- him to the goddess Hinglaj. Shah saw the Jogis feed Nani Hinglaj milk akh for reasons that will be made clear) dynasty of Nepal. One day, young that she promptly vomited out. Nani ignored him the first time, but the Prthivīnarayan Sāh, who would later lead the Gurkhas in their successful third time she obediently drank the milk. Taken aback by Shah's suc- conquest of the Kathmandu Valley in 1768, chances upon Gorakhnath, cess, the Jogis decided that night that they would devour Shah so that who is meditating in a cave. Gorakh sends Prthivīnārāyan Sāh back to his his powers, as evinced by his success with Nani, would pass on to

312 313 Chapter Ten Penetration, Perfection, and Immortality

them. Shah overheard their plan and prayed to God for help. So Shah sexual fluid), and "having pronounced 108 mantras, [says] 'put your mouth was swallowed by the earth instead and the good earth brought him in my mouth.'"35 In the Hindu Sarada Tilaka, the guru is said to transfer, to the safe haven of Bhit where he started to preach mysticism. through the mouth of his initiate, "that which has oozed inside his mouth"

The only way we can make sense of these accounts is by interpreting (vidruvaktrāntare), after which said fluid is returned back into the mouth of the guru.36 Elsewhere, the Yogakundali Upanisad describes the yogin's "pure them in the light of a tantric initiation ritual that was current in the period in which the Nath Siddhas were enjoying their heyday in northern India. phlegm, whose white color is that of the moon ... [which] when churned

We now turn to this ritual. by the breath and the energy of the kundalini, becomes a rich flow of nec- tar."37 Initiation into the Saiva Aghori sect is effected when the guru places a drop of his semen upon the initiate's tongue.38 Finally, in a verse perhaps inspired by the K7nN,3 the TA (29.127-32) prescribes the mouth-to- 2. Vedhamayī Dīksā: Initiation by Penetration mouth transfer of sexual fluids for the attainment of siddhis.40 In the light of this data, we may conclude that the yogurt that Gorakh Four tenth- to fourteenth-century tantric sources-the Tantraloka26 of "reverses" and offers to spit into prince Prthivīnārayan Sāh's mouth, the Abhinavagupta, the Kulārnava Tantra,27 the Sāradatilaka of Laksmaņadeśik- transmuting spittle that lands on king Vikramaditya's threshold and on endra,28 and the "Sakta" Goraksa Sambita29-describe an initiation known Bāppā Rawal's feet, the reverse sexual posture that the dancing Dombī girl as vedhamayī dīksā, "initiation having the form of penetration." This initia- (whose name evokes the kundalini serpent through which the transforma- tion basically consists of the penetration, by the guru's sakti, of the subtle tions proper to yoga and initiation take place) adopts in her seduction of body of the initiand. Abhinavagupta30 states that initiation by penetration King Cakravarman-may all be read as so many thinly veiled accounts of may take six forms, of which one is called binduvedha (penetration of "virile initiations by penetration, in which a guru attempts to transform his dis- potency," but also, of a "drop," bindu, of semen). In all of these variant ciple into a realized (siddha) being, by injecting him with sublimated forms, the guru, having entered the body of his disciple (whose kundalini semen. has been awakened) unites with that kundalini within the disciple's body Let us, however, go deeper into this matter of the apparent fluid transfer and subsequent raises it from the disciple's lower abdomen up to his cranial that takes place within the context of the so-called initiation by penetra- vault. The form the guru takes as he courses through his disciple's body tion, an initiation which, as we have seen, has the power to transform a boy may be that of a drop (bindu) of seed or of speech. In many descriptions of into a king and a disciple into a "second Siva." Here, the term vedha[na], this operation, the guru is said to exit the disciple's body through the "penetration," will guide us as we attempt to penetrate the plays of cor- mouth and thus return back into his own body through his own mouth.31 respondences proper to Siddha alchemy and to navigate the labyrinthine By means of this initiation, the guru reveals the supreme Siva to his dis- internal landscapes of the subtle body. ciple;32 the disciple whose bonds have been cut thereby becomes "a mani- The noun vedha, derived from vidh, the weak form of the Vedic root fest Siva."33 vyadh, "pierce," makes its first appearance in the medical literature, where What these various sources appear to be describing is a "procreative" it has the sense of "puncturing, wounding, a wound." Vedha, with or with- transfer of yogic seed between guru and disciple, a transfer that is charac- out its -na suffix, retains this sense of piercing or penetration throughout terized, in binduvedha, by a powerful seminal flow, in both guru and dis- the medieval traditions that concern us. In addition to these usages, vernac- ciple, that spreads through their bodies and rises to the cranial vault of ular Middle Indo-Aryan forms of these terms introduce a ba-/va- alterna- both.34 In a Buddhist account of the Mahāsiddha lineage of Nāgārjuna, tion, generating such terms as bedha and bedhana. These are in turn con- found in Nepal and edited by Giuseppe Tucci in 1930, a description of the flated, particularly in hathayogic sources, with the terms bheda and bhedana, Siddha initiation ritual climaxes when, the guru, having meditated on the nominalizations of the Sanskrit root bhid, "split, cleave, pierce." heart of the disciple, transfers from his own mouth into the mouth of his In addition to the use of this term for the form of initiation described disciple "that which is contained in the vajra vessel" (i.e., "transformed" in the literary sources we have passed in review, the term vedha[na] is also

314 315 Chapter Ten Penetration, Perfection, and Immortality

employed in the vocabulary of Hindu alchemy and hatha yoga. In all three hold in his mouth a solid "pill" (gutika) of mercury, which will gradually instances, an active agent penetrates and thereby fundamentally transforms penetrate his body and so transmute it into an immortal golden, diamond, its passive counterpart. As we will see, the dynamics of the processes desig- or perfected body nated by these three uses of the same term are quite identical. More than This transposition of vedhana, penetration/transmutation, from metal- this, the texts which describe these various penetrations explicitly parallel lic to human bodies, is directly addressed, in the terms with which we are or identify these processes or techniques with one another. already familiar, in the eleventh-century RA: "As in metal, so in the body. I have already broached the matter of vedhamayī dīksā, "initiation by Mercury ought always to be employed in this way. When it penetrates penetration." Let us now return to the Kulārnava Tantra, which, immedi- (pravian) a metal and the body, [mercury] behaves in an identical way. ately following its treatment of this and other forms of initiation, makes First test [mercury] on a metal, then use it in the body."43 The use of the the following statement: "Just as penetration by mercury brings about auri- term śarira-yoga for the application of perfected, tested mercury to the faction (suvarnatam) [in metals], so the soul, penetrated through initiation body is not fortuitous here: in both alchemical and hathayogic sources, the (dīksā), attains Siva-hood (sivatvam)."41 This identification was a common- complementarity of the two disciplines is stressed time and again. Whence place of medieval Indian thought, as evidenced in the echoes we find of the statement in the opening of the RA (1.18b): "Mercury and breath [con- this statement in a wide array of nonalchemical sources, ranging from the trol] are known as the Work in two parts."# Buddhist Sadhanamalā to the Vīrasaiva Bāsava Purāņa to the TĀ and Parā- As already noted many times, the goal of the practitioner, in both Hindu paraprakāśika of the Trika Kaula, to the Amanaska Yoga of Gorakhnāth.42 alchemy and hatha yoga, is the attainment of both supernatural powers and This direct comparison, even identification, of initiation by penetration bodily immortality. Nowhere are these concrete goals identified more ex- with a form of alchemical penetration, through the metaphor of physical plicitly than in alchemical depictions of vedhana. Thus, the RA (14.25-36) transmogrification (into gold/into Siva), leads us into the heart of the sec- states: ond field of application of the term vedha: Hindu alchemy. A mercurial pill [guțikā] capable of transmuting one hundred times its mass of base metals into gold [sata-vedba], when held in the mouth 3. Penetration, Transmutation, and Transubstantiation for one month, yields a life span of 4,320,000 years. One thousand- in Siddha Alchemy vedha mercury, held in the mouth for two months, permits one to lives as long as the sun, moon, and stars. Ten thousand-vedha mer- The Kularnava Tantra dates from the same period (tenth through twelfth cury, held in the mouth for three months, yields a lifetime of Indra. centuries) as the RHT, in which we find one of the earliest uses of the term One hundred thousand-vedha mercury, held for four months, yields a vedha[na] in the sense of "transmutation through penetration [by mer- lifetime of Brahma; one million-vedha divine mercury, held for seven cury]." As is common practice in the alchemical literature, the RHT lists months, places one on an equal footing [pada] with Iśvara. Smoke- vedha[na] as the seventeenth and penultimate samskāra. As we have seen, sight-vedha mercury, held in the mouth for eight months, renders the preceding eight alchemical samskāras involve the progressive envelop- one Svayambhu Maheśvara. One becomes the creator, destroyer, and ment by-and absorption of a given metal or mineral-into mercury. At enjoyer [of the universe], a maker of curses and boons, omniscient, this point, a threshold of critical mass and energy is reached, which auto- omnipotent, of subtle and immaculate beauty. Such a man acts at will, matically triggers vedha[na], the "piercing" or transmutation of said metal creates and destroys at will, moves at will, and himself becomes the by mercury, such that the mercury-metal amalgam is immediately trans- universal body [Viśvarüpa] worshipped by all of the gods, including formed into silver or gold. Mercury that is capable of transmuting base Brahmā, Visņu, and Maheśvara. metals into noble silver or gold may thereafter be taken internally. This is Progressions of this sort, in which perfected mercury's powers to trans- the eighteenth and final samskāra, known as śarīra-yoga, "body Work" or mute base metals into ever-increasing amounts of gold and transubstan- "transubstantiation." In this final operation, the alchemist will generally tiate the human body into ever-more exalted Siddha or divine bodies, are

316 317 Chapter Ten Penetration, Perfection, and Immortality

legion in the alchemical sources. Thus, for example, an adamantine pill, duction of excrements are greatly reduced; after six hours, the light of one's formed of a combination of mercury, gold, the "three fruits," the "three soul shines forth; after twelve hours, the light of the ether principle (akāsa- hot substances," and "rock water"45 and plunged in successive oil baths (sā- tattva) shines forth; after a day, one gains supernatural olfactory powers; ranas), has the following effects on base metals and the human body. After after two days, the supernatural gustatory powers; after three days, "tele- one such oil bath, this pill, held in the mouth, yields a life span of a thou- vision" (düradarsana); after four days, the sense of touch at a distance (dūra- sand years; after two oil baths, it transmutes ten thousand times its mass of sparśana); after five days, the sense of hearing at a distance (dūraśravaņa);51 base metals into gold and affords a life span of ten thousand years; after after six days, the arising of the mababuddhi; after seven days, the mystic three, the figures reach one hundred thousand; after four, ten million (and knowledge of the entire universe; after eight days, the cessation of all hun- the power of flight for the alchemist); after five, one hundred million. After ger and thirst; after nine days, supernatural speech (väksiddhi); after ten six oil baths, this pill transmutes one billion times its mass in base metals days, visions of hidden wonders; after eleven days, mind travel; after twelve into gold and affords the alchemist a life span of the same duration as the days, the power to move over the earth at will (bhucaratvam); after thirteen sun and moon and eternal force and virility. After a seventh operation, this days, the power of flight (khecaratvam); after fourteen days, the power to pill is given the name of khecari (the "aviator"): "this was proclaimed by render oneself infinitesimally small animä); after sixteen days, the power to Bhairava himself." He who eats a grain of this each day for one month render oneself infinitely great (mahimā); after eighteen days, infinite grav- becomes a diamond-bodied Siddha, sporting according to his whim with ity (garima); after twenty days, the unbearable lightness of being (laghimā); the other heavenly hosts, and attains Sivahood.46 after twenty-two days, the attainment of all one desires (prapti); after Progressions of a similar order are found in two sources, the one al- twenty-four days, the fulfillment of all desires (prakāmya); after twenty-six chemical (RA) and the other hathayogic (Amanaska Yoga 1.45-98), in which days, supremacy (isitva); after twenty-eight days, one bends the universe to a measure called a pala is employed as a means for calibrating the siddhis one's will (vasitvä);52 after a month, the dawning of release (moksa); after realized by the practitioner. In the RA (18.56-60) account of this progres- nine months, a diamond body and mastery over the element earth; after sion, the nature and quality of the siddhis realized by the alchemist vary eighteen months, mastery over the element water; after three years, mas- according to the number of palas of mercury that has been calcinated in tery over the element fire; after six years, mastery over the element air; mica one consumes. Eating one pala+7 renders one invulnerable to disease; after twelve years, mastery over the element ether. He who practices un- two palas produce an increase in semen; three palas confer heightened interrupted breath retention for twenty-four years gains dominion over powers; eating four palas cures baldness; with five palas, all wrinkles dis- the goddess Sakti and the entire universe and enjoys the siddhi of "Love's appear; six palas afford a telegraphic memory; seven palas destroy eye dis- Body" (kāmarūpa). Such a yogin, established in the mahātattva,53 retains his eases; eight palas endow one with a bird's-eye view (tarksyadrsti) and a life integrity and inviolability even after the universal resorption (pralaya) of span and powers equal to those of the god Brahma; nine palas make one the gods Brahma, Visnu, and Siva. the equal of the gods; ten palas make one a second Siva.48 The pala measure constitutes what we may characterize as a middle- In the Amanaska Yoga discussion of what it calls "salvific" or "stellar range case of the Hindu time-space continuum: the pala is at once a mea- yoga,"49 the length of time one holds one's breath (called the "time of ab- sure of time (24 seconds) and of mass (82.624 grams). This continuum is sorption": laya-kāla) determines the degree of success (siddhi) one realizes, operative at the two endpoints of the scale as well. At the infinitesimal in a mounting progression. Here, the term pala is used as a measure of end of the spectrum is the paramanu, "subatom," whose theoretical mass is time, rather than of mass.50 Laya for one full breath establishes the life force approximately o.00000017 grams,54 but which is also a "space of time," i.e., (präna) in the body; for four breaths, the replenishment of the seven bodily the time it takes for a beam of sunlight to travel across an atom having said constituents (dhatus); for one pala, reduction of in and out breaths. When mass. In its discussion of time and space, the Bhāgavata Purāna joins the the breath is restrained for twelve minutes, the kundalinī straightens; after infinitesimal to the infinite: "The subatom [parama-anu] is indeed [a mea- ninety-six minutes, a trance state is reached in which one has a vision of sure of] time, i.e., the time [it takes the sun] to pass through that which the "drop of light" (tejobindu); after three hours, one's food intake and pro- has the [spatial] measure of a subatom." A homologous case is the greatest

318 319 Chapter Ten Penetration, Perfection, and Immortality indivisible measure of space, which is also the "supergreat" (paramo mahān) penetration. According to the metaphysical teachings of Gorakhnāth, all that exists in the universe, from the most inert inorganic matter to the measure of time.55 Following the logic of this statement, a cosmic eon (kalpa) is the time it takes a theoretical light beam generated by the tran- abstract godhead, is ultimately composed of the same substance-i.e., the five elements.60 Sentient beings, and especially humans, consist of so many scendent absolute (having the form of the sun, in the mundane world) "overlays" of these elements, ranging from earth to ether. Yogic practice, to pass across the entire expanse of bounded space. The "two infinites" of Blaise Pascal, as well as the time-space continuum and light-years of mod- like the ritual technique of bhūtasuddhi, therefore consists of imploding lower elements into hierarchically superior elements until all has been ab- ern astrophysics and Douglas Adams's "Restaurant at the End of the Uni- sorbed into the empty space of the subtle ether. This vacuum of nothing- verse" all appear to have been theoretically anticipated by medieval Indi- ans, whose universe was at once spatial and temporal.56 ness in the heart of being-Gorakhnäth's child crying in the ether-is a There is yet another model for measuring time in the Hindu, and par- "clearing" of pure potential out of which supernatural powers, immortality, and a divine or siddha body may emerge.6! This "immanentism," the notion ticularly the yogic, context. This is the unit of breath called a mātra or that at the core of every being lies its perfect essence, is given an alchemical pala57 which, as we have seen in the context of laya-kala, the "time of breath retention," deploys into life spans equivalent to the duration of the uni- gloss in another of Gorakhnäth's mystic verses: "Just as within every stone lies the indestructible touchstone, so within every metal there is gold."62 verse. Here, we encounter a direct link between breathing and the duration of life. Four "measures" (mätra) of breath equal one pala (twenty-four sec- Of course, the prime technique for imploding the lower constituents of onds); sixty palas equal one ghatikā (twenty-four minutes), and sixty ghati- one's being into the higher is, in the hathayogic tradition, the pierc- kās one day and night. Fifteen days and nights constitute a lunar fortnight, ing (vedha, bhedana) of the cakras, through breath control and retention. As of which two make a human month. Six such months equal a solar semester noted in the laya-kala progression described in the Amanaska Yoga, re- (ayana), of which two make up a human year. One hundred such years are taining the breath for one pala reduces up and down breathing. Diaphrag- the allotted life span of a human being. Nearly every creature is allotted matic retention follows, and after twelve minutes (i.e., thirty palas) the kun- an optimal life span of one hundred years. However, in the case of the dalinī awakens, straightens, and begins her rise through the cakras, piercing them as she mounts toward the cranial vault. An important technique to ancestors, one "year" is the equivalent of 30 human years (thus the ances- tors live 3,000 human years); and a "year" of the gods 360 human years. this yogic end, called mahavedba, the "great penetration," is described in Twelve thousand divine years equal a mabayuga, of which one thousand the Goraksa Paddhati:63 "With one-pointed mind, the yogin should inhale equal a day of Brahma-who, like the mayfly, lives but for a day; i.e., a and retain the subtle breath within his body. Then he should halt the "day of Brahmã" is the equivalent of his hundred-year life span.58 A day of course of the breaths, through the use of the throat lock. Placing the two Brahma is the equivalent of a kalpa, a cosmic eon, of which one hundred palms flat on the ground, he should slowly strike the ground with the buttocks. Thereupon the breath, overflowing the two outer channels, constitute a mabakalpa, or great eon, a life span of Visnu, out of whose navel Brahma arises at the beginning of each kalpa. For the Saivas, how- bursts into the medial channel ... This ... surely generates immortality ... This is mahävedha, the practice of which confers the great siddhis." ever, a life span of Visnu is nothing more than an in breath and out breath of the supreme god Siva. In this perspective, all of time, from the infini- On a conceptual level, the piercing of the cakras effects a stadial resorp- tion or implosion of the lower elements into their higher emanates. Thus, tesimal to the infinite, is a continuum of breath. Extending one's breath is therefore tantamount to extending one's life span.59 when the second cakra (the svadhisthana) is pierced, the element earth- whose support is the lowest, the mūladbara cakra-becomes imploded into the element water, the second element on the hierarchy, and so on, until 4. Cakra Piercing and Ear Piercing in Nath Siddha Practice nothing remains but ether in the cranial vault. It is not a coincidence here that the term kha at once means "empty space," "ether," "the absolute," and "zero" in the Sanskrit language.64 The cranial vault, the locus of the Another conceptual model for effecting the yogin's return to and identifi- ether-both the site at which Siddha techniques for penetration end and cation with the Absolute brings us back to the transformative dynamics of

320 Chapter Ten 321 Penetration, Perfection, and Immortality

that at which mundane existence first begins, when the absolute first pene- trates the human microcosm65-is the zero point at which the two infinities Semen is yoga, semen is what pleases; semen averts the sixty-four diseases/

meet, the point at which "black holes" issue into "white holes."66 This is The rare dude who pierces semen's mystery, he's the creator, he's the di-

precisely the "end" of the RA's process for the creation of the Alchemical vinity."69

Man, which was outlined in an earlier chapter: when all that remains of the Gorakhnäth, the author of the hathayogic Goraksa Paddhati and the mys-

practitioner is the fifth element, ether, in the cranial vault, he arises-out tic banīs just quoted, is also the purported author of the alchemical BhP,

of the cauldron in which his lower being had been dissolved and imploded which in its colophon calls itself a portion of the Goraksa Sambita; in the

into its higher emanates-as a massive, powerful, perfected Siddha.67 Sākta work of the same name, the initiation of penetration is also de-

Parallels of this sort, between the alchemical and hathayogic techniques scribed.70 We therefore find in three works attributed to this great medieval

of the Siddha alchemist, extend to the ability to transmute: the highest Siddha references to all three uses of the term vedba[na]-the alchemical,

form of transmutation, sabda-vedba, is effected when the perfected alche- the yogic, and the tantric-I have been discussing here. This is also the

mist, holding a mercurial pill in his mouth, exhales over base metals. Else- same Gorakh who, in the Nepali royal chronicle we summarized earlier,

where, the yogic and alchemical Siddha is capable of using his bodily secre- initiated the young Prthivinarayan Sah by spitting yogurt he had "re-

tions to "pierce" and transmute base metals into gold.68 Penetration is not, versed" in his subtle yogic body into the young boy's hands. Elsewhere, in

however, solely a matter of breath, or of the implosion of the hierarchized the legend of his own birth and initiation, Gorakh is portrayed as the first

five elements. As we have already seen in the context of initiation by pene- human to have had his ears pierced (or punched, or split, or bored: phat)

tration, there is also a fluid component to this transformative process. by his guru, Matsyendranäth. Armed with this knowledge, it becomes pos-

Here, the technique of mabavedha just reviewed is but a particular instance sible for us to reproduce the symbolic context that made initiation in the

of the more general phenomenon of penetration or piercing that occurs narratives I have reviewed the transformative act that it was.

within the yogic body when the breath, energy, and seed of the yogin- In a number of these accounts, a boy becomes transformed into a king

embodied in the female kundalini serpent-pierce the six cakras. This when he is initiated by a Siddha who either spits in his mouth or bores his

piercing of the cakras-called cakra-vedbana or cakra-bhedana-is also a sex- ears or both. In the first instance, as we have seen, the guru enters and

ual penetration of sorts, albeit with sexual polarities reversed, given that it thereby transforms his disciple's body through the initiation rite known as

is a female kundalinī who awakens, stiffens, rises, even rushes upwards to- vedhamayī dīksā, initiation by penetration. But in the latter practice as well,

wards the cranial vault, the cavity that is the place of the passive male Siva something is being penetrated or pierced. This is the thick cartilage of the

As the kundalinī pierces each of the cakras, great quantities of heat are pro- initiate's ear. This practice of ear splitting, a hallmark of the Nath Siddhas,

duced, which refine and gradually transmute the seed that is the stuff of has also earned them the appellation of "Split-Eared Yogis," the Kānphata

her body: it is this transmuted semen that becomes the nectar that immor- Yogis. As we saw, however, in Banabhatta's seventh-century description of

talizes the yogin who holds it in his cranium. the Kāpālika Bhairavācārya, the kundala or mudra worn through the thick

Just as in the alchemical case, in which the practitioner holds a pill of of the ear was an external sign of Saivite sectarian affiliation even before

refined mercury in the mouth, so in the yogic case, in which his head is the emergence of the Nath Siddhas.

filled with nectar: it is refined, perfected semen (of the god Siva, of the What initiation by penetration and ear boring have in common is that

initiating guru, or of the yogin himself) that transmutes his mortal, gross both open up channels in the subtle body, channels which must be opened

body into an immortal, perfected one. Here, vedha refers as well to the if the practitioner is to succeed in his hathayogic practice. When the guru

transformation of the practitioner himself. We have already seen that the enters into his disciple's body in the form of a drop of saliva, semen, or alchemist, like the tantric initiate, becomes a second Siva. Similar claims sound, it is he who first pierces (bhed) the disciple's six cakras, before exiting

are made for the "transmuted" yogin, especially in the mystic verses attrib- from the latter's mouth back into his own mouth. So too, ear boring, be- uted to Gorakhnäth and others of his sect. So Gorakh says, "Now that sides opening a hole in the disciple's ear, also opens, as if by synecdoche, a you've pierced [bedhya] the lotuses six, go and drink that nectar mix/ ... subtle channel inside the head of the initiate.71 Once open, this channel too becomes a conduit to yogic powers and bodily immortality. In certain

322 323 Chapter Ten Penetration, Perfection, and Immortality

sources, it is further maintained that the guru actually penetrates, puts his 5. Kavi, Siddha, Vidyādhara mouth in the mouth of his disciple, somewhere inside the latter's left inner ear, at the opening of the channel called the "conch" (sankbinīnādi).72 What is the appearance of one who has transformed himself into an al- When we look closer, we find that the subtle body has at least ten chemical Siddha? As described in the RA (12.366-68), "mouths" (mukhas: a better translation would be aperture or orifice), that the kundalini serpent has two mouths, and that mercury itself becomes pos- he is the darling of fair-eyed vixens, slender-waisted with a compact,

sessed of a mouth when it reaches an optimum level of transmutation. This rock-hard body, as inflamed as a rutting bull elephant, as alluring as

multitude of mouths gives rise to all manner of complication in descrip- Love himself. The beau of lustful women, his body is a bolt of light-

tions of processes within the subtle body, as well as to all manner of varia- ning flashing amidst the storm clouds of his curling jet black hair. As

tions on the initiation process, some of which, of an overtly sexual order, immaculate as the chaplain of the gods (Brhaspati), he is an alchemi-

remind us of the perils of poor Cakravarman at the hands (or should we cal wizard (kavi) and wonder-worker. He moves with the imposing

say mouths) of the Dombi temptresses Goosie and Serpentina.73 gait of a great bull; his voice is deep and mellifluous. Like the divine

But so it always is with these Siddha and tantric traditions. It is precisely elephant Airavant, he surges ever forward into the world, indefatig-

this simultaneous play upon multiple levels of correspondence that makes able. Radiant as a lotus pond, he outlives even the sun, moon, and

these systems work. Each instance of penetration recalls another analogous stars of this world. A consummate logician and expert in all the sci-

instance, such that initiation, transmutation, and yogic integration all be- ences, he is a protector of the precepts of virtuous men because he

come a single process; every mouth refers to other mouths. These parallel knows, by his power of inference, what is righteous and proper. The

processes come to exert a cumulative effect as well, with one simultane- heroic [vira] equal of Visnu and Siva,75 he is as enduring as the sun,

ously catalyzing and being catalyzed by all the others. Thus, for example, moon, and sea.

the yogin who combines the internal piercing of his cakras with the prac- A number of terms (siddha, kavi, vīra) and allusions found in this passage tice of holding an all-penetrating mercurial pill in his mouth (gutikā ban- summarize both the historical origins and the ultimate goals of the Siddha dha) and/or the sexual penetration of vajroli mudrā becomes capable of alchemist. The last of these, vira, is a reference to the broader tantric tra- alchemically penetrating all matter with his bodily secretions, thereby dition within which the Siddha theoreticians often situated themselves. transforming minerals to gold and boys to princes or human replicas of The tantric "hero" is a practitioner who, through his initiation, practice, Śiva. Once inside the system, the various permutations on the process are and gnoseological transformation, has transcended the bounded, duality- mutually reinforcing. ridden world of lower creatures (pasus) of this world. The very same The Siddha or tantric initiation rites behind the narratives that were heterodox practices with which the tantric practitioner reaffirms his tran- passed in review in the first part of this chapter were the medieval cognates scendence and absolute freedom-caste-free sexual intercourse, the con- of the Vedic royal consecration of old. In the medieval cases, the institution sumption of forbidden power substances-are those which otherwise con- of priesthood remains, with the Vedic sacrificial priest being replaced by a demn the unwashed masses to hell.76 As we have noted, intercourse with tantric guru. The ritual reiteration of the priesthood's seniority and superi- Siddha maidens, yoginīs, even goddesses, is a desideratum for the Siddha ority vis-à-vis royalty is also retained. However, instead of the rather tame practitioner and a recurring theme in every tantric alchemical work.77 procedure of tapping an unpunishable (adandya) king on the shoulder with More specific to these Siddha traditions is the goal of becoming a kavi, the rod of punishment (danda), the tantric initiator actually renders himself a wizard. As indicated in the opening chapter of this book, the sense of the the biological father of his royal disciple, using tantric techniques to trans- term, as it is employed here and in alchemical works, differs from the more form the latter's mundane body into a perfected, tantric one and thereby broadly accepted usage of "poet." The Vedic kavis, of which the Asura afford him dominion over this world and worlds beyond. For the Siddha, chaplain Kavya Usanas was the paragon, were not only wordsmiths (poets) transforming a boy into a king is, in such a situation, child's play.74 but also death-defying, wonder-working wizards. It is this sense of the term

324 325 Chapter Ten Penetration, Perfection, and Immortality

that is intended here and in a number of Siddha works that evoke the kavi. Here, it should also be recalled that the "old" supernatural power of rasa- the final verses of the work, was adapted into Kaula ritual.84 Of course, the

rasayana was one that was to be wrested from the subterranean Asuras. revealer of this doctrine in the present age is none other than Macchanda/

This is precisely what the young Kaca did when he won the formula for Matsyendra himself, making him the ideal model for all humans who

immortality from Kavya Uśanas, who was, by virtue of his knowledge of would raise themselves up to the level of divine Siddha.85

healing and incantations of immortality, the Asuras' greatest wizard. The Matsyendra twice mentions Vidyadharīs (the female counterpart of the

Ayurvedic healers of Bengal continue to be called kavirājas down to the Vidyādharas) in his KJiN; in both cases he describes techniques for at-

present day.78 This goal or supernatural power of kavi-hood, shared by tracting and sexually exciting Vidyādharīs and other female demigods. His

alchemical and hathayogic sources alike, is evoked in the KJnN, KCM, sole mention of the Vidyadharas places them in the company of a number

RA, Goraksa Sataka, KhV, and Siva Sambitā.79 The successful alchemist is a of male demigods, all of whom the practitioner is to visualize within his

kavi because he, like the Asura Kāvya Uśanas (Usanas, son of a kavi) or the body.86

Vedic singer of hymns, is capable, through his mystic omniscience, to I have already reviewed a number of cases in which the matter of becom-

shape his own universal order, bend nature to his will, and thereby realize ing a Siddha or Vidyadhara, or gaining mastery over these aerial spirits,

wealth, invincibility, and immortality for himself.80 was broached in early Saiva literature. In the Candakausika of Ksemīśvara,

The RA compares the purity of the perfected Siddha to that of Brhas- the final apotheosis of Hariscandra, superintended by the god Dharma in

pati, the chaplain of the gods and divine counterpart of Kāvya Uśanas, a the form of a Kapalika alchemist, is effected through an aerial car, brought

comparison that leads us from the "subterranean" or "nocturnal" powers to him by the Vidyadharas.87 In Banabhatta's Harsacarita, the Pāsupata

of the kavi to those more closely identified with the light-filled worlds of Bhairavācarya's goal is to transform himself into the lord of the Vidya-

the gods and demigods. It is here that the notion of bodily transubstantia- dharas.88 So too, a Saiva "Great Vow-Taker" seeks to become a Vidyādhara in the Kathāsaritsāgara tale of Devadatta the gambler.89 tion, from human practitioner to divine Siddha or Vidyādhara, comes to the fore. Here, the intermingling, in the Siddha lists reviewed in chapter This theme, of transforming oneself into a Siddha or Vidyadhara, is also

four, of gods, demigods, and perfected humans, is reflective of the essential a standard fixture of the alchemical literature. So we read in the twelfth-

transformation realized by the Siddha alchemist. By becoming an immor- century RM that one who has succeeded in fully transmuting his body-

tal, invincible "second Siva," he literally enters the ranks of the semidivine by jumping into a cauldron of boiling oil, mercury, and other alchemical

Siddhas and Vidyädharas, if not of the gods themselves. ingredients-swoons and then recovers to find himself transformed into

There is ample evidence for the fact that cults of Siddhas and Vidyā- the three-eyed, four-armed Gananatha (lord of Siva's host). The Siddhas

dharas predated the emergence of those human practitioners and religious and Vidyadharas come en masse to view the transformed alchemist, and

orders who called themselves "Siddhas."81 In essence, the genius of these together with these denizens of the atmospheric region he flies through

orders and their authors was to appropriate for themselves these pre- the air and is brought before Mahesvara (Siva), whom he worships.90 In the

existing traditions and incorporate them into their own syntheses. This is origin myth of sulfur, presented in the same source, the Goddess is de-

transparently the case with Matsyendranäth, revered as the Siddha founder picted as sporting with Vidyadharīs, Siddha maidens, and a host of other

of a number of tantric cults, whose KJnN is replete with references to Sidd- semidivine women.91 According to the MBbT, the practitioner who follows

has both human and divine. It is to the latter that he clearly refers when its instructions gains siddhis and becomes a "Siddha who is the manifest

he states that "the planets, serpents, gods, yoginīs, and Siddhas, all of them equal of Siva."92 So too, the Siva Samhita, a relatively late Nath Siddha

worshipped, themselves worship the eight Mother Goddesses."82 Human guide to hatha yoga, ends with the promise that the yogin may see the (di-

Siddhas are evoked a few verses later, when Matsyendra speaks of the as- vine) Siddhas and gain control over the Vidyadharas through his practice.93

sembly (pankti) of gurus, Siddhas, and yoginīs.83 The four lineage Siddhas Finally, the AK concludes its description of the "lithe virgin on horseback"

(kula-siddbas) are semidivine intermediaries who have revealed the mystic technique94 for the extraction of mercury with the following statement:

doctrine to the world in the four cosmic ages: their worship, described in "Mercury is found in all of those places that the Siddhas and Vidyadharas caused it to fall [by using that technique]."95

326 327 Chapter Ten Penetration, Perfection, and Immortality

It is in the alchemical RA that we find the most extended references to 6. Mountains of Wisdom the matter of becoming a Siddha. In its discussion of "revivifying water" (sañjīvanījalam), it relates that the alchemist who has drunk three measures A number of tantric sources classify the Siddhas into three groups: gods, of this water swoons and then awakens to find himself transformed and semidivine Vedic sages, and humans who have realized a perfected (siddha) possessed of supernatural powers. He is able to "see into cracks in the earth body.100 In other words, a number of the Siddhas found in the lists reviewed [to find buried treasure] throughout the seven underworlds [haunts of the in chapter four began their ongoing careers as gods, while others are hu- Asuras]; he is invulnerable to the onslaughts of gods and antigods alike." mans who acceded to a divine station through their practice of Siddha After further treating this water with mercury, realgar, and the three noble alchemy. metals, "he suddenly disappears [from human sight] and becomes the lord Nāgārjuna, a Rasa Siddha who clearly belongs to the last of these three of the Vidyadharas, surrounded by a circle of Siddha maidens for a period groups, nonetheless shows divine pretensions when he states in his RM that of fourteen kalpas."% The same work concludes its description of khecari "in the perfection of rasa, I shall make the world devoid of poverty and jarana ("calcination of flying [mercury]") by stating that the alchemist who disease."101 Here, rasayana, in both the medical and alchemical senses of ingests said mercury by means of this technique is immediately uplifted the term, is the only means by which humans may enjoy the immortality into the presence of the gods, Siddhas, and Vidyadharas, with whom he that the gods, demigods, and denizens of the underworld possess by their flies through the air at will.97 The work concludes on a similar note:98 very nature. This appears to be the intended sense of a passage from the venerable Caraka Samhitā (6.1.78): "Like the ambrosia (amrta) of the im- He mounts into an aerial car made of divine gold studded with divine mortal gods and the nectar (sudha) of the serpents, so the principle of rasā- gemstones and rubies and filled with flowered garlands and banners yana was to the great seers of old ... he who uses rasayana in the prescribed and the roar of conches and musical instruments. Thereupon, a di- manner not only obtains long life on earth but also, upon death, betakes vine maiden, a singer of nymphen melodies and an alluring dancer, himself to that light-filled realm within which the divine sages dwell." 102 decked out in divine finery and garlands-a lusty beauty and a veri- There are thus three sorts of elixirs of immortality, the access to which table image of Love in a female form-comes to him and takes the depends upon one's ontological status; those humans, however, who make consummate practitioner to dwell in the world of the Siddhas. Then, use of the elixir available to them can hope to accede, after death, to a level after having bathed, wined and dined him, and clothed him in divine of existence normally reserved for a certain class of deities. This notion finery, the love-starved Siddha maiden sports with him for hundreds of an exclusive afterlife in the company of semidivine beings, already and thousands of years. glimpsed in this medical source, becomes a central concern of the later Siddha traditions. This is the purport of the RA's final word on alchemy, The passage and the entire work end on the statement, "When all the which seems to imply that in order for the alchemist to fully join the ranks fixed and moving beings in the universe have been annihilated in that ter- of the Siddhas and Wizards, he must physically enter into his alchemical rible flood [of universal dissolution], the Siddha is absorbed into the same apparatus. This is a principle also found in a number of Chinese alchemical place as are the gods."9 The place in question here is, according to the sources-sources in the light of which this and a number of other elements most widely held Hindu beliefs concerning the fate of the universe, one of of Hindu alchemical lore come to stand out in higher relief. First of all, the two highest "worlds" or "heavens" (tapoloka, and siddhaloka or brabma- transforming oneself into an immortal by entering into one's alchemical loka), which together serve as a holding tank of sorts for liberated souls. apparatus is a commonplace of Taoist alchemy. In the Chinese case, the Suspended high above the general conflagration, they are saved from uni- apparatus in question, called bu-lu, was formed out of two superimposed versal dissolution and, most important, the necessity of reincarnation into spherical gourds, a configuration at once identified with the head and torso a transmigrating body upon the reordering of the cosmic egg, by Brahmā, of the subtle body, a double mountain located to the far east or west of at the beginning of the next cosmic age. Once one accedes to these sublime China, and an abode of the Immortals (hsien).103 Entering into one's al- levels, one need never return. chemical apparatus is, in these sources, somewhat less dramatic than the

328 329 Chapter Ten Penetration, Perfection, and Immortality

operation described in the RA: its results are, however, wholly as trans- hemmed in by a ring of peaks so as to have the form of a basin or saucer;107 formative as those promised in the Hindu alchemical tradition. Taoist al- the Candakausika (4.34) states that this mountain abode of the Siddhas may chemical sources dating from the first half of the first millennium A.D. are be reached through alchemical practice. particularly rich in legends on this subject, one of which was reproduced It is perhaps incorrect to state that the Hindus made no explicit iden- in chapter eight, on the subject of a certain Che Ts'ouen. It will be recalled tification between their alchemical wizards or immortals, their alchemical that the vessel this Taoist alchemist suspended from his roof beam, called apparatus, and their sacred mountains. The Sanskrit term dhara, generated bu or hu-lu, was at night transformed into Heaven and Earth. Che Ts'ouen, from the root dhr, means, as has already been noted, "bearer" or "recipi- who passed his nights there, called himself "Heaven in a [Double] Gourd" ent." Another sense of dhr, however, is "to support," a meaning that gener- (bu-t'ien); people called him the "Old Man of the Gourd." Following this ates another reading of dhara: a mountain (dhara) is that which supports he realized the Tao (and became an immortal).104 (dbr) the earth or the mineral riches within the earth. In this case, vidyā- In the Chinese sources, these identifications-between a bicameral dhara may be read as "Mountain of Wisdom," while the Vidyādharas, the alchemical apparatus, a configuration within the subtle body, a double Wizards, may be considered to be not only the denizens of such moun- mountain, and an abode of the immortals-are made much more explicitly tains, but also the mountains themselves. What we are suggesting here is than they are in the Hindu sources reviewed to this point. To be sure, the that behind the medieval Indian cults of divine Siddhas and Vidyādharas Hindu use of the term vidyadhara, Wizard, applies equally to a bicameral as denizens of mountains there lay a more archaic cult of these mountains apparatus, a hathayogic technique, and the mountain dwelling alchemists themselves as a group of demigods. who mastered them. What appears to be lacking in the Hindu material is A number of works have much to say on this subject. An important the explicit identification of these semidivine figures and alchemical appa- example is the great tantric opus entitled the Svacchanda Tantra, much ratus with these sacred mountains themselves, an identification which the of whose cosmology is adapted nearly verbatim by Abhinavagupta in Taoists make with especial reference to two immortal abodes. These are his TA.108 Here, in an enumeration of the atmospheric levels located above the mountains H'un-lun and K'un-lun, located to the far east and west of the terrestrial disk and separated from one another by distances of either China respectively. The names of both of these peaks are derived from the one hundred or five hundred yojanas, this source describes cloud masses same root (hu) as the term for the double-chambered gourd of Taoist al- (megha) (1) that strike down trees on earth with thunderbolts; (2) that cause chemy: more than this, both of these mythic mountains retain the shape rains of fishes, frogs, and turtles to fall; and (3) that cause disease-inducing of the alchemical apparatus: H'un-lun has the form of two superimposed poison rains to fall (and into which are born the subgroup of embodied spheres, while K'un-lun that of two superimposed cones joined at their beings called the Plagues; in this multitude dwell the divine ghouls who apex. These mountains are, moreover, situated within the subtle body of form Skanda's retinue, as well as the Obstructors, who are born of Siva the alchemist, in his lower abdomen and head, respectively. As such, they himself). Still paraphrasing the Svacchanda Tantra, Abhinavagupta con- are further identified as the Gates or Wells of Earth (H'un-lun) and tinues: Heaven (K'un-lun).105 Now, the Hindus also know of at least one mythic double mountain: Five hundred yojanas higher is [the abode of] the Wind [named] this is Meru, the pivot of the Hindu universe, which, located to the north "Lightning-Streak." Here at [the level of] "Lightning-Streak" are of India, has the form of two cones, the one inverted and the other upright, stationed ... the "lowest-level Vidyadharas." These are beings who, joined at their tapered ends.106 Like the Chinese Hun-lun, Meru has its when in the [prior] form of human wizards (vidyāpauruse) carried out microcosmic homologue as well: it is identified with the spinal column cremation ground-related practices. When they died, that siddhi (called the meru-danda, "Meru rod") of the subtle body, along whose ver- [rendered them] Siddhas, stationed in the midst of the "Lightning- tical axis the nectar of immortality is drawn upwards. General descriptions Streak" wind ... 10° Five hundred yojanas higher ... there at Raivata of Meru locate gods and demigods of the likes of the Vidyadharas and itself are the primal Siddhas (adisiddhab) [named] Yellow Orpiment, Siddhas on its slopes and describe its summit as an extensive plateau, Black Antimony, and Mercury Ash.110

330 331 Chapter Ten Penetration, Perfection, and Immortality This passage goes on to describe ever-higher levels, inhabited by supe- ground for their Siddha appellation, in spite of an increasing Saiva (among rior (visesa) Vidyādharas who, together with the celestial musicians, sing the Hindu Siddhas) influence that never ceased to diminish and dilute the the praises of Parameśvara, the highest god; and still higher levels in which Siddha coloring of their gnosis. So it is that we find quite nearly as many the most elevated (uttama) Vidyadharas are stationed. We will, however, references to "becoming a second Siva" as to becoming an atmospheric dwell for a moment upon the particular case of Raivata and the clearly Siddha or Vidyadhara; in certain sources, the issue is resolved by trans- alchemical siddhis and Siddhas with which it is associated. Concerning the forming those same semidivine Siddhas and Vidyadharas into so many latter, I first reproduce the SvT passage upon which Abhinavagupta based members of Siva's celestial retinue. his description: "On Raivata are indeed stationed those great-souled Sid- In an earlier chapter, I argued that certain mountaintops, lofty pivots dhas. Having undertaken the practices [dealing] in yellow orpiment, the between the worlds of the gods and demigods and the world of men, have black antimony [of invisibility], the ash [of transmutation], the shoes [of been singled out since the medieval period as sites at which to realize the magical flight], the hairy skin [of the tiger, goat, or dog worn or carried by various siddhis that enabled one to become a Siddha. Among these were the Saiva "Vow-Taker" or renouncer], etc., these great-souled ones became Srisailam, Kedārnāth, Mount Abu, and Girnar. This brings us back to the Siddhas, having Love's body (kāmarūpināb)."111 passages just cited from the SuT and the TA which, in the midst of their Here we can see that, whereas the SvT evokes a certain number of min- descriptions of atmospheric levels located thousands of miles above the eral preparations instrumentally-to explain that the Siddhas of Raivata earth's surface (a yojana equals approximately nine miles), suddenly present are those beings who, while human, gained the siddhis of invisibility, trans- the reader with a terrestrial toponym which they identify, precisely, with mutation, magical flight, etc., through the practices of alchemy and renun- the alchemical "primal Siddhas" Yellow Orpiment, Black Collyrium, and ciation-Abhinavagupta transforms these alchemical staples into a group Mercury Ash. This is the toponym Raivata which was in fact a medieval of demigods whom he terms the "primal Siddhas." Here we are reminded name for the cluster of peaks known today as Girnar, in the Junagadh dis- of the Siddhas named Fire, Sun, and Moon who, according to the Kubji- trict of Gujarat. In praise of this site, the Jain "Raivatacala Mahatmya" calls kānityāhnikatilaka, aided Srīnātha in founding the kula, "at the beginning it the fifth of the twenty-one Jain siddhadris (Siddha mountains) and states of the Kali Yuga," at a site called Candrapurī.112 In both cases, these found- that "[here] sages who have ceased to eat and who pass their days in devo- ing Siddhas are nothing other than elements of the Siddha gnosis itself tion ... worship Nemīnäth. Here divine nymphs and numerous heavenly which, like the Nine Näths, are hypostasized into abstract deities. Homol- beings-Gandharvas, Siddhas, and Vidyadharas, etc .- always worship ogous to this ambiguous treatment of the Siddhas, who are here portrayed Nemīnāth."114 A number of Puränas beginning with the ca. ninth-century as humans and here as atmospheric demigods, is that accorded to the Wiz- Matsya Purana115 also devote long descriptions to the site, which they term ards. In these sources, it is clearly stated that the "lowest level of Vidyād- Raivataka. Two to three centuries later than the Matsya, the Skanda Purāna haras" is composed of those beings who, when still human, had been Vid- devotes fifteen chapters to the wonders this site, which it alternatively calls yāpurușas ("men of occult wisdom") whose cremation-ground practices Raivāta or Vastrāpatha.116 had won them an atmospheric station after death."3 It is to this level that We clearly appear to be in the presence, in this wide array of sources, such figures as Bānabhatta's Bhairavācārya would have acceded in the me- of a direct identification of Girnar as both a terrestrial site to which human dieval literature. experts in the esoteric sciences come to perfect themselves through Siddha These descriptions are for us proofs of what we have been arguing techniques and an atmospheric or celestial site at which they dwell in their throughout this book: that the human practitioners of the Siddha disci- definitively transformed state of semidivine Siddhas. This pedigree of Gir- plines of alchemy, hatha yoga, and erotico-mystical ritual considered them- nar goes back further still, mentioned as it is by the seventh-century selves to be divine or semidivine Siddhas in potentia and that beyond the Hsuan-tsang, who describes a meeting there with "rishis endued with spir- supernatural powers and bodily immortality that were their immediate itual faculties [siddhis]."117 It is also present in earlier Hindu texts, but here goals lay the higher end of apotheosis to the loftier realms of the Siddhas by yet another name, this being Gomanta."18 We find this early toponym and Vidyädharas. This archaic component of their practice was the prime for Girnar mentioned once in the Mahabharata,119 in the context of the

333 332 Chapter Ten Penetration, Perfection, and Immortality

episode of Jarāsandha, a regicide king and very early devotee of Rudra- Data from other regions of the Indian subcontinent appear to support Siva, who hailed from the Kathiawad region of present-day Gujarat.120 this argument. Gorakh, who is called a Vidyädhara in the "Song of Manik A much more detailed description of the site is given in the ca. fifth- Candra" is depicted in the "Legend of Püran Bhagat" as flying through the century Harivamsa "appendix" to the epic, which also relates it to the figure air at the head of fifty-two hundred visible and invisible disciples.126 Popu- of Jarasandha.121 While this appears to be no different from a great number lar tradition maintains that the founding Nath Siddhas are still living of other praises of heavenly mountains, this text is important for two rea- in the Himalayas, as the guardian spirits of the Himalayan peaks. In the sons: on the one hand, the mountain in question is Girnar; on the other, mountainous Gulmi district of central Nepal, a nondescript "god of the this is perhaps the earliest Hindu source to bring the Siddhas and Vidyā- summit" is named "Siddha."127 In the mountainous Chamba district of the dharas "down to earth" and to make them the inhabitants of a lofty peak Punjab, generally nameless "Siddhs" are worshipped, in the same fashion rather than the atmospheric regions. "The [mountain] called Gomanta, a as serpents and minor goddesses, in rustic temples; an exception is Gugga solitary heavenly peak surrounded by a group of [lesser] peaks, is difficult Mundalikh Siddh, identified with an eleventh-century historical figure to scale, even by the Sky-goers ... its two highest horns have the form named Gūgā Chauhān and with the Nath Siddha known as Guru Gūga or of two shining gods ... 122 The interior of this mountain is frequented Gūgā Pīr, whose shrine is located at Shālū in Himgari.128 As noted in chap- by Siddhas, Caranas, and Raksasas, and the surface of the peak is ever ter four, the "historical" Gorakhnäth is said to have discovered a shrine to thronged with hosts of Vidyādharas."123 the Nepali (Gurkha) deity Gorakh, into whose service he devoted himself. The Girnar peak which the Jains identify as Neminath has long been Over time, this mountain godling and the human yogin became fused into known to Hindu pilgrims by the name of Dattatreya,124 the semidivine a single figure: Gorakhnäth, founder of the Näth sampradāya and the site founder and leader of the Nine Näths of western Indian, especially Mahar- of Gorakhpur.129 ashtran tradition. Dattatreya is in fact one of a pair of rocky crags that In the light of this evidence, we must conclude that such historical Sid- dominate the formation of Girnar, the other being Gorakh: these twin dhas as Gorakh, Dattātreya, and Nāgārjuna, whose acts and lives defined peaks, having heights of 3,450 and 3,470 feet respectively, are by far the the religious landscape of a certain Gupta and post-Gupta India, became highest and most impressive features of the rim of mountains that form a fused in the Hindu imagination with certain classes of gods and demigods, basin some six miles in circumference. Girnar has furthermore constituted in whose number the divine Siddhas and Vidyadharas must be counted; one of the most important centers of Nath Siddha activity in western India, and that it was at such peaks as Girnar, in western India, the heartland of as evidenced in references to it, from the fourteenth century onwards, in early Saivism, that such identifications first emerged.130 While we cannot legends concerning their founders.125 In an earlier chapter I argued, on the say to what extent these figures were and are identified with the sacred subject of these twin peaks, that they were called "Gorakh" and "Dattā- peaks themselves, my reading of Vidyadhara as "Mountain of Wisdom" treya" not because they featured shrines to these Näth Siddhas but because should not be entirely far-fetched, especially in the light of the Harivamsa they were these Siddhas, i.e., that the bedrock of the presence and ven- description of Gomanta. eration of "human" Nath Siddhas at Girnar were the semidivine Siddhas This peak, which I have identified with Girnar, is said to be inhabited whose ranks the latter succeeded in joining through their practice. The both inside and out by Siddhas and Vidyādharas. Like many sacred moun- semidivine Siddhas dwelling inside Gomanta (within the basin formed tains, Girnar is a site riddled with caves, of which at least two are identified by the rim of peaks?) and the Vidyadharas dwelling on its surface (on with Nath Siddhas (Bhartrhari and Dattatreya); and what is a mountain the outer slopes of those peaks?) preceded the likes of the twelfth- to cave, if not the macrocosmic replica of the cranial vault of the meditat- thirteenth-century Gorakh and Dattatreya, who replaced them, and ing yogin, the tumulus (samädhi) in which deceased yogins are interred, or thereby became the demigods identified with its two highest twin peaks. the upper chamber of an alchemical apparatus within which the alchemist In this light, it is tantalizing to note that the Kulārnava Tantra (6.66) lists transforms himself into the opus alcbymicum? As in the Taoist case, the Dattātreya and Raivataka in immediate succession as two of the ten gurus Möbius universe of the Siddhas is so constructed as to permit its prac- of the Kaula siddhaugha. titioners at once to identify cosmic mountains with their own subtle bodies

334 Chapter Ten

and to enter into those mountains to realize the final end of their practice, EPILOGUE the transformation into the semidivine denizens of those peaks. So it is that these myriad allusions, found in a wide array of Saiva, Sid- dha, tantric, and Jain sources, are so many literary vestiges of an archaic The Siddha Legacy in Modern India cult of divine Siddhas and Vidyadharas who, like the immortals (hsien) of Taoism, came to be joined in their ranks by heroic humans (Pasupatas, Kapalikas, Nath Siddhas, and Rasa Siddhas) who, through their dangerous "My dear chap, I ask you!" said Masood, as if it were my fault. and difficult trials, transcended their human condition. This is the Siddha Those sadhus would sometimes come to Dewas and bless the foundation of tantrism: the archaic goal of gaining power over divine Sid- Palace, and demand a hundred rupees each. Malarao would dha and Vidyadhara wizards and nymphs funneled into the tantric cults of speak as fair as he could and give each of them one rupee. the yoginis, which were in turn internalized into the practices of batha yoga They then cursed the Palace and returned to Ujjain.' and alchemy as practiced by superhuman Nath and Rasa Siddhas and the more refined and abstract ritual practices of "high" Tantra. The medieval Siddhas of India were self-styled imitators of divine and semidivine Siddhas who, through their conjoined practice of alchemy, hatba yoga, and tantric ritual (erotico-mystical or "sublimated," when it took the form of worship of the divine Siddhas), sought to join the ranks of the latter at the end of their practice (sadhana) or at the end of their lives. Over the centuries, the cults of the divine Siddhas and Vidyādharas, denizens of the atmospheric regions or of lofty peaks, have gradually diminished and are today found only in remote mountain regions of the Indian subcontinent. In other parts of India, these demigods have been supplanted by or conflated with such historical figures as Matsyendra- nāth, Gorakhnāth, Nagarjuna, or Dattātreya, figures who self-consciously aligned themselves with the former, taking their names as honorary titles or indications of various degrees of realization in their mystic arts.2 A case in point is "Gorakhnäth" who, having discovered the shrine of the godling Gorkha (divinity of the Gurkhas of the Himalayan region of Gurkha in Nepal), took the name of Gorakhnath, i.e., "he whose Lord is Gorakh."3 Such a conceptual shift-from "shrine served by a holy man' to "shrine of the holy man" himself-was effected in the Indian popular imagination, and in the generation of legends that grew up around this composite figure, in the same way as such has occurred in Europe. A local godling, identified with a spring of healing waters or one or another geo- graphic anomaly gradually becomes transformed into a "saint" whose relics are the source of the magic that cures the sick or drives away plagues. In India, the nocturnal sight of hermits' fires, tiny points of light illumi- nating the brooding silhouettes of dark mountains, would also have fueled an identification of the semidivine Siddhas with their human emulators.

335

336 337 Epilogue The Siddha Legacy in Modern India Today, the majority of those dark hillsides are crisscrossed by metaled ent that he could also produce gold by the same procedure, by merely roads, over which buses full of pilgrims and tourists travel. The hermits substituting ammonium chloride for borax. Later on, at Birla's insti- have withdrawn deeper into the mountains or have disappeared altogether gation, he produced gold in this way. He continued to make gold in from the sight of men. Their departure has left a void in the modern Indian this way, at the rate of three grams per week, to cover his laboratory soul, perhaps not unlike that which has marked the American soul since and personal expenses. the disappearance of the frontier some one hundred years ago. Something fundamentally real, something whose mere existence has been a millen- Marble plaques, in the Visvanath Temple on the campus of Ben- narian source of spiritual solace, has been lost or is in the process of being ares Hindu University and at the Birla Temple in Delhi, bear witness to lost. How has twentieth-century India reacted to the "disenchantment of Shastri's alchemical feats and state that the gold he manufactured over time totaled seventy thousand rupees in value. I was told by another Banarsi the world"? Ayurvedic scholar, Siddhinandan Misra, that Shastri used up all his gold playing the horses and that he died of poisoning by persons attempting to extort his secrets from him.5 I was unable to corroborate either of these 1. Transmutation in the Twentieth Century statements. Similar accounts of alchemical aurifaction have continued to appear I have met Siddhas who claimed to be alchemists but who would not show periodically in the Indian press. The 8 June 1968 issue of the Navabharat me their powers. I met many more who claimed to know alchemists with- Times ran a story, out of Ahmedabad, concerning an Ayurvedic pharmacist out ever being able to divulge to me their exact whereabouts. There was, named A. C. Acharya, who had produced pure gold from mercury in four however, a well-documented case of transmutation, carried out in 1940 by days, at Jamnagar (Gujarat). This story was corroborated by Siddhinandan an Ayurvedic pharmacist in a Benares Hindu University laboratory before Misra, who added that the experiment was carried out under the eyes of a group of scholars. One of the last surviving members of this group, Yadu- ten goldsmiths, ten chemists, and six government ministers. Dharm Yug's nandan Upadhyaya, gave me an account of the experiment at his Benares 7 September 1975 edition states that the "secret ingredient" used in these home-cum-Ayurvedic dispensary in the spring of 1984: experiments was "perfected mercury" (siddhasūta).6 Krishna Pal Shastri, a vaidya from Jamnagar, in Gujarat, performed Even if authentic alchemists have become a scarce commodity in the experiment in the laboratory of Benares Hindu University chem- twentieth-century India, the Hindu alchemical tradition seems to have left istry professor Phaldevasahaya Varma, in the presence of nine or ten its mark on the Indian psyche in a number of often-unexpected ways. One Ayurvedic scholars and practitioners and the great industrialist Birla. of these is a tendency to attribute a certain order of consciousness to metals. Shastri hollowed out a soap nut, which he filled with mercury, two Such a tendency ought perhaps not to surprise us, if we understand tantric or three grams of borax, and a grain of a secret powder. He sealed thought in the same way as does Sanjukta Gupta when she states that: the nut with a paste of lime and molasses and put it inside a crucible, ultimately the conscious bits of the universe, like stones, are also God which he placed on a charcoal fire. He fanned the fire until the nut and hence consciousness, but a consciousness that has decided to inside the crucible began to burn. When the smoke cleared, he split conceal itself (atmasamkoca). Here we come to the double conceal- the nut open with an iron wedge. Inside was a metal that looked like ment which God decides on; firstly, He conceals the fact that His silver. Half of this metal was taken by Professor Varma and the other true form is identical with the individual soul; and secondly He con- half by Mr. Birla. Varma tested it at BHU and Birla at one of his ceals His true nature as consciousness to manifest Himself as uncon- firm's laboratories in Calcutta. In both cases, the metal tested out as scious phenomena. The world of the Tantric, then, is ultimately all pure silver, with only the spectroscopy showing a slight variation God, but it contains a vast range of things, from things as gross as from that of natural silver. Sastri had informed those who were pres- stones to things as subtle as God.7

338 339 Epilogue The Siddba Legacy in Modern India

A number of twentieth-century Indian scientists have been sympathetic tain sources identify the Nine Naths with the eight metals or nine planets; to this viewpoint. In his renowned Autobiograpby of a Yogi, Yogananda gives in an interview he granted me in 1984, Mangalraj Joshi, the royal astrolo- an account of the Bengali scientist Jagadis Bose, who demonstrated to him ger to King Birendra of Nepal, identified the nine planets with the eight an apparatus he had invented in the 1930s:8 metals.10

Graphs of my delicate apparatus have proved that trees possess a cir- Of greater and more wide-ranging impact than pseudoscientific theo- culatory system; their sap movements correspond to the blood pres- rizing of this order are certain effects the medieval Siddha traditions appear sure of animal bodies ... The more deeply we perceive, the more to have had-if only by their transformation or absence-on a number striking becomes the evidence that a uniform plan links every form of medical notions and political strategies in late twentieth-century India. in manifold Nature ... I shall show you experiments on a piece of These are the subject of the next two sections of this chapter. tin. The life-force in metals responds adversely or beneficially to stimuli. Ink markings will register the various reactions. [Yogananda then continues, in his own voice:] Deeply engrossed, I watched the graph that recorded the characteristic waves of atomic structure. 2. Impotence: The Ravages of Gupta Rog When the professor applied chloroform to the tin, the vibratory writ- ings stopped. They recommenced as the metal slowly regained its High in the snow-clad Himalayas, in a cave above Joshimath, lives a five normal state. My companion dispensed a poisonous chemical. Simul- hundred-year-old man whose eyebrows have grown down over his eyes. taneously with the quivering end of the tin, the needle dramatically The four hundred-year-old Sundarnäth Siddha, an expert in the alchemi- wrote in the chart a death notice. cal arts living in the hills above Kathmandu, disappeared and reappeared at will before definitively disappearing in the 1970s." The alchemist Na- A similar body of assumptions appears to underlie the theories of T. R. gārjuna lived for eight hundred years before giving up his head and life at Anantharaman, who was, when I interviewed him in December of 1984, the request of a young prince. Gorakhnath, the yogic superman, has stood the dean of the prestigious Engineering College at Benares Hindu Univer- beside kings and heroes for millenia, abandoning one body for another at sity. In a lecture entitled "Transformations-Metallurgical and Mental," will. He and his brother Nath Siddhas, immortal guardians of the Himala- Professor Anantharaman proposes that the four states of human conscious- yan peaks, are as old as the mountains themselves. In India, whose recorded ness (matter, life, mind, and supermind) correspond to the four metallur- traditions of the deeds of gods and humans are among the oldest on the gical phases (gas, liquid, single-phase solid, multiphase solid). Here, the planet, examplary humans have ever rivaled the gods in their longevity and three culminating stages of Pātañjala yoga correspond to phase transfor- power. Yet this is the same India upon whose sweltering plains "young- mations (nucleation and growth) in the metallurgical solid phase. In such men" are warned in "clinical literature" that they will "look old at the age transformations, "embryos" or small nuclei are generated, which are ca- of 25 [through] the horrors of the wastage of semen ... Many youngmen pable of further growth and steady increase in size until transformation have sex many times at night and thus waste this essence of life recklessly. occurs. This process of nucleation is to be compared, in yogic practice, to With small production and heavy drainage, supply will exhaust soon and the realization of samadhi. In the end, the transformations undergone by critical consequences will have to be faced."12 the meditating yogin are to be viewed as so many monotectoid transforma- What connections can we draw between the grandeur and the squalor, tions, effected on the level of human consciousness." between immortality in the highlands and misspent youth in the lowlands On a more popular level, we find the pervasive notion in India that the of India? A glance at Ayurvedic theories of digestion, metabolism, and eight metals (astadhätu) are intimately related to the movements and astro- the production of sexual fluids provides us with a key to this puzzle. The logical effects of the nine celestial bodies (navagraha). By wearing an eight- Upanisads inform us that humans are constantly being cooked by the fires metal ring, one wards off the baneful influences of such dread planets as of time; and as the myth of King Moon demonstrates, the cooking is never Saturn, while enhancing the salutory effects of the auspicious planets. Cer- so fierce as when man sheds his seed, his very staff of life.This sentiment

340 341 Epilogue The Siddha Legacy in Modern India

is expressed in the beautiful poetry of Tayumanavar, a Tamil Saiva poet process, is in constant danger of being burned up and entirely consumed who lived in the first half of the eighteenth century: in a fiery uterine ocean of female sexuality.

Ecstatically, you think, "sex is bliss." The mythic ground for this cultural obsession, or for what Gananath Obeyesekere has termed cultural diseases involving the loss of bodily fluids, This embracing becomes more frequent, is the Vedic myth of the origin of royal consumption (rajayaksma), in which Growing to excess the dissipation of semen in sexual intercourse results in the "death" of King Moon.18 Elsewhere, the yogic sources abound in admonitions to celibacy Like the waning moon, (ürddhvaretas, "semen held upwards," in the lunar region of the cranial Your intellect becomes exhausted. vault), sometimes comparing the vulva to a vampiress that drains a man of And your body shrivels up his life and virility.19 A "limit situation" of this attitude was stated with Like a monkey's wrinkled skin. dramatic clarity by the very this-worldy Aghori Vimalananda just over a You grow old soon [ ... ] decade ago:

When the dark Lord of Death comes, You must have heard of women in the Place Pigalle in Paris who can Who will protect you, pick up one-franc coins from tables with their vaginas; I have seen O sinful mind? this with my own eyes. And I am told that in Laos and Vietnam, some bargirls can smoke cigarettes vaginally ... A woman who It would appear that the male sex drive puts males in an impossible bind, knows Vajroli [here, vaginal suction] can "milk" an unsuspecting man in which they find themselves in constant danger of losing their seed, a of his semen with her suction. She can make him eject over and over seed that is accumulated only very slowly and in minuscule quantities again until there is nothing left to eject, which will sap him of all his through the long process of digestion. According to Ayurvedic theory, this ojas [human vital fluid, of which there only exist eight drops]. Her is a seven-step process, in which the food we eat is serially "burned" or glands will be well benefited by this, at his expense.20 refined into rasa (chyle), blood, flesh, fat, bone, marrow, and finally, in men, In a general sense, all women are, according to such popular wisdom, semen (it becomes uterine blood or breast milk in women).13 Here, it takes so many vampiresses whose mere presence, even as an image in the male some twenty-eight days-a lunar month, and the same time it takes a imagination, is sufficient to drain a man of his seed: indeed, this is the woman's body to produce an ovum-for the food males ingest to become cause of male nocturnal emissions (svapna dosa, the "sleep flaw"). The long- transformed into semen.14 Moreover, it requires a prodigious quantity of term effect of this nefarious influence is death: Hindu India holds widows food to produce a single gram of semen.15 Ayurvedic theories of concep- implicitly guilty for the premature deaths of their husbands, whose very tion16-according to which male children are produced by a relative pre- lives they are viewed as having consumed in their insatiable feminine pas- ponderance of semen to uterine blood at the moment of conception, while sion.21 More than once I have been told by Indian men that my bachelor- females are produced by an opposite ratio17-compound the perils of se- hood (which they have equated with celibacy) was a good thing, because men depletion. thanks to it I would not die young. These medical notions have given rise, in modern India, to a male con- On a more immediate and visible level, the upshot of these concerns is cern, if not obsession, with semen retention, in which the garnering and that Indian males, finding themselves constantly beset with the cultural conservation of this, the end product of digestion and the most vital of disease of gupta rog, the "secret ailment," have been pouring, in ever- bodily fluids, is quite nearly synonymous with the maintenance of good increasing numbers, into so-called Ayurvedic gupta rog clinics. These are health and the prospects for long life. Male virility, indeed, male life expec- clinics that treat male "sexual disorders," of which the prime symptom, tancy, depends upon a careful conservation of this rare and precious vital judging from their advertising and literature on the subject, is the birth fluid which, in addition to being depleted through the inexorable aging of daughters. Putra hoga, "you will have a son," is the promised cure.22 In a

342 343 Epilogue The Siddha Legacy in Modern India

society in which male children bring wealth into a household (through the cultural disease of gupta rog. I have already suggested that this sexual dowry), while females take it out, chronic "fuel shortages" on the part of disorder was "sociocultural" inasmuch as it assumed a cause-and-effect males can take on financially catastrophic proportions. As a result, the relationship between Ayurvedic and yogic notions of physiology and gupta rog clinics do a land-office business throughout India, as evinced in metabolism on the one hand and that form of "impotence" signaled by ubiquitous billboard advertising for their remedies. the birth of daughters as opposed to sons on the other. If it is the case The economic pressures of a viciously hyperinflated dowry system, that this disease has arisen out of a fundamental misunderstanding of compounded with these pervasive notions concerning male and female medical and yogic notions of the digestive and reproductive processes, sexuality, have only increased the perceived, indeed the real, need in India this does not mean that these venerable traditions have no treatments for the remedies sold by the gupta rog clinics (the term covers all manner to offer for those suffering from such ailments. As has been noted, the of sexual disorder, including impotence, premature ejaculation, venereal bodily symptoms of the medical condition I have been discussing are disease, etc.). Within a conceptual context in which women are vam- treatable through celibacy, continence, and, should these fail, Ayurvedic piresses and the loss of seed is tantamount to financial ruin if not death, therapy and remedies. What of the more pervasive ailment of sociopolit- psychological pressures arising from the urgency of producing male off- ical impotence? Does India know of a cure for this ailment? A growing spring cannot help but have a powerful negative impact on male sexual chorus of "men in saffron," armed with paradigmatic proofs for their response. Can there be any outlet for this cultural disease? treatment, are saying yes, and a growing army of young men are follow- ing them, in a quest to recover their social, political, and religious po- tency. In order to understand the nature of their proofs, we must return 3. Omnipotence: India's "Men in Saffron" to certain of the legends of princes and Nath Siddhas recounted in the last chapter. For a large segment of the male Indian population, life is in some ways like The first of these is the legend of young prince Prthivinarayan Sāh and a bomb with a short fuse, which none save the distant Siddhas and yogins the yogurt-spouting Gorakhnath. As we have seen, that legend described have ever been capable of defusing before an otherwise inevitable ex- the empowerment, through tantric initiation, of an otherwise powerless plosion. It is in this way that Siddha longevity and male sexual disorders, figure, a petit prince without a kingdom. It is a commonplace of Nath seemingly situated at opposite poles of the Indian experience, may be Siddha legend to "Gorakh-ize" the acts of any powerful member of the viewed as intimately related in Indian thought. Indeed, it is a single mil- sect, regardless of his name. So it is that we find that the Nath Siddha in lenarian symbol system that joins them together-a system that embraces Prthivinarayan Sah's life was in fact named Bhagavantnäth, and that the Ayurvedic rejuvenation therapy (rasāyana), the "yoga of violent exertion" young prince himself was the adult leader, in the mid-eighteenth century, (hatha yoga), and elixir alchemy (debavāda). In a sense, this ancient legacy of the diminutive kingdom of Gurkha, which was vying with no fewer than has taken the form of a conundrum (unavoidable aging, loss of virility, forty-five other kingdoms for control over western Nepal. In 1763, Bhaga- death, all identified with failure to retain semen) for which a satisfactory vantnäth commended himself to the ambitious young prince, quickly be- solution is lacking. This false problem of semen depletion may be rooted came his chief strategist, and set about to vastly strengthening the prince's in a deeper "paradigm depletion." Yet, a solution of sorts has been emerg- position by using his tantric charisma to negotiate a series of crucial matri- ing in India of late, a solution that has the potential to trigger the fall of monial and military alliances between Gurkha and other of the forty-five the "secular state" in the world's largest democracy. kingdoms. In 1768, five years of strategy were crowned with success: an Here, I wish to go one step further, in the certain knowledge that some army led by Prthivinārāyan Sāh and Bhagavantnāth, and composed of their will find my speculations offensive, and hypothesize that the culturally de- western Nepali allies, conquered the kingdom of Kirtipur, in the Kath- termined form of impotence known as gupta rog is the somatization of a mandu Valley, inaugurating the reign of the Gurkha dynasty. situation of sociopolitical powerlessness or, conversely, that the acceptance A grateful Prthivinärāyan Sāh wished to reward Bhagavantnāth by mak- of certain forms of sociopolitical powerlessness is in some way related to ing him a vassal king; the Yogi refused temporal rule, but allowed that he

344 345 Epilogue The Siddha Legacy in Modern India

would be satisfied if he were given Vilās Kumārī, one of the king's daugh- Näth Siddha presence in the kingdom, constructed at an expense of over ters, in marriage. This was granted, and Bhagavantnäth received, together fifty lakhs of rupees to the royal treasury, could house as many as twenty- with his new bride, a number of major land grants which served to truly five hundred persons.25 In Tod's words once again, "During the few years solidify the place of the Näth Siddha order on the religio-political land- [that Deonäth] held the keys of his master's conscience, which were con- scape of Nepal. With this, Bhagavantnäth was elevated to the head of his veniently employed to unlock the treasury, he erected no less than eighty- order in Nepal. After his death in 1786 (eleven years after that of Prthivi- four temples ... with monasteries adjoining them, for his well-fed lazy ... nārāyan Sāh), the Näth Siddhas continued to thrive under the Gurkhas- disciples." None but Män Singh was amused, however, and after a meteoric to the extent that they were frequently called upon by their kings to loan career, Deonäth was gunned down in 1815.26 them money, which, thanks to their land revenues, they were able to do That these two cases were not aberrant is borne out by Peter van der and for which they were repaid with interest.23 Veer's provocative Gods on Earth,27 a recent study of the social, political, This is not the sole case in which the miraculous transformation of a and economic role played by the Hindu religious orders in medieval and powerless prince into a victorious king, mythically portrayed as issuing modern north India. As van der Veer demonstrates, the "Yogis," Nāgas, directly from the intervention of (and often initiation by) a Näth Siddha, and other ascetic orders, prior to the coming of the British and the railroad, turns out to be the result of efforts of a more mundane order. In the pre- used the annual pilgrimage circuit between Hardwar, Bengal, and Jagan- ceding chapter, I mentioned the case of Mãn Singh, the king of Marwar nath Puri (Orissa) as a trading network, lived in fortified temples, fought who gained his throne, in the first years of the nineteenth century, through as mercenaries in the armies of the highest (at times Muslim) bidder, bank- the miraculous intervention of another Nath Siddha. According to legend, rolled kings and generals in their political conquests, and played an invalu- young prince Män Singh, whose army was under siege in the fortified city able and extremely lucrative role in long-distance commerce.28 Indeed, the of Jalore by that of his evil cousin Bhim Singh, was met one night by a earliest religious group to take up arms after the Muslim conquest may mysterious Nath Siddha named Mastnath, who had suddenly materialized have been the Nāth Siddhas.29 out of nowhere. Mastnath led Man Singh to a "hidden well" within the More than this, as Dirk Kolff has demonstrated in his carefully docu- city walls, within which the prince found provisions enough to feed his mented Naukar, Rajput and Sepoy,30 putting on the garb of an itinerant yo- starving army. Then, cryptically, Mastnäth predicted that Män Singh's tor- gin seems to have been standard practice for all sorts of mercenaries and menter Bhīm Singh was not long for this world. Miraculously, Bhīm Singh traders in precolonial India. Kolff moreover notes that the most commonly died that very night.24 recurring time frame for such pseudoascetic peregrinations was a period Colonel James Tod, who toured Rajasthan in the first decades of the of twelve years, a figure that, as we have seen, is standard for the comple- nineteenth century and who actually met Man Singh, tells a slightly differ- tion of yogic sadhanas in the traditions under study.31 In certain cases, his- ent story, based on the testimony of members of the royal household. Mān torical figures clad themselves in the guise of the yogic god Siva. So, for Singh-who was quite demented by the time Tod met him-did indeed example the late thirteenth-century Rānā of Mewar, Rānā Samarsi, was credit the intercession of the high priest of Marwar, whose name was Ayas known as the "Regent of Mahadeva": "a simple necklace of the seeds of the Dev Nath or Deonath, for the twenty-fourth-hour death of his cousin. lotus adorned his neck; his hair was braided, and he [was] addressed as Jogin- Others saw matters differently. In the words of Tod, "a dose of poison, it is dra, or chief of ascetics.""2 Alternatively, Siva himself was portrayed in the role said, was deemed a necessary adjunct to render efficacious the prayers of of the yogin-cum-itinerant merchant. A Bhojpuri folk song says of the the pontiff." god that:33 Man Singh's gratitude knew no limits. He offered Deonath a share of the throne, which this Nath Siddha accepted. Over the years that followed, Mahadev has gone to the East to trade Deonäth himself and the Nāth sampradāya in general became fantastically And four months are passing away. wealthy in the kingdom of Marwar, with Deonath's income amounting to Gaura [sic] sits on a chair and watches the road, saying one tenth of the revenues of the state. The Mahämandir, centerpiece of the When will my ascetic [tapasi] come home?

346 347 Epilogue The Siddha Legacy in Modern India

The principal enterprise of many of these itinerant ascetic groups- built over the birthplace of the Hindu god Rama, whose temple the Mus- who were often indistinguishable from tribal confederations or trading lims had torn down to build their mosque. This was anything but a spon- cartels in the precolonial period-was the long-distance trade of horses taneous demonstration: it was a symbolic gesture and show of strength and elephants, etc. Their this-worldly ambitions of wealth and power were, planned and executed under the leadership of a group known as the Viśva however, rudely brought to a halt by the advent of the British raj and the Hindu Parisad (VHP), the Hindu World Council. This group, comprised railroad. Political centralization under foreign rulers who had no use for exclusively of members of Hindu renunciant orders, the heirs to the Siddha tantric power brokers to underwrite their authority further undermined and other sectarian movements of the Indian middle ages, are a resurgent their financial, military, and spiritual ambitions. Stripped of their tradi- nationalist Hinduism's "men in saffron." The political cousin of the VHP tional sources of income, power, and prestige, these orders began to lose is the BJP, the Bharatīya Janata Party (Indian People's Party), which has, ground well before the end of the nineteenth century. Over the past hun- over the past years, appeared to be poised to replace India's secular democ- dred years, they have become increasingly sedentary,34 such that even when racy with a Hindu theocracy. While the national BJP leaders pedal a softer the Nath Siddhas send out itinerant groups of ascetics (called jhandīs) from line, this is the outspoken agenda of the VHP, the organizational core of such centers as Gorakhpur, they have nothing to trade, save for songs militant Hindu fundamentalism. glorifying their powerful forebears. Independence has not helped matters; Given its reputation as the birthplace of Rama, the seventh incarnation and now, some fifty years into the India's democratic experiment, the Nath of the god Visnu, Ayodhya has long been a major religious center in north- sampradaya appears to be on the verge of collapse. The Mahamandir, the ern India. Its location, in eastern Uttar Pradesh, places it just about 120 centerpiece of Deonath's hegemony in Jodhpur, the capital city of old miles from Benares, the major Saiva religious center of north India, and Marwar, has been converted into a public school, the last monk of the once 80 miles from Gorakhpur, the political epicenter of the Näth Siddha or- powerful Gorakh Țilla monastery of Benares died over a decade ago, and ganization. Sitting on the nine-man VHP brain trust is Avedyanāth, who the list goes on. is, in addition to serving as the leader of the Nath sampradāya at Gorak- Or does it? Since at least the 1920s, the monastery of Gorakhpur in hpur, also the most powerful political figure in that city: he presently repre- northeastern Uttar Pradesh has been the "Rome" of Nathdom, and its ma- sents Gorakhpur district as its member of the Indian parliament in Delhi. bant, its abbot, the "pope" of the sect. In 1969, Digvijaynāth, this monas- Who is Avedyanäth if not the Gorakhnäth, the Bhagavantnath, the tery's charismatic but rather ineffectual abbot, died and was replaced by a Deonäth of late twentieth-century India? And if he is indeed nothing other dynamic figure, already active in regional politics, who took the initiatory than a modern power broker in a long line of Näth Siddha power brokers, name of Avedyanäth. When I visited the Gorakhpur monastery in the win- what compunctions need he have to respect religious or political conven- ter of 1985, the same Avedyanath was a most gracious host and helpful tion? In addition to his many sacred and secular functions and titles, Aved- informant to me, giving me the run of the monastery and impressing me yanāth is also the chairman of the Rām Janmabhūmi Mukti Samiti, the with his exceptional organizational abilities. Committee for the Liberation of the Birthplace of Rama.35 As anyone Under Avedyanath's leadership, the monastery of Gorakhpur has been familiar with Hindu sectarian theology knows, this is highly ironic. For, transformed into a force in the community and even the Gorakhpur dis- whereas Rama, the "boy scout" of the Hindu pantheon, is the god whose trict as a whole. But this is small change in comparison to what appears to adherents have historically constituted the "right wing" of Hindu religious be the national agenda of Avedyanäth, if not of a much broader confedera- belief and practice, the Nath Siddhas have, together with the tāntrikas tion of monastic leaders who have been flexing their political muscles of whose legends and initiatory ideology I have been treating in these pages, late in India. long figured among the most "left-handed" (vāmacāra) sects of all of Hin- On 6 December 1992, a Hindu mob, composed to a great extent of kār duism. sevaks, "servants of the cause"-marginalized young men in search of Avedyanäth and the VHP, even if they are championing the symbolically social, political, and religious potency-reduced the 450-year-old Babrī powerful cause of restoring a Hindu site to Hindu hands, are not, in the mosque of Ayodhya (Uttar Pradesh) to rubble, claiming that it had been final analysis, fighting for issues. Rather, they are fighting for power, for

348 349 Epilogue The Siddha Legacy in Modern India

a sociopolitical power that is the macrocosmic homologue of the power proach to life has always featured not a small amount of carefree playful- the tantric practitioner gains over his bodily microcosm, and by extension ness. It is this quality that has always impressed me the most among Nath over the entire universe, through his violent (the hatha in hatha yoga) anti- Siddhas whom I have met: they always seem to be at play-playing with nomian practice.36 The goals of the tantric practitioner were and remain words, playing with other people's minds, playing games with the world. immortality and unalloyed power in the world. Once possessed of this This is an attitude that many before me have noted. For Bernier, the Yogis power, the practitioner may use it to blast or bless, to raise untested boys were people "who scoff at everything, and whom nothing troubles,"38 while to the throne (Bhagavantnāth and Prthivinārayan Sah) or to curse entire Oman provides turn-of-the-century photographs of Yogis in carefree and cities to destruction (as Dharamnäth did to Pattan).37 The men in saffron uninhibited repose in front of camera, Yogis whose easy and well-adjusted have not forgotten these goals, and they seem to be prepared to use certain attitude he notes with a mixture of respect and suspicion.39 tried-and-true methods from that past to realize them, in the firm knowl- The ideal Nath Siddha is a god-man who plays with the entire universe, edge that such will be mythologized into idealized accounts of transforma- with the lives of the great and small alike, as he pleases. Secure in the tive initiation rites which naive western orientalists will recount as "amus- knowledge of the identity of microcosm and macrocosm, of the imma- ing anecdotes" to audiences across the world, thereby assuring them of nence of the Absolute in every creature and stone, he takes the universe to their immortality. be his plaything, with its every element (nectar and ashes, cloaks and bod- At times I think that a Hindu theocracy could not be more harmful to ies, earrings and the power of flight) interchangeable according to his India than the cynical bureaucracy that the secular democracy has spawned whim. Like the holder of any sort of power, the Nath Siddha can use his over the past half century. Indeed, such a reversal would just be another knowledge for good or for evil; most I have known spend much of their epicycle in the broad historical sweep of what Jan Heesterman has termed time "getting high" in ways I evoked in chapter four.40 Their credo is per- as India's "inner conflict of tradition," an inner conflict in which two camps haps best encapsulated in a passage from the hagiography of Mastnath: inexorably trade places in filling the roles of world maintainers and world "Whose friend is a Yogi when he plays? It takes so little to please him. He renouncers. Here let us recall that Gandhi, political mastermind that he doesn't give a thought to what's high or what's low. Whatever he wants to was, won independence for India by playing the role of world renouncer; do, he just does it ... The world's four cornerstones are [his] playground; once that independence was won, his Congress Party activists established when you're carefree you want for nothing. From a pauper to a king, from themselves in the role of world maintainers. The secular Indian state, cor- a king to a pauper, [he's] never bothered over the difference between the rupted by its own power, seems to be no longer capable of fulfilling this two."41 dual role. Has the time then come, once more, for the world-renouncing men in saffron to try their hand at world maintenance? They certainly think so, because the this-worldly ideology of over a thousand years of 4. Bhandarinath and Me tantrism tells them so. One need only think of the medieval myths of coop- eration between kings and täntrikas: power and transcendence are within Soon after I began my fieldwork in India, in the spring of 1984, I met reach. Yet, should the men in saffron gain power, they will immediately the droll and bearded Bhambhulnath, the kindly abbot of the Nath Siddha find themselves faced with the same conundrum as has faced every world monastery at Hardwar, to whom I expressed my interest in meeting living renouncer who went the world-maintenance route. As soon as the world alchemists. While he allowed that he didn't know any such persons, he sug- renouncer becomes involved in the world, playing the role of royal chap- gested that I should go to Nepal, where the great Nath Siddha scholar lain, minister, or even surrogate king, he forfeits all claim to transcendence, Narharinath, abbot of the Caughera monastery, would be able to give me to the very ground upon which he founded his prior authority. He becomes some pointers. It was October before I finally managed to have an audience just another player in the power game, as Prthivinārāyan Sāh's Bhagavant- with Narharinath, a most impressive figure, who told me he had heard näth perhaps foresaw, but Mān Singh's Deonāth never comprehended. rumors of alchemists living in the region of Mount Abu, in western Rajas- Between the two extremes I have outlined lie those Siddhas whose ap- than. I was not to reach Abu until my next research tour of India, in the

350 351 Epilogue The Siddha Legacy in Modern India

winter of 1993. There, Samundranath, the custodian of the Gorakhnath dundā came rushing back to me. I was in Nepal, talking to a Sampela who shrine at Oriya village, told me there were no such people at Abu. How- had come there from Hardwar. Had I hit the mother lode? Bhandarinath ever, Inder Dan Detha, a Rajasthani bard (Caran) and scholar-and part- was typical in many ways. He was on the road most of the year, taking a time resident of Acalgarh at Abu-whom I met in Jodhpur the next month, side trip from the Himalayan high roads to the four dhams (Kedarnath, in March, told me of an encounter he had had with a certain Motinath, Badarīnāth, Gangotrī, and Yamunotrī) to visit Kathmandu this time. And whom he had met at the Rathadunda monastery, in the Meerta City district by the way, could I give him some money? of eastern Rajasthan, some twenty-five years earlier. Motinäth, he said, had At this point, we entered into concerted negotiations. His main concern had a reputation as an alchemist and was someone who could tell me about was money,45 and mine was information. So back and forth we went. "How the use of mercury and herbs as means to immortality. much money you gonna give me? A hundred and one rupees would be So it was that on a hot day in April, I found myself at Rāthadundā talk- nice," to which I replied, "Ever meet any alchemists?" I asked him if he ing to the two Naths-one very senile and the other very stoned-who knew any Nath Siddhas who ate herbs and mercury, and he allowed that appeared to constitute the entire population of that monastery. When I they all used herbs to improve their quality of life. Then, quickly changing asked about Motinäth, the younger of the two laughed and said, "He's out- the subject, he told me to pick up a pinch of dirt from the side of the road side, underneath the new samadhi.42 He gave up his body twenty years ago." on which we were standing. I did so, and noted that here, as in many parts So much for my immortal alchemist, I thought. I was, however, intrigued of the valley, the soil was very sandy, with a high white mica content. "The by the Nath Siddha's aside that at the time he abandoned his body, at an seed of Prajapati46 and the sexual fluid of the Goddess,"47 I thought. "I've age of about seventy-five, he still had the appearance of a twenty-five-year- got a handful of jizz!" old. I then asked whether there were any alchemists living in Rajasthan "Now," said Bhandarinath with a theatrical flourish, "put the dirt in my that he knew of, to which he responded with a loud guffaw. "Alchemists in hand and say 'Om jai Pasupatinath!'"48 He then returned the dirt to my Rajasthan? Rajasthan's a desert! There's no alchemists here, because there's hand and instructed me to repeat the mantra. This I did, at which point he no herbs here. You've gotta go to the mountains, up to Nepal, up to Hard- told me to open my hand. There, together with the sand and the mica, war, to find alchemists!" Hardwar was where I had begun my search, some were grains of unhusked rice! "Shades of Cauranginath,"49 I thought, nine years earlier. It seemed I had come full circle in a futile quest for "more sexual fluids!" I fell all over Bhandarinäth, telling him how impres- something that no longer existed .. sed I was at his piece of magic. Encouraged, he reached into the folds of One cool premonsoon morning in the Thamel neighborhood of Kath- his still-capacious saffron tunic and pulled out a jholi50 from which he pro- mandu, a month after I had left Rāthadunda, I nearly ran over a Nth Sid- duced a plastic bottle, out of which he spilled two pea-sized pellets into dha on my bicycle. There could be no doubt that he was a Näth: the crystal my hand. They appeared to be of mineral composition and were covered earrings in the thick of his ears were a dead giveaway. "Ades!" I cried out to with vermilion. him, as I tumbled down off my bike, "ades!"4 Without missing a beat, the These, Bandarinath assured me, would bring me good health, good life, Jogi's answer came: "Will you have a vision of the Lord of Serpents?" "A and success in all that I did, warning me that only I could benefit from Sampela," I thought, and more or less knowing what was to follow, said their powers: they were nontransferable. Then he asked me to feed him, "Sure!" at which point he pulled a wicker basket out from beneath the and I gave him 15 rupees. He said he'd done a lot of talking for 15 rupees, folds of his capacious saffron tunic, opened the lid-and a cobra that was and that 101 was more like what he had in mind. I promised him more the supposed to rear up and scare the dickens out of me didn't. Prodding and next time I met him. He told me where he was staying, and I promised to poking, the Sampela tried to coax his serpent into showing more signs of come and see him in the coming days. As it turned out, that was the last I life but, poor guy, it was a cool morning and the snake was still snoozing. saw of Bhandarinath ... Seizing the moment, I tried to engage the Nath in some small talk. I raced home on my bicycle, tapping my shirt pocket all the way to be "What's your name? Where're you from?" "My name's Bhandarinath and sure that my alchemical (?) booty was still there. As soon as I arrived, I I'm from Hardwar," he replied. The words of the laughing Näth of Ratha- took out a pocketknife and-not without some trepidation-cut into one

352 Epilogue

of the two gutikas31 Bandarināth had given me. The knife went through the NOTES pellet, but not without difficulty, since it was indeed of mineral composi- tion. It was also, clearly, a man-made pill, the product of some chemical operation. My eyes were filled with images of the alchemical Grail. I had passed the test! The Nath Siddhas on high, approving of my sincerity and diligence, had sent their man to me to deliver the mystic goods! One of the first things I did upon my return to the west was to take my gutikas to a laboratory for analysis." A week later, the results of the electron-microscopic tests came back. My sample was composed of an amalgam of mineral and vegetable matter. The minerals in question were sulfur (!), silver (!), silicon, phosphorus, potassium, calcium, iron, and chlo- Preface rine-but no mercury! Was the silver some base metal that had been "di- 1. Thomas Pynchon, Gravity's Rainbow (London: Picador, 1975), p. 590. Ac- gested" and transmuted by mercury, which had left no trace of itself be- cording to our best evidence, the vast majority of practitioners of tantric alchemy hind? Without a guru, how could I ever know whether this was in fact and batha yoga have always been males. So it is that I employ the masculine pronoun the philosopher's stone I had cut in half and had fried under an electron be, rather than she or s/be when referring to such practitioners. microscope or just a medical tablet for chilblains? The second gutikā was 2. Cf. the akam genre of classical Tamil love poetry, whose "five landscapes" are still intact. discussed in A. K. Ramanujan, The Inner Landscape: Love Poems from a Classical Tamil Our research continues. Antbology (Bloomington: Indiana University Press, 1975), pp. 105-8. 3. Sudhir Kakar, The Inner World: A Psychoanalytic Srudy of Childhood and Society in India (New York: Oxford University Press, 1978); Ashis Nandy, The Intimate Enemy: Loss and Recovery of Self under Colonialism (Delhi: Oxford University Press, 1983). Given that many of the interpretive connections made between les mots et les choses in this book are my own, some may be moved to turn their analytical lights on my own psychological profile as well. 4. La poétique de l'espace (Paris: Presses Universitaires, 1974). A recent western study of an eastern tradition, which I feel to be most respectful of the phenomeno- logical approach, a book which moreover devotes many of its pages to charting mystic landscapes, is Norman Girardot's admirable Myth and Meaning in Early Tao- ism (Berkeley: University of California Press, 1983). 5. Even if, as Gadamer and others have demonstrated, such is patently impos- sible: one always comes to a text with a forejudgment of its meaning. This includes, for the alchemical portion of the books, the models of the modern science of chem- istry. Many are the historians who have treated tantric alchemy as a protochemistry or iatrochemistry and who have, working from this methodological assumption, projected the writings of the alchemical tradition upon a modern chemical grid. Such attempts are, I believe, tentative at best, given our uncertainty of the modern equivalents of medieval (and often local) terminology. This is especially the case with botanical names, in which the sources abound. Similarly, modern interpreta- tions of the subtle body of batha yoga which would see in the medieval texts proofs

353

354 355 Notes to Chapter One Notes to Chapter One

for the advanced state of Indian knowledge of physiology and psychology force the (= Vidyā[purușas?]), a group of more or less suicidal mercury-drinking Theravāda textual data into impossible and highly prejudicial directions. Buddhist monks: Patrick Pranke, "On Becoming a Buddhist Wizard," in Don Lo- 6. On the theological uses of anagogy, as a complication of analogy and allegory, pez, ed., Buddbism in Practice (Princeton: Princeton University Press, 1995), pp. in the medieval west, see Marie-Dominique Chénu, Nature, Man and Society in the 343-58; Maung Htin Aung, Folk Elements in Burmese Buddbism (London: Oxford Twelfth Century, tr. and ed. Jerome Taylor and Lester K. Little (Chicago: University University Press, 1962; reprint Westport, Conn .: Greenwood Press, 1978), pp. of Chicago Press, 1969; reprint Chicago: Midway Books, 1979), pp. 123-24. 41-50; Melford Spiro, Buddhism and Society: A Great Tradition and Its Burmese Vicissi- 7. Symbolic language is employed with such great success in these traditions tudes (New York: Harper & Row, 1970), pp. 163-71; and Joseph Needham et al., because the name of a thing bears the efficacity of a given object: Jean Filliozat, Science and Civilisation in Ancient China, 6 vols. in 17 tomes (Cambridge: Cambridge "Taoïsme et yoga," Journal Asiatique 257 (1969): 63. University Press, 1954-88), vol. 5, pt. 3 (1976), "Chemistry and Chemical Tech- 8. Douglas Renfrew Brooks, The Secret of the Three Cities: An Introduction to nology: Spagyrical Discovery and Invention: Historical Survey, from Cinnabar Hindu Sākta Tantrism (Chicago: University of Chicago Press, 1990), p. 209. Elixirs to Synthetic Insulin," p. 166. 9. Betty Jo Teeter Dobbs, The Foundations of Newton's Alchemy or "The Hunting of 7. They are also called Nath Yogis, the Nath Sampradāya, Kanpbata ("Split the Greene Lyon" (London: Cambridge University Press, 1975), p. xi. Eared") Yogis, Jogis, and Gorakhnatbis in modern northern India. IO. RRĀ 3.11b-12a: na krameņa vinā šāstram na śāstreņa vinā kramaḥ/šāstram 8. Most fully initiated members of the Nath Siddha order are given names, upon kramayutam jñātvā karoti sa siddhibhāk//. Cf. BbP 3.96; 4.139; 9.140b-141a; and initiation, that end in the suffix -nath. Throughout this book, however, I often RRS 6.2. shorten the names of these figures, and call Gorakhnath "Gorakh," Matsyendranāth 11. Mircea Eliade, Forgerons et alchimistes, 2d rev. ed. (Paris: Flammarion, 1977); "Matsyendra," etc. Furthermore, in those cases in which I retain the suffix, I gener- idem, Yoga: Immortality and Freedom, 2d ed., tr. Willard R. Trask (Princeton, N.J .: ally transliterate it as -nath, following modern usage, rather than -natha, the San- Bollingen, 1973). skritic ending. Only in those cases in which I am referring to a given Nath Siddha 12. "Why Gurus Are Heavy," Numen 33 (1984):40-73- as the author of a Sanskrit-language work do I retain the -natha suffix (as in "the Rasaratnākara of Nityanātha"). Chapter One 9. On the overlap between the terms tantra and sakta (applied to those sects I. Reproduced with translation and notes in B. N. Goswamy and J. S. Grewal, whose immediate object of devotion was the Goddess, Sakti), see HT, pp. 5-7. Most The Mughals and the Jogis of Jakbbar: Some Madad-i-Ma'ash and Other Documents Hindu tantric sects were in fact Säkta-Saiva, worshipping the Goddess as a more (Simla: Indian Institute of Advanced Study, 1967), pp. 120-24. Jakhbar monastery accessible form of Siva than the god himself. is located in the Gurdaspur district of the Punjab. 10. For the hundreds of references to the semidivine Siddhas and Vidyadharas 2. The root tan is a cognate of the English tension and tensile, in which the senses in Somadeva's monumental 11th c. Kathāsaritsāgara, see the index to Ocean, vol. 10, of stretching and weaving are also present. On tantra as ritual framework, see Brian pp. 302, 351 [s.v. "Siddhas, independent superhumans" and "Vidyādharas, inde- K. Smith, Reflections on Resemblance, Ritual, and Religion (Oxford University Press, pendent superhumans"]. 1989), p. 126, citing Asvalayana Śrauta Sūtra 1.1.3. 11. So, for example, according to a Bengali recension of the Râmāyana, the para- 3. In order to avoid confusion between Tantra as written work and tantra as dise land of Uttarakuru was reached by crossing a river called Sailoda ("Rock Wa- religious phenomenon, I follow general modern usage and refer to the latter as ter"), whose touch turns men into stone. On either bank of that river grow reeds "tantrism." I further adjectivize "tantra" into "tantric" to speak, for example, of called kichaka, which carry Siddhas to the opposite bank and back. This is a country tantric ritual. I often refer to tantric practitioners as tantrikas, following Sanskrit where the Siddhas live together with divine nymphs in forests whose trees and usage. flowers, composed of precious stones, exude a miraculous resin that is nothing 4. This and all the other terms discussed here are generated from the verbal other than the nectar of immortality itself. On this, see Encyclopedia of Religion and root sadh/sadh (weak form sidh), which means "to realize, succeed." Etbics, 12 vols., ed. James Hastings (New York: Scribner's, 1908), s.v. "Abode of the 5. On the uses of this appellation, see below chap. 4, nn. 112-14. Blest (Hindu)," by H. Jacobi. On "rock-water," see below, chap. 4, n. 189. 6. To these groups we might also add the Burmese Zawgyis (= Yogis) or Weikzas 12. See below, chap. 10, nn. 82-85.

356 357

Notes to Chapter One Notes to Chapter One

13. Paraśurāmakalpasūtra 4.10 (ed. A. M. Sastri and S. Y. Dave, Gaekwad's Ori- p. 199. The perspective of all Hindu tantric texts of which I am aware is that of ental Series, no. 22 [Baroda: Oriental Institute, 1950], p. 149). Significant here is the male practitioner. Miranda Shaw (Passionate Enlightenment: Women in Tantric the redundancy in the term employed for the second category of Siddhas, mediat- Buddbism [Princeton: Princeton University Press, 1994]) has recently advanced the ing between the divine (divya) and the human (manava): these are the perfected theory that it is the female practitioner's perspective that originally predominated (siddha) Siddhas. The Lalitā Sabasranāma (cited in Bagchi, introduction to Kaula- in Buddhist tantric practice. While I find her arguments improbable, I leave it to jñānanirnaya, p. 20) delineates the same divisions. The earliest enumeration of the my colleagues in Buddhist studies to confirm or disprove her hypothesis. three oghas may be that of the Kularnava Tantra (6.63-68), which divides the found- 18. For further discussion, see below, chap. 5, sec. 2. ing Kaula gurus along these lines (Kulārnava Tantra, ed. Taranatha Vidyaratna with 19. Whence the maxim "privately a sakta, outwardly a saiva, and a vaisnava in an introduction by Arthur Avalon [John Woodroffe][Madras: Ganesh and Com- court; bearing various outward appearances, the followers of the kula-system pany, 1965; reprint Delhi: Motilal Banarsidass, 1975]): see below, chap. 4, n. 62. spread over the earth": Yoni Tantra 4.20, cited in The Yonitantra, critically edited The Tantraraja Tantra (2.2-5) divides a group of nine Nathas (identified with bodily with an introduction by J. A. Schoterman (Delhi: Manohar, 1980), p. 16. orifices) into three groups of the same three "streams": in John Woodroffe, Tan- 20. This relationship was first charted in the Samkbya Kārikas (39, 42) and com- trarāja Tantra: A Short Analysis, 3d ed. (Madras: Ganesh and Co, 1971), p. 19. Janar- mentaries. For a discussion, see Lakshmi Kapani, La notion de samskāra, 2 vols. dana Pandeya, in his introduction (p. ja) to the second volume of his edition of the (Paris: De Boccard, 1991, 1993), vol. 2, pp. 422-28. Goraksa Sambitā, 2 vols. (Sarasvatibhavana Granthamala, vols. 11O, 111 [Benares: 21. For further discussion of the origins and history of hatha yoga, see below, Sampurnananda Sanskrit Visvavidyalaya, 1976, 1977]), divides certain of the Rasa chap. 5, nn. 83-92, and chap. 8, sec. za. Siddhas into these three categories. Four of the eight directions in the Kaula "Cir- 22. RA 17.165a. cle of Siddhas" (Siddha Cakra) are assigned to the Siddhas (south) and the Sid- 23. Whereas Siva is the name that takes precedence over all other names (Mahā- dhaugha (west), Mänavaugha (northwest), and Divyaugha (southwest): Mark S. G. deva, Bhairava, etc.) for this male divinity, there is no such order of precedence for Dyczkowski, The Canon of the Saivagama and the Kubjika Tantras of the Western Kaula the Goddess's many names (Durgā, Pārvatī, Sakti, Caņdī, Kālī, Devī, etc.). I there- Tradition (Albany, N.Y .: SUNY Press, 1988), p. 81. See also ibid., p. 90, on the fore generally refer to her as the Goddess (which is also a literal translation of the three ogbas in the Ciñcinimatasarasamucchaya, a text of the Western Transmission. name Devi). 14. Atharva Veda 1.4.4; 3.31.10; 10.2.17; and 19.53.1. Rasa and the waters are 24. Bhudeb Mookerji, Rasa-jala-nidhi or Ocean of Indian Chemistry and Alchemy, further identified with notion of "healing remedy" (bhesaja), which becomes the 5 vols. (Calcutta: K. C. Neogi, 1926-38), vol. I, p. xiv. foundation of traditional Indian medicine (ayurveda) and elixir therapy (rasāyana). 25. RA 1.19. For variant readings, from other alchemical sources, see below, See below, chap. 7, sec. 1. chap. 5, n. 161, and chap. 9, nn. 54-55. 15. The foundational works of Indian medicine (Ayurveda), the Caraka and Su- 26. Satapatha Brahmana 7.4.1.15. The all-absorbing capacities of mercury are śruta Sambitās, date from this period: see below, chap. 2, n. 15. See bibliography those of any element at the summit (although it is more accurate to say the center)

for editions. of a Hindu hierarchy (on this see David N. Gellner, Monk, Householder, and Tantric 16. The earliest excavated Siva linga is the Gudimallam image from Andhra Priest: Newar Buddbism and its Hierarchy of Ritual [New Delhi: Cambridge Univer- Pradesh. On its dating, see Gritli v. Mitterwallner, "Evolution of the Linga" (who sity Press, 1992], pp. 45-48). Concomitant to this ability is also the inherent power dates the image to the first century B.c.) and Doris Meth Srinivasan, "Significance of expansivity, the power to become all-encompassing. This is precisely the nature and Scope of Pre-Kusana Saivite Iconography" (who dates it to the fourth to third of brábman, the Absolute of Vedanta metaphysics: the verbal root of brabman is brb, centuries B.c.), both in Michael Meister, ed., Discourses on Siva: Proceedings of a Sym- "to magically expand." This is what mercury does when it transforms millions of posium on the Nature of Religious Imagery (Philadelphia: University of Pennsylvania times its own mass of base metals into gold. Press, 1984), pp. 18-19, 34, and plate 18. 27. The two important modern exceptions to this rule are the Śrī Vidya tradi- 17. Alexis Sanderson, "Purity and power among the Brahmins of Kashmir," in tion of Tamil Nadu and the tantric form of Hinduism that continues to predomi- The Category of the Person: Anthropology, Philosophy, History, ed. Michael Carrithers, nate in Nepal. Steven Collins, and Steven Lukes (Cambridge: Cambridge University Press, 1985), 28. See below, chap. Io, sec. I, and chap. I1, sec. 3.

358 359 Notes to Chapter One Notes to Chapter One

29. This scorn and mistrust carries down to nineteenth-century Britain, in tity of (King) Soma and moon is made: Satapatha Brabmana 10.4.2.1; 11.2.5.3. See which, according to the ethnographer H. A. Rose (A Glossary of the Tribes and Castes below, chap. 2, sec. 2 and 4, and chap. 7, n. 2. of the Punjab and Northwest Frontier Province, 3 vols. [Lahore: Superintendent, Gov- 42. Satapatha Brābmana 6.1.2.4. ernment Printing, Punjab, 1911-19], vol. 2, p. 407), certain elements of the Rawal 43. Satapatba Brābmaņa 6.2.2.6. branch of the Näth Siddhas had made their way to Europe, where they were notori- 44. RV 1.23.19a; 10.137.6; Atharva Veda 1.4.4. See also Kenneth Zysk, Religious ous purveyors of quack medicines and cures! On the modern-day Indian confusion Healing in the Veda, Transactions of the American Philosophical Society, vol. 75, between the magicians and tantrikas, see Lee Siegel, Net of Magic: Wonders and De- part 7 (Philadelphia: The Society, 1985), pp. 90, 92. ceptions in India (Chicago: University of Chicago Press, 1991). 45. See below, chap. 2, n. I. 30. Joseph Needham et al. Science and Civilisation in China vol. 5, pt. 4 (1980): 46. On this verse, see Louis Renou, "Études védiques," Journal Asiatique 243 "Chemistry and Chemical Technology: Spagyrical Discovery and Invention-Ap- (1955), p. 437, n. 2. Cf. Jaiminīya Brāhmana 3.335; and Kāthaka Sambitā 8.5 and paratus, Theories and Gifts," pp. 497-98. 12.13, discussed in Stephanie W. Jamison, The Ravenous Hyenas and the Wounded 31. Francois Bernier's Voyages were originally published in Amsterdam, in two Sun: Myth and Ritual in Ancient India (Ithaca, N.Y .: Cornell University Press, 1991), volumes, in 1699, by David-Paul Marret. The present passage is translated from a pp. 191-93. These homologizations are assumed by the Ramayana, in its account modern edition of the work, Voyage dans les Etats du Grand Mogol, with an introduc- of the birth of Skanda: see below chap. 7, sec. 2. tion by France Bhattacharya (Paris: Fayard, 1981), p. 245. 47. Yajur Veda 7.4.1.10. 32. RV 7.59.12. Cf. RV 3.3.7; 9.100.1; 2.38.5; 7.90.6; Atbarva Veda 18.3.62; Śata- 48. Präna, breath or "life force" is further identified with the soul (ätman), a patha Bräbmana 2.6.1.12; and Wendy Doniger O'Flaherty, Tales of Sex and Violence: term etymologically linked to the German atmen sich and the Latin animus. It may Folklore, Sacrifice, and Danger in the Jaiminiya Brabmana (Chicago: University of further be equated with "spirit," which is present in "re-spir-ation." Breath is further Chicago Press, 1985), p. 20. identified with the universal self or soul, brabman, in Kausītaki Upanisad 2.1. 33. RV 4.26-27. For Indo-European and ancient Near Eastern parallels, see 49. Zysk, Religious Healing, pp. 8, 241. Cf. Louis Renou, Etudes védiques et pani- David M. Knipe, "The Heroic Theft: Myths from Rgveda IV and the Ancient Near néennes, 17 vols. (Paris: de Boccard, 1955-69), vol. 2, p. 66, note (quoted in André East," History of Religions 6 (1967): 328-60. Padoux, Vac: The Concept of the Word in Selected Hindu Tantras [Albany, N.Y .: SUNY 34. Charles Malamoud, "Le corps contractuel des dieux: remarques sur le rite Press, 1989], p. 6) "In fact, the creation of the cosmos is similar to that of a work védique du tanūnaptra," in Cuire le monde, p. 238. of art, either being the kavi's deed." An equally important link between the Atharva 35. Charles Malamoud, "Les dieux n'ont pas d'ombre: remarques sur la langue Veda and the medical and alchemical traditions lies in one of the terms employed secrète des dieux dans l'Inde ancien," in Cuire le monde, p. 250. On the brahmanic by the former to designate the physician. The image of the kavi, the inspired poet definition of sacrifice, see idem., "Cuire le monde," in ibid., p. 47, citing of the Veda, stands behind both the Ayurvedic physician (called kavi-rāja in Bengal) Āpastambba Yajña Paribhāsa Sütra 1 and Kātyāyana Śrauta Sūtra 1.2.1. and the alchemical wizard (kavi) who ultimately creates his own universe. See be- 36. Satapatba Bräbmana 10.1.5.4, with the commentary of Sayana who states low, chap. 9, nn. 105-6 and chap. 1o, nn. 78-80. that one hundred years is how long amrta keeps the gods immortal! Cf. RV 50. Surendranath Dasgupta, A History of Indian Philosophy, 5 vols. (Cambridge: 10.161.3b; Atbarva Veda 2.29.2. Cambridge University Press, 1922; reprint Delhi: Motilal Banarsidass, 1975), vol. 37. Charles Malamoud, "La théologie de la dette dans le brahmanisme," in Cuire 2, p. 284 and n. 1. According to the Taittirīya Sambita (6.4.9), the itinerant physi- le monde, pp. 125-27. cian was impure; Brahmans therefore were not to take up medicine as a profession. 38. Compact Edition of the Oxford English Dictionary, s.v. "mortgage." According to the RV (1.180.2), the divine physicians, the Asvins, are drinkers of 39. RV 8.48.3a, 4b, 12a, translated in Wendy Doniger O'Flaherty, The Rig Veda "mead" (madbu) rather than soma: Jean Filliozat, La doctrine classique de la médecine (Harmondsworth: Penguin, 1981), pp. 134-35. indienne: ses origines et ses parallèles grecs, 2d ed. (Paris: Ecole Francaise d'Extrême 40. Śatapatba Brābmana 9.5.1.7-8,I1. Orient, 1975), pp. 15-16. 41. RV 9.86.39; 10.107.2. Later, the mediating "drop" is dropped, and the iden- 51. Caraka Sambitā 1.30.20. Cf. Dasgupta, History, vol. 2, pp. 273-84-

360 361 Notes to Chapter Two Notes to Chapter Two

Chapter Two 14. On Buddhist terminology in Caraka, see Priyavrat Sharma, Ayurved ka vai- 1. Smith, Reflections, pp. 46, 203, 224. jñānik itibās, 2d ed., Jaikrishnadas Ayurveda Series, no. I (Benares: Chowkhambha 2. The term mesocosm was coined by Paul Mus (who further uses the terms "pro- Orientalia, 1982) p. 102. It is widely held that the replacement of surgery (salākya) tocosm" for "microcosm" and "metacosm" for "macrocosm" in his study of Indian by pharmacy (rasa śästra) in Indian medicine was also due to Buddhist influence: Buddhism) in his lengthy foreword to Barabudur (Hanoi: Imprimerie d'Extrême see below, chap. 3, n. 21. Orient, 1935), p. 100. Mus's terminology has been further expanded by John S. 15. Dasgupta, History 2:393-99. The dating of the foundational Ayurvedic texts Strong (The Legend of King Asoka: A Study and Translation of the "Asokavadāna" continues to be a matter of some discussion. The Chronology Committee of the [Princeton: Princeton University Press, 1983], pp. 104, 118-20, 131-33). On this National Institute of Sciences of India dates the composition of the Caraka Sambitā see my "Dakkhina and Agnicayana: An Extended Application of Paul Mus's Typol- to ca. A.D. 100, and that of the Susruta Sambita to the 3d-4th centuries A.D .: D. M. ogy," History of Religions 26:2 (November 1986), pp. 191-95. Bose, ed., A Concise History of Science in India (New Delhi: Indian National Science 3. On this fascination with number, see Charles Malamoud, "Exegèse de rites, Academy, 1971), p. 223. exégèse de textes," in Cuire le monde, p. 282 and throughout. 16. Caraka Sambitā 4.1.1-156. For a discussion, see Dasgupta, History 1:212- 4. By "bipolar" I mean two polar opposites which, in spite of, or indeed because 19. of, their opposition, interact dynamically. Male and female are bipolar, as are moon 17. Jean Filliozat, "Les mechanismes psychiques d'après les textes de yoga," and sun, semen and blood, etc. See HTSL, pp. I-2. Annuaire du Collège de France (Paris: 1970-71), p. 416. 5. Satapatba Brābmaņa 7.2.1.4: Taittirīya Brabmana 1.5.9.4. 18. Ibid., p. 416. 6. Mus, Barabudur, pp. 53, 59, 203-4. 19. Caraka Sambitā 4.5.5. 7. Brbadāranyaka Upanisad 1.4.1-6. 20. Caraka Sambitā 4.4.13. 8. Chandogya Upanisad 6.8.7; Hathayogapradīpikā 4.50, 56 (Hathayogapradīpikā of 21. A very early use of the term dhätu is Atharva Veda 11.3.78. Cf. the ca. 11th c. Svätmaräman, with the commentary of Brahmānanda, ed. and tr. Srinivasa lyengar A.D. Śārada Tilaka (6.7) [Sardatilakam of Laksmanadesikendra with the Padārthadarsa [Madras: Adyar, 1972]). For an early mythological account of the "bottling up" of Commentary by Raghavabbatta. edited by Mukunda Jha Bakshi, 3d ed. [Varanasi: the absolute brabman within the human body, wherein it becomes identified as the Chowkhamba Sanskrit Series Office, 1986]), a tantric ritual guide, which identifies atman, see Aitareya Aranyaka 2.4.1-3. the seven dhätus with the seven mother goddesses (mātrkās). On digestion in gen- 9. On the kalas, see below, sec. 4 of the present chap. eral, see Caraka Sambita 6.15.3-38, especially vv. 32-35. 10. Sanjukta Gupta, "The Mandala as an Image of Man," in Richard Gombrich 22. C. Dwarkanath (Introduction to Kayachikitsa [Bombay: Popular Book Depot, ed., Indian Ritual and its Exegesis. Delhi: Oxford University Press, 1988, p. 35. On 1959], pp. 318, 320) defines rasa as plasma, including interstitial fluid and lymph; the five elements, see below, sec. 3 of the present chap. and chap. 7, sec. 5. and the dhatus as "intermediary metabolites," intended for the maintenance of 11. Cited in Jean Filliozat, "La discipline psychosomatique du yoga et ses fonda- stable formed tissue. See below, chap. 7, sec. I, for an extended discussion of the ments theoriques," Annuaire du College de France (1965-66), p. 384 term rasa. 12. RV 10.16.1. See also Gopinath Kaviraj, Bharatiy Samskrti aur Sādbanā 23. In digestion the lower dhätus nourish the higher ones either directly or in- (Patna: Bihar Rashtrabhasha Parishad, 1977-79), vol. 2, p. 281; and Malamoud, directly with rasa irrigating each of the higher dhatus consecutively. This latter (and "Cuire le monde," in Cuire le monde, p. 52 n. 52, citing the Yajñavalkya Smrti (3.84). later) conceptualization holds that each of the last six dbatus are nourished "di- 13. Robert C. Lester, Theravada Buddhism in Soutbeast Asia (Ann Arbor: Univer- rectly" by the food one eats through its own particular channel, via the rasa dhātu sity of Michigan Press, 1973), pp. 23-24, 53; Caroline A. F. Rhys-Davids, "Original (Caraka Sambitā 4.28.4, with the commentaries of Cakrapanidatta and Śivadāsa Buddhism and Amrta," Mélanges chinois et bouddhiques [Brussels] 6 (1938-39), p. 378. Sena, discussed in Dwarkanath, Kāyachikitsā, pp. 319-26). Cf. Caraka Sambitā On the medical language of the later Saiva Pasupata Sūtras (ca. Ist c. A.D.), see 6.15.15-16; Śārngadbara Sambitā 1.5.11-12; and Yājnavalkya Smrti 3.84. Nagendranath Bhattacharyya, History of the Tantric Religion (A Historical, Ritualistic 24. Caraka Sambitā 6.15.17; MBbT 2.5-6. and Philosophical Study). Delhi: Munshiram Manoharlal, 1982, pp. 16, 201-2. 25. Jan Gonda (Triads in the Veda, p. 210, cited in Georges Dumezil, Les dieux

362 363

Notes to Chapter Two Notes to Chapter Two

souverains des Indo-Européens, 3d ed. [Paris: Gallimard, 1986], p. 253) calls dosas 21) and January 14 is due to the 23.27° angle between the equator and the ecliptic "morbific entities." A more general translation for this term, which would take into in central India. account its uses in other spheres of Indian discourse, is "break in continuity." 37. Hārīta Sambitā 4.20-30 (La Hārītasamhitā: texte médical sanskrit, ed. and tr. 26. Edeltraud Harzer, s.v. "Sāmkhya" in Encyclopedia of Religions, ed. Mircea Eli- Alix Raison [Pondicherry: Institut Français d'Indologie, 1974]). Cf. Zimmerman, ade (Chicago: University of Chicago Press, 1986). "Rtu-Sātmya," p. 94. 27. Caraka Sambitā 1.12.13; Suśrūta Sambitā 4.33.3. The earliest mention of the 38. Caraka Sambitā 6.8.2-I1; Sušrūta Sambitā 6.41.4-5. tridosa is useful to the dating the emergence of Ayurveda per se in India. Whereas 39. Raghuvamsa 19.37-47, cited in Zimmermann, "Rtu-Sātmya," p. 96. wind and bile are already mentioned in the Vedas, the term for phlegm (ślesma) 40. Taittirīya Sambitā 2.3.5.1-3; Maitrāyanī Sambitā 21.7; Kāthaka Sambitā 11.3. does not appear until the ca. 1000 B.C. Satapatha Brahmana. Following this, it is not The myth has its origins in RV 10.85.2. until the ca. 313 B.C. vārttika of Kātyāyana on Pāņini (Astādhyāyī 5.1.38) that the 41. Rohinī may alternatively be identified with the sun: Filliozat, Doctrine three are mentioned together in a compound (vātapittaslesman). One may thus sur- classique, p. 85. mise that the definitive formation of Ayurvedic medical theory occurred between 42. Astānga Samgraba 6.49.245 (Śrīmadvāgbhattaviracita Astāngasamgraba, ed. 1000 and 313 B.c .: Filliozat, Doctrine classique, pp. 154-59. with a Hindi translation by Kaviraj Atrideva Gupta, 2 vols. [Bombay: Nirnay Sagar 28. The dhätus are the material causes for health, while the dosas are the dynamic Press, n.d.]). This verse is directly borrowed into the RA 18.14; similar formulas causes: Dasgupta, History, 2:334. are found in Caraka Sambita 6.7.69; Brbat Sambita 75.3; and RHT 19.19. For a 29. Louis Renou and Jean Filliozat, L'Inde classique: manuel des études indiennes discussion, see Siddhinandan Misra, Ayurvedīya Rasašāstra (Benares: Chowkham- (Paris: Imprimerie Nationale, 1947-53), vol. 2, pp. 152-53. On the symbolic role bha, 1981), pp. 19-20. The classic account of the Ayurvedic healing of rājayaksma is of the five breaths, from the Chandogya Upanisad (3.13; 5.19) to tantric Buddhism, found in Caraka Sambita 6.8.1-191. Cf. Palmyr Cordier's translation, with synoptic see Alex Wayman, Yoga of the Guhyasamajatantra: The Arcane Lore of Forty Verses passages from the works of Bhela, Vrnda, and Cakrapāni, in "Histoire de la méde- (New York: Samuel Weiser, 1977), pp. 70-71. cine indienne. La phtisie pulmonaire," Annales d'hygiène et de médecine coloniales 15 30. The higher dhatus are heavier than the lower: Caraka Sambita 1.27.337. (1912), pp. 255-66 and 535-48. 31. Francis Zimmerman, "Rtu-Sātmya: Le cycle des saisons et le principe d'ap- 43. Suśruta Sambitā 1.14.14-15; but see 6.15.32 for a figure of six to seven days. propriation," Purusārtha 2 (1975), pp. 87-105. 44. According to popular traditions in Rajasthan and elsewhere in India, it takes 32. Gopatha Bräbmana cited in Sharma, Ayurved, p. 501. forty days and forty drops of blood to produce one drop of semen: G. Morris Car- 33. Zimmerman, "Rtu-Sātmya," pp. 87-92. Also in the Atharva Veda (8.1.11-12), stairs, The Twice-Born: Study of a Community of High-Caste Hindus (London: Ho- Time is identified with the sacrificial fire that can cook and mature (when it is the garth Press, 1961), pp. 83-84. Cf. SSP 1.73 (in Kalyani Mallik, ed., Siddba Sidd- Jatavedas or Vaisvanara) or consume and devour (when it is the Kravyāda fire). banta Paddhati and Other Works of the Natha Yogis [Poona: Oriental Book House, 34. In the "complete" system, the six Indian seasons correspond to the six 1954]). "tastes" (rasas), each of which is most appropriate to one of the seasons: Zimmer- 45. On the use of the terms saumya and agneya for the male and female sexual man, "Rtu-Sātmya," pp. 94-95, 100-104. fluids in conception, see Suśrūta Sambitā 3.3.3. 35. The effect on the macrocosm of wind, the third element in the triad, is 46. Vaja is the Sanskrit cognate of the English (and Latinate) "vigor." It is also termed viksepa, "dispersion." An explicit paralleling of sun, moon, and wind with related etymologically to the Sanskrit ojas (Monier Monier-Williams, A Sanskrit- their macrocosmic functions and the three dosas is found in Susruta Sambita 1.21.8. English Dictionary [London: Oxford University Press, 1899; reprint Delhi: Motilal 36. January 14, called makara samkränti, is the sole fixed solar date of the luniso- Banarsidass, 1984], s.v. "vāja"), the so-called eighth dhātu (Atharva Veda 2.18.1, with lar Indian calendar: Jean-Luc Chambard, "Les trois grands dieux aux enfers: Tradi- the commentary of Sayana) which is nothing other than the fluid of life itself, of tion orale et cycle des fêtes hindoues dans un village de l'Inde centrale (M. P.)," in which eight drops are located in the heart: Caraka Sambita 1.17.74-75; 1.30.7. Catherine Champion, ed., Littérature populaire et tradition orale en Inde (Purusārtha 47. E. Anantacharya, Rasayana and Ayurveda (Vishakapatnam: World Teacher 17) (Paris: Éditions de l'E.H.E.S.S, 1994), appendix 2. According to Chambard's Publications, 1935, 1982), pp. 12-14, summarizing discussion from the Caraka and informants, the twenty-four-day discrepancy between the solstice date (December Suśruta Sambitās.

364 365 Notes to Chapter Two Notes to Chapter Two

48. Caraka Sambitā 6.1.16-24; Suśruta Sambita 4.29.10-12; and the A.D. 1131 (3.27.1-6) five "world protectors" (lokapala), the gods of the cardinal directions, are Manasollāsa (2.14-51). Dalhana's commentary to Suśruta Sambitā 3.3.4 provides a identified with five colors and five elements. discussion of the three-chambered female reproductive system. For further discus- 58. It is found, for example, in the alchemical Rasendra Mangala (1.9): pañcat- sion, see Arion Roşu, "Yoga et alchimie," Zeitschrift der Deutschen Morgenländischen vam gati. Gesellschaft 132 (1982), p. 374; and idem., "Consideration sur une technique du Ra- 59. Eliade, Yoga: Immortality and Freedom, p. 32; and Haraprasad Shastri, ed. sāyana ayurvédique," Indo-Iranian Journal 17 (1975), 7-28, which indicates parallels Advayavajrasamgraba, Gaekwad's Oriental Series, no. 40 (Baroda: Oriental Insti- with the Vedic initiation hut (diksasala) and similar Taoist structures. Archaeologi- tute, 1927), p. xxviii. cally excavated pre-4th c. A.D. Buddhist ascetics' cells in the Barābar, Nāgārjuni, 6o. Giuseppe Tucci, Tibetan Painted Scrolls, 3 vols. (Rome: Libreria dello Stato, and Guntupalli Caves clearly reproduce such a double-womb structure: Alexander 1949), vol. I, pp. 233, 243. Cunningham, Archaeological Survey of India, Four Reports Made during the Years 1862- 61. Louis de la Vallée Poussin, "Le Bouddhisme et le Yoga de Patañjali," Mé- 63-64-65 (Simla: Government Central Press, 1871), plate 19 and pp. 44-53; De- langes chinois et bouddhiques [Brussels], vol. 5 (1936-37), pp. 224-27. bala Mitra, Buddbist Monuments (Calcutta: Sahitya Samsad, 1971), pp. 44-46 and 62. See Kāthaka Upanisad 3.10-11 and 6.7-8, and its discussion by Madeleine figs. 14-15. Biardeau in Asian Mythologies, s.v. "Puranic Cosmogony," p. 44 49. See Goraksa Śataka (verse 133 of Das Goraksaśataka, ed. by Karl A. Nowotny 63. Gerald James Larson, Classical Samkbya: An Interpretation of its History and [Wetterschlick b. Bonn: Richard Schwarzbold, 1976]): nābhideśe vasati eko bhās- Meaning, 2d ed. (Delhi: Motilal Banarsidass, 1979), pp. 171-76. karo dahanātmakah/ amrtātmā sthito nityan tālumūle ca candramāḥ. 64. Ibid., p. 236. 50. Aghehananda Bharati, The Tantric Tradition (London: Rider and Co., 1965; 65. There is a dynamic relationship that obtains between the tanmätras and the New York: Grove Press, 1975), p. 264, citing Vansidhar Sukul Vaidyaraj, Vama- mababhūtas: sound generates ether; touch, catalyzed by ether, generates air; color, mārg (Allahabad: Kalyan Mandir, 1951), p. 11I. catalyzed by air, generates fire, etc. On this, see Priyanaranjan Ray, History of Chem- 51. Lawrence Babb, The Divine Hierarchy: Popular Hinduism in Central India istry in Ancient and Medieval India Incorporating the History of Hindu Chemistry by (New York: Columbia University Press, 1975), pp. 128-40. Babb bases his analysis Acharya Prafulla Chandra Ray (Calcutta: Indian Chemical Society, 1956), p. 264- on fieldwork he carried out in village Madhya Pradesh (Chattisgarh district). On 66. T. A. Gopinath Rao, Elements of Hindu Iconography, 2 vols. in four tomes the "meteorological" origins of Goddess worship in the Vedic literature, see A. L. (Madras: Law Road Printing House, 1914-16), vol. 2, part 2, p. 374. Basham, "Notes on the Origins of Saktism and Tantrism," in Religion and Society in 67. Shashibhushan Dasgupta, An Introduction to Tantric Buddbism, 2d ed. (Cal- Ancient India: Sudhakar Chattopadbyaya Commemoration Volume (Calcutta: Roy and cutta: University of Calcutta Press, 1958), pp. 19, 25, 29-33. Chowdhury, 1984), pp. 148-53. I am grateful to Douglas Brooks for bringing this 68. Bhattacharyya, History, p. 224- article to my attention. 69. A transcendant sixth, Vajrasattva, was equated with pure consciousness and 52. Brhadāranyaka Upanisad 6.2.13-16; Chandogya Upanisad 5.10.1-7; Praśna Ādi Buddha, the "primal" Buddha: ibid., pp. 232-33. Upanisad 1.9-10; Kausītaki Upanisad 1.2-3; Satapatba Brābmana 2.1.3.1-9. 70. Dasgupta, Introduction, pp. 85-86. For an expansion of this discussion, see 53. Sankara'a Commentary on the Chāndogya Upanisad (5.10.4), translated in Mi- also Tucci, Tibetan Painted Scrolls, 1:235-39. chel Hulin, La face cachée du temps: L'imaginaire de l'au-delà (Paris: Fayard, 1985), p. 71. Ibid., p. 239. 369, who also refers to Sankara's Commentary on the Brahma-Sūtras (3.1.6). 72. Taittirīya Sambitā 7.1.6.2, cited in Filliozat, Doctrine classique, p. 85. 54. Sankara's Commentary on the Chandogya Upanisad (5.10.6), ibid., pp. 369-70. 73. Satapatba Brābmana 1.6.4.5; Brbadāranyaka Upanisad 1.5.14. Sixteen (or sev- 55. See below, sec. 4 and nn. 88-96. enteen) is also the number of priestly technicians employed in a soma sacrifice (Jan 56. On the evolution from three to five, see David M. Knipe, "One Fire, Three Heesterman, The Inner Conflict of Tradition [Chicago: University of Chicago Press, Fires, Five Fires: Vedic Symbols in Transformation," History of Religions 12 (1972), 1985], p. 210 n. 33). Elsewhere, the soma plant is said to be sixteen petaled: Shashib- pp. 28-41. hushan Dasgupta, Obscure Religious Cults, 3d ed. (Calcutta: Firma KLM Limited, 57. Satapatha Brahmana 6.1.2.19; 10.1.3.1-4. Already in the Atharva Veda 1976), p. 251.

366 367 Notes to Chapter Two Notes to Chapter Two

74. Brbadāranyaka Upanisad 1.5.15-16. On such "participation in number" see 88. This is stated most graphically in the five-fires doctrine of the Chandogya Betty Heimann, Facets of Indian Thought (London: Allen and Unwin, 1964), pp. Upanisad (5.4.1-5.9.2), which constitutes the earliest systematic Indian statement 95-99, "The Discovery of Zero and Its Philosophical Implications in India." See on the cycle of death and rebirth in its relationship to the cycle of the year and the above, n. 3. food chain. 75. Monier-Williams, Sanskrit-English Dictionary, s.v. "kala." This carries over 89. In this regard, the Mababbarata (12.293.3-7) statement that the other fifteen into the hathayogic physiology, in which the adbaras, the "supports" of the subtle digits of the moon are merely "vestures" (kosas) of the immortal digit is a significant body, are sixteen in number: SSP 2.10-25; HYP 3.73. one. For a discussion of the fifteen fiery and one lunar kalas in the microcosm 76. Dasgupta, History, 2:317. and macrocosm, see Ramlal Srivastav, ed., Yogavānī: "Gorakb" Visesānk (Gorakhpur: 77. Aitareya Brābmana 1.3. Gorakhnath Mandir, 1977), p. 40. 78. Satapatha Brābmaņa 10.4.1.17. Prajāpati is also called șodasin, "having sixteen 90. The SSP (1.64) lists seventeen (16 + 1) digits, of which the last is alternately parts," in Yajur Veda 8.36. The person (purusa) is said to consist of sixteen parts, of called amrta or nivrtti. Cf. Amaraughaśāsana, p. 12. which the sixteenth is immortal, in the Chandogya Upanisad (6.7.1-3) and Praśna 91. The SSP (1.65) lists thirteen (12 + 1) solar digits. Upanisad (6.2). 92. Gorakb Bānī Sabadī 89. Cf. the Gyān Chautīsā (v. 2) which, after evoking the 79. Mababbarata 12.224-43; 12.298.15. It is in this context that the use of the sixteen lunar and twelve solar kalās, speaks of the four remaining digits as the im- term soma for the herbal medicines ingested in the kutīpravesa technique takes on mortal abode of the gurudev. The SSP (1.66) however, enumerates eleven (10 + 1) its greatest fullness of meaning: Suśruta Sambita 4.29.4-10. On the duration of the fiery digits in the body. In basic tantric worship practice as well, fire is said to be menstrual cycle, see Yajñavalkya Smrti 79, quoted in the 13th c. Smrticandrika of composed of ten kalās: HT, p. 146. The Kulārnava Tantra (6.37-40) lists and names Devannabbatta, Āhnika Kāņda (an English translation with notes) by J. R. Ghar- sixteen lunar, twelve solar, and ten fiery kalas, which it identifies with the pho- pure, Ist ed. (Bombay: V.J. Gharpure, 1946), pp. 25-26. Cf. Manu Smrti 3.46 nemes. (Manu Smrti, with the commentary of Kulluka Bhatta, ed. Gopala Sastri Nene, 93. This process is also cast, in yogic parlance, as a raising of the yogin's female Kashi Sanskrit Series, no. 114 [Benares: Chowkhamba Sanskrit Series Office, energy (sakti), of the indwelling Goddess herself, towards union with her beloved 1970]). Siva who, identified with the absolute brabman, resides in the moon of the cranial 8o. Jan Gonda, Medieval Religious Literature in Sanskrit. History of Indian Liter- vault. The mounting of the Goddess (sakti-calana) is also described as the awaken- ature, II, I (Wiesbaden: Harassowitz, 1977), p. 171. ing and raising of a female serpent, named kundalini, from the lower abdomen to 81. The association with Siva is especially powerful if the new moon (amāvāsyā) the head. In Buddhist Siddha sources, this female energy is termed avadbūti, can- falls on a Monday, in which case he is called sompati, "lord of the moon": Babb, dali, or dombi, all names for "outcaste" women: Kaviraj, Bharatiy, 2:262. Divine Hierarchy, pp. 109-10. A north Indian guide to the "sixteen Mondays vow" is 94. Švetāśvatara Upanisad 2.12. Kālāgnirudra literally means "Rudra [i.e., the the anonymous Solab Somvar tatha Somvar aur Saumya Prados Vrat Katha (Hardwar: destroyer] of the fire of time." Ranadhir Booksales, n.d.). 95. The number is variable: certain systems describe a system of nine, twelve, 82. RV 8.47.17, in Monier-Williams, Sanskrit-English Dictionary, s.v. "kalā." or even twenty-seven cakras, of which six extend beyond the top of the head into 83. The classical study of this subject is Jan Gonda, "The Number Sixteen," space: "The System of Cakras According to Goraksanātha," in Gopinath Kaviraj, in idem., Change and Continuity in Indian Tradition (The Hague: Mouton, 1965), Aspects of Indian Thought (Calcutta: University of Burdwan, 1966), pp. 229-37. pp. 115-30. 96. HYP 3.4. 84. Heesterman, Inner Conflict, p. 210, n. 33. 97. Asian Mythologies, s.v. "Purānic Cosmogony." See below, chap. 7 nn. 181-86. 85. Jaiminīya Upanisad Brāhmana 3.38.8. 98. In this perspective, kala is at once the inactive aspect of the goddess Sakti 86. Mabanirvana Tantra 10.127 (Arthur Avalon, Tantra of the Great Liberation as principle of manifestation and the active aspect of the god Siva as principle of [London: Luzac & Co., 1913; reprint New York: Dover Publications, 1972]), cited nonmanifestation: Padoux, Vac, p. 89; Gopinath Kaviraj, Tāntrik Sādhanā aur Sam- in Gonda, "Sixteen," p. 116. skrti (Patna: Bihar Rastrabhasa Parisad, 1979), pp. 209-10. 87. Mahabbārata 12.293.4-8. 99. Dasgupta, Obscure, p. 243.

368 36g Notes to Chapter Two Notes to Chapter Three

100. Gorakb Bānī Sabadī 219; Pad 56.2. Cf. HYP 3.3; KJnN 6.16, 28; 14.23, the twentieth century, a basic monetary subunit has been the sixteenth fraction of 26; 17.17. a whole: Gonda, "Sixteen," p. 117. Just as the ancient kārsapana was divided into IOI. Harvey Alper, "The Cosmos as Siva's Language Game: Mantra According sixteen māsa[ka]s, so the pre-1947 Indian rupee was divided into sixteen annas, to Ksemarāja's Šivasūtravimarșinī," in idem., Mantra (Albany: SUNY Press, 1989), whence expressions referring to gold or river water as "sixteen annas pure": Milton pp. 264, 274, 289 n. 44. Singer, ed., Krishna: Myths, Rites, and Attitudes (Chicago: University of Chicago 102. The matter is much more complex than this, bringing into play the bindu Press, 1966), p. 84; Briggs, Gorakbnāth, p. 130. Cf. BhP 6.408b-409a, which refers ("drop" of energy, sound, and light) and bija ("seed"), which are identified in a to mercury that has been combined with sixteen "seeds" of other elements and number of ways with Siva and Sakti, according to the textual sources. On the inter- becomes capable of producing gold. relationship of nāda, bindu, and bīja in phonematic emanation, see HT, p. 94, and III. BbP 6.485; 7.11. Padoux, Vac, pp. 96-121, esp. p. 116. In the tantric "initiation of penetration" 112. The mätra (twenty-four seconds) is defined in Brahmananda's commentary (vedbamayī diksā), the guru completes his acoustic penetration of his initiate by tele- to HYP 2.12. Pranayama begins with the holding of the breath for 121/2 matras, scoping the final letters of the Sanskrit alphabet into the bindu. This he unites with with all further yogic practice merely being so many multiples of this basic unit, the kala, which he in turn unites with the nāda: Sarada Tilaka 5.137. See below, culminating in samädbi, total yogic integration, when the breath is held for twelve chap. 1o, sec. 2. days. A similar geometrical progression, working from a basic unit, occurs in the 103. Paul Eduardo Muller-Ortega, The Triadic Heart of Siva: Kaula Tantricism of alchemical domain as well: see below, chap. 10, sec. 3. Abbinavagupta in the Non-dual Shaivism of Kashmir (Albany: SUNY Press, 1989), 113. Basava, vacana no. 537, translated in A. K. Ramanujan, "The Myths of pp. 152-54. The sixteen vowels, especially the visarga, are identified with Siva, the Bhakti: Images of Siva in Saiva Poetry," in Meister, ed., Discourses on Siva, p. 213. consonants with Sakti: Padoux, Vāc, p. 417. Cf. Anne-Marie Gaston, Siva in Dance, Myth and Iconograpby (Delhi: Oxford Uni- 104. Padoux, Vac, P. 79. versity Press, 1982), p. 164; and Stella Kramrisch, The Presence of Siva (Princeton: 105. Pratapaditya Pal, Hindu Religion and Iconology According to the Tantrasara Princeton University Press, 1981), pp. 275-78. (Los Angeles: Vichitra Press, 1981), pp. 18-19. 114. Monier-Williams, Sanskrit-English Dictionary, s.v. "atman." See above, 106. Mababbarata 12.328.13. chap. I, n. 48. 107. Dasgupta, Obscure, p. 47. 108. Padoux, Vac, p. 91, n. 12. On the creation of new "niches" to accomodate Chapter Three an ever-expanding metaphysics, see HTSL, p. 14; Dyczkowski, Canon, p. 125. In a 1. Aung, Folk Elements, pp. 68-69. On the Zawgyis and Weikzas, see ibid., veritable explosion, the number of kalās is quadrupled in some sources to sixty- pp. 41-54 and Needham, Science and Civilisation, vol. 2, "History of Scientific four, itself a widely employed tantric "round number." A Nath Siddha text speaks Thought" (1956), p. 174; and vol. 5, pt. 3, p. 166. See also above, chap. I, n. 6. of the sixty-four lunar digits, which have as their homologues sixty-four locations 2. Bāna's Kādambarī (Pūrvabbāga Complete), ed. by M. R. Kale, 4th revised ed. within the body: Sarasvati Sahagal, Gorakb Darsan (Hyderabad: Madhu Printers, (Delhi: Motilal Banarsidass, 1968), pp. 338-39. Cf. Kādambarī: A Classical Sanskrit 1979), p. 123. The sixty-four practical, mechanical, and fine arts, have also been Story of Magical Transformations, tr. with an introduction by Gwendolyn Layne referred to as kalās since at least the time of the Tantras. (New York: Garland Publishing, 1991), pp. 225-26. 109. Raj Bali Pandey, Hindu Samskaras (Socio-Religious Study of the Hindu Sacra- 3. Alberuni's India, ed. with notes by Edward Sachau, 2 vols. (London: Kegan ments), 2d ed. (Delhi: Motilal Banarsidass, 1969), pp. 17-23. Paul, Trench, and Trubner, Inc., 1910; reprint Delhi: Munshiram Manoharlal, 11O. The end result of the alchemical samskāras can be "sixteen-varna gold," the 1983), vol. 1, pp. 191-92. Variants on this myth are found in both Telugu and Indian homologue of twenty-four-carat gold: Rasabrdaya Tantra (Śrimadgovinda- Cambodian sources: the 15th c. Telegu Bhojarājiyamu of Ananta names the alche- bhagavatpadaviracita Rasbrdayatantram with the commentary of Caturbhuja Misra, mist Sarpați [Carpați]: Arion Roșu, "Alchemy and Sacred Geography in the Medi- tr. into Hindi by Daultarama Rasasastri (Benares: Chaukhamba Orientalia, 1989) eval Deccan," Journal of the European Ayurvedic Society 2 (1992), p. 155. The Cam- translator's introduction, p. xxxi. This carries over into the Indian monetary sys- bodian source calls the alchemist a "brahmin" and the king the "leprous king of tem. Since at least the time of the Manu Smrti (8.136) and down to the middle of Angkor Thom": Jean Filliozat, "Al-Biruni and Indian Alchemy," in Debiprasad

370 371

Notes to Chapter Three Notes to Chapter Three

Chattopadhyaya, ed. Studies in the History of Science in India, 2 vols. (New Delhi: 12. Astānga Samgraba 6.49.245. The Astānga Hrdaya (6.13.36) prescribes the use Educational Enterprises, 1982), vol. I, p. 342, citing G. Porée and E. Maspero, of mercury for timira roga. Arion Roşu (Gustave Lietard et Palmyr Cordier: Travaux Moeurs et coutumes des Khmers (Paris: 1938), p. 73. sur l'histoire de la médecine indienne, Publications de l'Institut de Civilisation In- 4. The Nath Siddha Gopicand's capital city is called Dhara (Dharanagar) in dienne, no. 56 [Paris: De Boccard, 1989], p. lxxxix n. 148), who takes his citations Richard Carnac Temple, Legends of the Panjab, 3 vols. (London: Turner and Com- from Kunte's 1912 Bombay (4th) edition of the AH, and Athavale's 1980 Poona pany, 1884-86; reprint Patiala: Department of Languages, Punjab, 1963), vol. 2, p. edition of the AS respectively, notes the following classic references to mercury in 63 (legend no. 18, v. 68o). these works: AS 6.30.80 = AH 6.13.36; 6.32.31; 6.25.61b-62a; and AS 6.49.392 = 5. Needham, Science and Civilisation in China vol. 5, pt. 4 (1980), pp. 497-98. AH 6.39.162. On the dates of Vagbhatta the Elder, and the closely related Astanga Less than forty years after Marco Polo, the Muslim traveler Ibn Battûta describes Samgraba and Astanga Hrdaya, see Sharma, Ayurved, pp. 172-90. yogis, encountered near Gwalior and in coastal Tamil Nadu, who specialized in the 13. Even prior to the medieval emergence of rasa sāstra, Āyurveda had its own preparation of mineral pills: Ibn Battûta, Voyages, 3 vols., tr. into the French by form of elixir or rejuvenation therapy, called rasāyana, with which it claimed to be C. Defremery and B. R. Sanguinetti with an introduction and notes by Stéphane able to prolong life for one hundred, even five hundred years. Unlike dehavāda and Yerasimos (Paris: Maspero, 1982), vol. 3, pp. 171, 275. rasa šāstra, however, rasāyana preparations are nearly exclusively herbal: Subbara- 6. Bernier, Voyage dans les Etats du Grand Mogol, p. 245. This detail, of Yogis yappa, "Chemical Practice," in Bose, Concise History, p. 316; and Roşu, "Yoga et being great eaters, seems to run against the accepted image of such persons as as- alchimie," pp. 365-66. cetic in their eating habits. Yet, such seems not to be the case with these Yogis, who 14. The earliest list of the eight siddhis is found in the pre-3d. c. A.D. Buddhist have no use for self-mortification. On eating sulfur at daybreak, see Khecarī Vidyā Subābupariprcchašāstra (par. 199, translated by W. Wassiljew, in Der Buddbismus, [KbV] 4.13 (NNA MSS no. 5-6568). All KhV references are to this manuscript. seine Dogmen, Geschichte und Literatur [Saint Petersburg, 1860], vol. I, p. 218). Vari- Elsewhere, as I show in chap. 9, mercury's powers of "consumption" are also prodi- ant lists are found in the 3d-11th c. A.D. Buddhist Sādbanamāla (ed. Benoytosh gious. Bhattacharya, Gaekwad's Oriental Series, 51 [Baroda: 1928, vol. 2, pp. lxxxV, 350 7. One of Oman's later works is entitled Cults, Customs, and Superstitions of India [vv. 3-5], and 509 [vv. 14-16], as well as in the 9th-1oth c. A.D. Hindu KJiN (Kau- (London: 1908). lajnananirnaya and Some Minor Texts of the School of Matysyendranatha, ed. by Pra- 8. John Campbell Oman, The Mystics, Ascetics and Saints of India (London: 1903; bodh Chandra Bagchi, Calcutta Sanskrit Series, no. 3 [Calcutta: Metropolitan, reprint Delhi: Oriental Publishers, 1973), pp. 60-61. I have modernized some of 1934]) of Matsyendranath (5.3-4), and the slightly later Hindu alchemical Tantra, Oman's language. For other accounts of alchemists, see ibid., pp. 56-59. the KCM (NNA MSS no. 5-3969, fol. 2b.3-4). Unless otherwise indicated, all 9. The ca. Ist c. A.D. Artbaśāstra (2.12.2; 2.13.3) of Kauțilya appears to refer to KCM references are to this manuscript. All of these three lists include the siddhi of the amalgamation of gold with mercury in mining, metallurgy, and coinage, when rasa-rasayana. A parallel list, from a Tibetan Buddhist source, the Dragpa jets'en, is it uses the terms rasaviddha and rasapāka: R. P. Kangle, The Kautilya Arthašāstra, 2 found in Stephan Beyer, The Cult of Tāra (Berkeley: University of California Press, vols. (Bombay: University of Bombay, 1960), vol. I, pp. 121-22 and notes; vol. 2, 1973), p. 252. The eight siddhis are defined in Benoytosh Bhattacharya, An Intro- pp- 55-56. Rasa may, however, mean "smelted ore" rather than "mercury" in this duction to Buddhist Esoterism (Oxford: Oxford University Press, 1932), pp. 85-86. case: Subarayappa, "Chemical Practice," in Bose, Concise History, p. 305. In his 6th Three of the eight are found in the 6th c. A.D. Astanga Samgraba of Vagbhatta the c. Brbat Sambita, a work devoted mainly to astrology, Varāhamihira (75.3) mentions Elder: Sharma, Ayurved, p. 179. an aphrodisiac preparation containing mercury (parada) and other minerals. 15. Perhaps the earliest occurance of the term may be construed from the Ma- I0. Caraka Sambitā 6.7.70-71. babbarata (1.165.10) description of Vasistha's wish-fulfilling cow, which in addition II. Suśruta Sambitā 4.25.39; 5.3.14; 6.35.7. Sharma (Ayurved, pp. 76, 89, 113) to milk gives "the six flavors and the fluid ambrosial elixir" (sadrasam cāmrtarasam maintains that the "original" Susruta is more ancient than the Caraka, but that Na- rasāyanam). Cf. Rāmāyana 4.27.3. An alternative Hindu enumeration of the eight gärjuna's 6th-9th c. recension (and addition of a sixth section, called the "Uttara- siddbis, which differs entirely from the eight "magical" siddhis, is first found in tantra") of the Susruta is later than the final recension of the Caraka. Cf. Majumdar, Åpastambha Dharma Stra 2.9.23.6-7. A siddbi not generally included in such "Medicine," in Bose, Concise History, p. 223. Hindu listings is the power to transmute base metals into gold with one's bodily

372 373

Notes to Chapter Three Notes to Chapter Three

secretions. An early mention is verse 74 of the Yogatattva Upanisad (tr. in Jean Vare- 1978), p. 137. Mādhava quotes the RHT, RA, RRS, a lost work entitled Rasasid- nne, Upanishads du yoga [Paris: Gallimard, 1971]), p. 81. dhanta, and that of an author named Sarvajna Ramesvara. 16. The ninth-century Jain author Ugraditya devotes the twenty-fourth and 7. See Dasgupta, Obscure, pp. 254-55. penultimate chapter of his medical work, the Kalyānakāraka, to rasa-rasāyana: 28. The Rasaratnakara of Nityanātha and the Bhūtiprakarana of Goraksanātha: Bhagiratha Prasada Tripathi's foreword to Goraksa Sambitā (Part Two), ed. Pandeya, see below, chap. 5, secs. 4f and 4h. p. 1. Cf. Sharma, Ayurved, pp. 194-95, 225. 29. Amarakośa I.I.1I: vidyādharāpsaro yaksarakso gandharvakimnarāh/piśāco 17. The Chou I Tshan Thung Chhi (A.D. 142), attributed to Wei Po-Yang, is the guhyakah siddho bhūto 'mi devayonayah (Śrīmad Amarasimbaviracitam nāmalin- earliest Chinese canon of theoretical alchemy: Needham, Science and Civilisation, gānušāsanam Amarakośa, 2d ed., ed. with Hindi and Sanskrit commentaries by vol. 5, pt. 4 (1980), p. 248. See below, sec. 3. Brahmananda Tripathi, Chaukhamba Surbharati Granthamala, no. 52 [Benares: 18. RA 18.211-21. This alchemical operation is discussed in chap. 5, sec. 4b. Chaukhamba Surbharati Prakashan, 1982]). The Bhāgavata Purāna (5.24.4) locates 19. RA 18.211. the Siddhas and Vidyadharas at the highest atmospheric level, immediately below 20. RA 17.165a, 166a. The latter part of the Sanskrit phrase translated here the spheres of the sun and Rahu, the "descending node" of the moon. In the Bud- reads: pūrvam lohe parīkșeta tato dehe prayojayet. dhist Mañjuśrī Mūlakalpa, the bodhisattva Vajrapāņi, the foremost of the Vidyā- 21. So, for example, Nägārjuna, the 6th-9th c. final redactor of the Suśruta Sam- dharas, is iconographically placed to the left of Säkyamuni: David Snellgrove, Bud- bita, removed salyatantra from the canon: for a survey of the question, see Arion dhist Himalaya: Travels and Studies in Quest of the Origins and Nature of Tibetan Roşu, "Le renouveau contemporain de l'Ãyurveda," Wiener Zeitschrift für die Kunde Religion (Oxford: Cassirer, 1957), p. 287. See also above, chap. I, n. 10. Sidasiens 26 (1982), p. 65; and Misra, Ayurvedīya, pp. 9-10, 18; and Sharma, 30. See above, chap. I, n. 13. Ayurved, p. 540. 31. Translation in Swami Hariharananda Aranya, Yoga Philosopby of Patañjali, 22. On rasa sästra, which Prafulla Chandra Ray and others have termed "iatro- with the commentary of Vyasa (Calcutta: University of Calcutta, 1981), pp. chemistry," see Subbarayappa, "Chemical Practice," in Bose, Concise History, p. 314. 333-34- Cf. Roşu, "Yoga et alchimie," p. 365. 32. Priyavrat Sharma, Indian Medicine in the Classical Age, Chowkhamba Sanskrit 23. On rasacikitsā, see Roşu, "Renouveau," pp. 65-66; and Majumdar, "Medi- Studies, vol. 85 (Benares: Chowkhamba Sanskrit Series Office, 1972), p. 99. As an cine," in Bose, ed. Concise History, p. 233. Both of these scholars relate rasacikitsā Ayurvedic category, bhütavidya is limited mainly to mental disorders and complica- to the south Indian school of Siddha medicine. See below, chap. 4, nn. 112-14 for tions in childbirth, both of which arise from demonic influences. a discussion of the medieval Mãheśvara Siddhas, who were the likely pioneers of 33. The classic work on the subject is Jean Filliozat, Étude de démonologie in- this tradition. dienne: le Kumāratantra de Rāvana et les textes parallèles indiens, tibétains, chinois, cam- 24. In the words of Alberuni, "The greediness of the ignorant Hindu princes bodgien et arabe, Cahiers de la Société Asiatique, rè série, vol. 4 (Paris: Imprimerie for gold-making does not know any limit": Sachau, Alberuni's India, 1:193. Nationale, 1937). This work is an expansion of an earlier article by Filliozat, "Le 25. The complementarity of mercurial alchemy and breath control is already Kumāratantra de Rāvaņa," Journal Asiatique 226 (1935):1-66. The Kumāra Tantra present in the RA (1.18) and other texts from the tantric period. It is the greatly as an independent work is certainly pre-11th century; much of the material in it increased emphasis on the latter that distinguishes Siddha alchemy from tantric draws on the Atharva Veda, the epics, and the Caraka and Suśruta Sambitās. alchemy. On this comparison, see Needham, Science and Civilisation, vol. 5, pt. 5 34. A Sittar alchemical work, the Bogar 7000 (probably of late date) names Räv- (1983), "Spagyrical Discovery and Invention: Physiological Alchemy," p. 257. ana's son Indrajit as an alchemist: personal communication from N. Sethu Raghu- 26. Sarvadarsana-samgraba of Sāyana-Mādhava, ed. with commentary by Vasu- nathan, Madurai, February 1985. dev Shastri Abhyankar, Government Oriental Series, Class A, no. I (Poona: Bhan- 35. Ausadhibbih asura-bbavanesu rasāyaneti: Vyāsa's commentary on Yoga Sūtra darkar Oriental Research Institute, 1924; reprint 1978), p. 202; English translation 4.1 (this verse has a parallel in the Buddhist Abbidhammakośa [7.122]). It has been in The Sarva-darsana-samgraba or Review of the Different Systems of Hindu Philosopby argued that Yoga Sütra 4.1 was a late non-Patanjalian addition made some time after by Madhava Acharya, tr. E. B. Cowell and A. E. Gough (London: Kegan Paul, the 4th c. A.D. On this, see Roșu, "Yoga et alchimie," pp. 369, 371; Vallée Poussin, Trench & Trubner, 1882; 7th reprint ed. Benares: Chaukhamba Amarabharati, "Le Bouddhisme et le Yoga de Patañjali," p. 241.

374 375 Notes to Chapter Three Notes to Chapter Three

36. upasamprāptaḥ kamanīyābhir asurakanyābhir upanītam rasāyanam: Sāmkbya 44. N. Dutt, "Notes on the Nagarjunikonda Inscriptions," Indian Historical Yogadarsana or Yogadarśana of Patañjali, ed. Gosvami Damodara Sastri et al. (Be- Quarterly 7:3 (Sept. 1931), p. 639; Rahula Sankrtyayana," (Recherches Bouddhiques nares: Chaukhambha Sanskrit Sansthan, 1990), p. 392. The ca. A.D. 1040 king II) L'Origine du Vajrayana et les quatre-vingt siddhas," Journal Asiatique 225:2 Bhoja of Dhära is the first commentator to gloss ausadhi as "such elixirs as mercury, (Oct .- Dec. 1934), p. 212. etc." (pāradādi rasāyanādi): for a discussion of these commentaries on Yoga Sūtra 4.1, 45. āryaka nāmā Gopāla-dārakaḥ siddhādeśena samādisto rājā bhavisyati: see Roșu, "Yoga et alchimie," pp. 370-72. See also RA 12.102-3 on the method Mrcchakațikā, Act 2 (Mrcchakatikā of Śrī Šūdraka, ed. by Jagdish Chandra Mishra used by these semidivine and demonic females to recover said fluids; and KCM (fol. [Benares: Chaukhamba Surbharati, 1985], p. 117). For discussions, see P. Lal, Great 6b.4-7a.1) on alchemical means of gaining access to such creatures. Sanskrit Plays in Modern Translation (New York: New Directions, 1964), p. 79; Diva- 37. For a discussion, see Roșu, "Yoga et alchimie," pp. 376-78. The Bhagavata kar Pandey, Gorakhnäth evam unkī Paramparā kā Sābity (Gorakhpur: Gorakhpur Purana is dated to the 6th-12th centuries A.D. University, 1980), p. 67; and Sankrtyayana, "Recherches," p. 213. 38. V. R. Madhavan, Siddha Medical Manuscripts in Tamil (Madras: International 46. Vāsavadattā, vv. 78-79, 87 (in Fitz-Edward Hall, The Vāsavadattā of Suban- Institute of Tamil Studies, 1984), p. 27. While asuras (antigods, "Titans"), bbūtas dhu [Calcutta: Asiatic Society of Bengal, 1859]). (beings, bogeymen), and rāksasas (brahmin ghouls; technically, Rāvaņa was one of 47. The two may have been separate toponyms, for mountains in the same re- these) are all different types of demons according to Hindu demonology (see gion. The Srisailam of the Mahabharata and other Hindu sources would be none Amarakoa 1.1.11), they are often conflated in both popular and literary discourse. other than the site of one of the twelve self-generated (svayambhu) lingas of Siva 39. See for example the rich epic mythology of Kavya Usanas, the chaplain of and modern-day Hindu religious center in the Kurnool district of Andhra Pradesh, the Asuras, whose sorcery was more powerful than the sacrificial knowledge of while the Buddhist Sriparvata would correspond to the modern Nagārjunikonda, Brhaspati, chaplain of the gods. On this, as well as the semantic field of the term some sixty miles to the east. On this question, see Arion Roșu, "A la recherche kavi, from which Kavya Usanas' name as well as terms for "(alchemical) wizard" d'un tirtha énigmatique du Deccan médiéval," Bulletin de l'Ecole Française d'Extrême are generated, see below, chap. 9, sec. 5 and chap. Io, sec. 5. Orient 6o (1969), p. 46. See below, chap. 4, sec. 5a. 40. RA 2.10-II. 48. Mababharata 3.83.16-21; David Lorenzen, The Kāpālikas and Kālamukbas: 41. Hasan Mujtaba and Nafisa Shah, "Taming of the Serpent," Newsline (Paki- Two Lost Saivite Sects (New Delhi: Thomson Press, 1972), p. 51. Other early San- stan), August 1992, p. 82. Gūgā Chauhan, also known by the names of Guru Gūgā skrit references to Srisailam are the Suśrūta Sambitā (4.29.27), the 6th c. A.D. Brhat (or Goga), Gūgā Pīr, Gūgā Vir, or Zahīr Pīr, is counted by the Nath Siddhas as Sambitā of Varāhamihira (16.3), and the Padma Purāna: Roșu, "A la recherche," one of their own. While his principal shrine is located in Rajasthan, Gūga is also P. 39; and idem, "Alchemy and Sacred Geography," p. 153. revered throughout north India, in Pakistan, and the Deccan. On his cult and 49. "Brahmi Inscriptions at Nagarjunikonda," South Indian Epigrapby. Annual legend, see Pushpa Bhati, Rajasthan ke Lok Devta evam Lok Sabity (Bikaner: Kavita Report for the Year Ending 31st March 1927 (Madras: Archaeological Survey of India, Prakashan, 1991), pp. 89-97; Jacky Assayag, "Pouvoir contre 'puissances': Bref essai 1927), pp. 73-74 (Brahmi inscription no. 214). See also Etienne Lamotte, Histoire de démonologie hindou-musulmane," L'Homme 131 (July-Nov. 1994), p. 47; E. C. du Bouddhisme indien, des origines à l'ère Saka (Louvain: Université de Louvain, Lapoint, "The Epic of Guga: A North Indian Oral Tradition," in S. Vatuk, ed., 1958), pp. 377, 382, 583, 589. American Studies in the Antbropology of India (New Delhi: Manohar, 1978), pp. 281- 50. According to Kaviraj (Bhāratīy, vol. I, pp. 411-12, citing Vinaya Pitaka 2.35), 308; and Stuart H. Blackburn, Peter J. Claus, Joyce B. Fluekiger, and Susan Wad- the Buddhists of Sriparvata knew of the rddhis of "transmutation" and "transub- ley, eds., Oral Epics in India (Berkeley: University of California Press, 1989), pp. stantiation." 26-27, 224-27. 51. The Siddhathakas were one of four Sriparvata-based schools to form the 42. On the hathayogic symbolism of this passage, see above, chap. 2, nn. 95-96; Andhra contingent of the 3d c. A.D. Mahasanghika heresy. The Mahasanghikas on a similar Näth Siddha tradition, a legend involving Jālandharanāth and Gopī- were branded heretical because they maintained that nirvana was, rather than the cand, see below, chap. 9, nn. 149-50. extinction of suffering as taught by the Buddha, a "positive faultless state": Dutt, 43. On this great Serpent, called Sesa or Ananta, see below, chap. 7, sec. 6; and "Notes," p. 650, citing Kathavatthu 9.2, 19.6. chap. 8, sec. I. 52. Roșu, "A la recherche," pp. 47-48; idem., "Alchemy," p. 153. I cannot say

376 377 Notes to Chapter Three Notes to Chapter Three

whether the Siddhas of whom Siva is made the Lord here were of the divine or the vol. 5, pt. 5 [1983], p. 285) and Shanmuga Velan (Siddhar Science, p. 18) see him as human variety. The Saivas who apparently eased the Buddhists out of the Andhra a Chinese (Taoist) philosopher who came to south India to study medicine. country included the Pāsupata sect (Bāna's alchemist is called a "Mahā-pāsupata"). 61. Maritime trade between China and ports on India's west coast (Muzuris and These would later be followed by the Kāpālikas and Vīrasaivas: see below, chap. 4, Broach) was under way by the end of the 4th c. A.D .: Needham, Science and Civilisa- n. 163. tion, vol. I (1954), pp. 171, 174, 178-79; vol. 5, pt. 4 (1980), pp. 354, 406-7; and 53. Tirumūlar, Tirumantiram 1463 (1490) quoted in Kamil V. Zvelebil, The Poets vol. 5, pt. 5 (1983), p. 282. of the Powers (London: Rider and Company, 1973), p. 27. 62. Indeed, the very term "alchemy" is likely derived from the Chinese root 54. Thiru N. Kardaswamy, History of Siddha Medicine (Madras: 1979), p. 311, kim/chim ("aurifaction") or the term chin i ("potable gold"): Needham, Science and calls Tirumūlar the disciple of Nandhi Devar, whom he says later incarnated him- Civilisation, vol. 5, pt. 4 (1980), pp. 351-53, 355. For a history of Taoist alchemy in self as the 12th c. Virasaiva saint (and alchemist, in some sources) Allama Prabhu this period, ibid., pp. 213-59; 320-31. (ibid., p. 354). A certain Nandi is cited in three post-12th c. works, the Rasendracū- 63. It should be noted that mercury naturally occurs, in traces, in areas of geo- dāmaņi, the Rasaprakāasudbākara, and the Rasaratnasamucchaya: see below, chap. thermic activity, where (often sulfurous) hot springs throw up small quantities of 5, nn. 16, 211, 220. A certain Nandikeśvara is termed an "authority on alchemy' cinnabar from deep in the earth: Shorter Oxford Economic Atlas of the World, 2d ed. (rasādbikārika) in a 9th c. work; the Jain ayurvedic writer Ugrāditya (see above, n. (London: Oxford University Press, 1959), p. 88. Such traces may have sufficed, 16) calls his guru Srī Nandi: Sharma, Ayurved, pp. 194-95, 225, 480. Nandi is called prior to the importing of foreign mercury, for the fledgling alchemists of the Indian the "god of alchemy" in Kavyamimamsa 6: idem, Indian Medicine in the Classical Age, subcontinent. See below, chap. 7, sec. 3, for the symbolic importance of this geo- p. 42. thermal and geological phenomenon. 55. Eighteen independent lists of the eighteen Sittars are culled from manu- 64. Filliozat, "Taoïsme et yoga," p. 46, citing I-Ching, Ta Tang Hsi-yu chiu fa kao scripts by V. R. Madhavan in his Siddba Medical Manuscripts in Tamil, pp. 20-26. seng chuan, ed. and tr. Edouard Chavannes, Mémoire composé de la grande dynastie 56. Zvelebil, Poets of the Powers, p. 36 and n. 44; and Renou and Filliozat, L'Inde T'ang sur les religieux éminents qui allèrent chercher la Loi dans les pays d'Occident (Paris: classique, vol. 2, p. 163. Sittar medical works are older. Ernest Leroux, 1894), pp. 23-24; and Arthur Waley, "Notes on Chinese Alchemy," 57. Moti Chandra, Sarthavab (Patna: Bihar Rashtrabhasha Parishad, 1953), p. Bulletin of the School of Oriental Studies 6:1 (1930), pp. 22-24. 188, quoted in Sharma, Ayurved, p. 48o n. 1. 65. Needham, Science and Civilisation, vol. 1 (1954), p. 212; ibid., vol. 5, pt. 3 58. There, at the Dhandhayuthapani shrine, associated with Bogar, worshippers (1976), p. 160. The case of Nārayanasvāmin is found in the A.D. 863 Yu-yang Tsa obtain a sacred ash which has curative powers: A. Shanmuga Velan, Siddbar's Science Tsu of Tuan Chêng-Shih: ibid., vol. 5, pt. 4, p. 197. of Longevity and Kalpa Medicine of India (Madras: Sakthi Nilayam, 1963), p. 49. Bo- 66. Bhattacharyya, History, p. 99; Prabodh Chandra Bagchi, "On Foreign Ele- gar is also said to have created the famous image of Murugan at his Hill Temple at ment in the Tantra," Studies in the Tantras (Calcutta: University of Calcutta, 1975), Palani from an alloy of nine poisonous metalloids: Manuel Moreno, "Pañcāmirtam: pP. 48-49. God's Washings as Food," in The Eternal Food ed. Ravindra Khare (Albany: SUNY 67. Su yu ki: Buddhist Records of the Western World, Translated from the Chinese of Press, 1992), p. 159. Hiuen Tsiang (A.D. 629) by Samuel Beal, 2 vols. (London: Trubner & Co., 1884; 59. On Konganar, see below, chap. 4, sec. 2. reprint ed. Delhi: Munshiram Manoharlal, 1969), 2:212-16. For a discussion, see 60. Bogar's principal work, the Sattakantam, which is certainly much later than Thomas Watters, On Yuan Chwang's Travels in India, 629-645 A.D., 2 vols. (London: the dates of the Bogar of legend, has him travel to Rome, Jerusalem, and Mecca as Royal Asiatic Society, 1904-05), 2:200-8. On Hsuan-tsang's role in the translation well: Filliozat ("Taoïsme et yoga," p. 78 [citing Paramananda Mariadassou, Medecine of the Tao te ching, see Filliozat, "Taoïsme et yoga," p. 41. traditionelle de l'Inde, 4 vols. (Pondicherry: Sainte Anne, 1934-38), vol. 3, p. 6 n. 68. Bānabhatta also describes a prince who, "mad for the elixir of life," is dy- 2]) and P. C. Bagchi ("Compte-rendu de Benoytosh Bhattacharyya, édition de la ing from the medicines given him to counter the disease of royal consumption Sadhanamala" Indian Historical Quarterly 6 [1930], pp. 584-87) consider Bogar to (rājayaksma); and a young physician named Rasyana who, although he had mas- have been a south Indian who went to China; Needham (Science and Civilisation, tered Äyurveda in its entirety, takes his own life when he is unable to heal an ailing

378 379 Notes to Chapter Three Notes to Chapter Three

king: Harsacaritra 6.224 and 5.178 (The Harsacarita of Banabhatta with Exhaustive 77. For Tāra, see p. 172 of Hall's edition of the Vasavadatta, cited in Beyer, Cult Notes [Ucchvāsas I-VIII], ed. Pandurang Vamana Kane [Bombay: n.p., 1918; reprint of Tārā, p. 7; pārada [-pinda] is mentioned in verse 181. Delhi: Motilal Banarsidass, 1973]). 78. Cited in Sharma, Ayurved, pp. 178-79, 470. Varāhamihira's Brbat Sambitā, 69. I-Ching, A Record of the Buddhist Religion as Practiced in India and the Malay a northwestern text, is the source of Vagbhatta's early reference to mercury. See Archipelago (A.D. 671-695), tr. J. Takakusu (London: Clarendon Press, 1896; reprint above, n. 9. ed. Delhi: Munshiram Manoharlal, 1966), pp. 134-35. 79. Two images of "To-lo" (Tara) are reported by Hsuan-tsang in the early 7th 70. The two pilgrims were named Hui-sêng and Sung yün. The date was A.D. c .: Beal, tr., Buddhist Records, vol. 2, pp. 103, 174. Patna had overland trade links 520: Needham, Science and Civilisation, vol. I (1954), pp. 207, 209; and vol. 5, pt. 4 with Kathmandu (Alastair Lamb, British India and Tibet, 1766-1910, 2d ed. [Lon- (1980), p. 354. don: Routledge and Kegan Paul, 1986], pp. 3-4, also a locus of Tră worship in the 71. China also imported Indian medicine at the same time as it did Buddhism: early 7th c. (Beyer, Tara, p. 5), and with Taksasila and Khotan, the great hubs of a text entitled Buddha-vaidyarāja-sastra was translated into Chinese in A.D. 230: the Silk Road in the northwest. Needham, Science and Civilisation, vol. 5, part 4 (1980), p. 336. Manuscript frag- 8o. The king of Kanci erected a temple to the "Chinese goddess" in about A.D. ments on Hindu sorcery of the "Kaksaputa" variety, dating from the Tang period 720: Needham, Science and Civilisation, vol. 5, pt. 5 (1983), p. 286. Kāñci, the mod- (A.D. 618-906) and found in Kucha, in Chinese Turkestan, indicate that the Chi- ern Conjeevaram, located forty-three miles south of Madras, was the capital of nese were interested in this Indian tradition as well: Filliozat, "Taoïsme et yoga," p. the Cola empire. Buddhism flowered there briefly, in the period under study here: 76. On the Kaksaputa Tantra, a 13th c. Hindu work attributed to Siddha Nāgārjuna, Lamotte, Histoire, p. 383. see below, chap. 5, sec. 5j and nn. 60-61. 81. Kalyān Sakti Ank (Gorakhpur: Gita Press, 1934), p. 227. This was, moreover, 72. In the 5th c. A.D. Mabāyānasūtrālamkāra Sāstra, for example: Needham, a period in which a great number of royal ambassadors were sent by sea from south Science and Civilisation, vol. 2 (1956), p. 428; and vol. 5, pt. 5 (1983), pp. 270, India to China: one was sent from Kñci, in 720 A.D .: Needham, Science and Civi- 274-75 lisation, vol. 5, part 5 (1983), p. 286. 73. For a full discussion of these practices and their Taoist symbolism, see below, 82. Ernst Obermiller, History of Buddhism (Chos-bbyung) by Bu-ston, 2 vols. (Hei- chap. 7, nn. 87, 92-101. Cf. Needham, Science and Civilisation, vol. 2 (1956), pp. delberg: Harrassowitz, 1931-32; reprint Delhi: Sri Satguru Publications, 1986), 148-50, and vol. 5, pt. 5, pp. 269-75. The Ayurvedic rejuvenation (rasāyana) tech- vol. 2, p. 122; Misra, Ayurvedīya, p. 8; Dutt, "Notes," p. 637, citing Pag Sam Jon nique of kuti-pravesa, "entering the hut," has a close Taoist parallel. See above, Zang by Sumpa Khan-po Yese Pal Jor; 2 vols., ed. Sarat Chandra Das (Calcutta: Presi- chap. 2, n. 48; below, chap. 10, nn. 103-5; and David Gordon White, Myths of the dency Jail Press, 1908), vol. I, pp. 85-86. Dog-Man (Chicago: University of Chicago Press, 1991), pp. 163, 176. 83. Lucette Boulnois, Poudre d'or et monnaies d'argent au Tibet (Paris: Editions 74. Needham, Science and Civilisation, vol. 2 (1956), pp. 425, 427; vol. 5, pt. 5 du C.N.R.S., 1983), p. 124. (1983), p. 283. 84. Needham, Science and Civilisation, vol. 2 (1956), p. 132. Cinnabar also occurs 75. Indeed, the Persian traders who did much of the actual transporting of naturally on islands in the Bay of Bengal, perhaps the Andamans, according to the goods along the Silk Road, especially between the 5th c. and 1oth c., were popularly Suleiman Saudagar, a 9th c. travel account cited in Fitz-Edward Hall, Bibliograpby regarded as alchemists in the Chinese capital of Chang-an, due to the magical pow- (p. 363) summarized in the commentary to the RRS [Śrīvāgbhatācāryaviracitah Ra- ders and precious stones in which they traded: Needham, Science and Civilisation, saratnasamucchaya, ed. with Hindi commentary by Dharmananda Sharma (Delhi: vol. 2 (1956), p. 187. Motilal Banarsidass, 1962)] p. 208) 76. Sharma, Ayurved, pp. 179, 470, 473. Nagarjuna's connection with Tārā may 85. F. E. Treloar, "The Use of Mercury in Metal Ritual Objects as a Symbol of derive from the fact that she was identified, at some level, with Prajñāparamitā, the Śiva," Artibus Asiae 34 (1972), p. 239. deified "Perfection of Wisdom," the eponymous sütra of which he wrote the defin- 86. Sharma, commentary to RRS, pp. 186, 208-9; Sharma Ayurved, p. 477; and itive commentary: Tara-Tantram with an introduction by A. K. Maitra (reprint ed. George Watt, Dictionary of the Economic Products of India, 6 vols. (Calcutta: Superin- Delhi: Bani Prakashan, 1983), p. 19. For a general discussion, see Bharati, Tantric tendent of Government Printing, 1889-96), 2d reprint, in 10 vols. (Delhi: Periodi- Tradition, pp. 58-84. cal Exports, 1972), vol. 5, p. 232.

38o 381 Notes to Chapter Three Notes to Chapter Three

87. Needham, Science and Civilisation, vol. 5, pt. 5 (1983), pp. 280-86. 94. Sharma, commentary to RRS, p. 199; Watt, Dictionary, 5:233. See also 88. There was in fact an ancient overland trade route which linked Szechuan Ching-lang Hou, Monnaies d'offrande et la notion de Trésorerie dans la religion chinoise with India, via Yunnan, Burma, and Assam, and a highly syncretistic alchemical (Paris: Mémoires de l'Institut des Hautes Etudes Chinoises, 1975), p. 124. The tradition arose in Burma in about the 5th c .: Needham, Science and Civilisation, vol. richest modern-day deposits in the entire region are found in Khaydarkan (Shorter 2 (1956), p. 174; vol. 5, pt. 3, p. 166. Oxford Economic Atlas, p. 88), in the central Asian republic of Kyrgyzstan, which 89. Sādbanamālā, ed. Bhattacarya, vol. I, pp. 193-94; vol. 2, pp. 265-66. Both would correspond to the mercury-rich region of Farghana mentioned in the Hudūd of these passages are followed by the colophon: "extracted from the land of the al-'Alam 25-45-49 (Hudüd al-'Alam, "The Regions of the World": A Persian Geog- Bhoțas [Tibet] by the Arya-nāgārjuna-pādas." Cf. ibid., vol. 2, pp. xlv-xlvi, for a dis- raphy 372 A.H .- 982 A.H., tr. and explained by V. Minorsky [1937; reprint ed. Kara- cussion. chi: Indus Publications, 1980], p. 116). 90. While the Dragon Kaju was not founded until the early 12th c., the Ny- 95. Giuseppe Tucci, "Animadversiones Indicae," Journal of the Royal Asiatic Soci- ingma sect to which the "ancient" divinity Quicksilver belongs dates from the ori- ety of Bengal n.s. 26 (1930), p. 147. gins of Tibetan Buddhism itself: Beyer, Tāra, pp. 42-43, 54; Michael Lee Walter, 96. The Tibetan Tanjur contains fifty-nine works by "Nagarjuna" on Buddhist The Role of Alchemy and Medicine in Indo-Tibetan Tantrism (Ph.D. thesis, Indiana Tantra (listed in Bhattacharya, Sādbanamālā, vol. 2, pp. cvi-cviii); Palmyr Cordier University, 1980), pp. 44, 58-60, 90, 215, 216. For a brief survey of the Nyingma lists twelve magico-medical and nonmedical works, in Sanskrit and Tibetan, au- alchemical systems attributed to the 8th c. founders of Tibetan Buddhism, Vima- thored by Nāgārjuna (Nāgārjuna et l'Uttaratantra de la Suçrutasambitā [Antana- lamitra and Padmasambhava (but which in fact date from the 12-13th centuries), narivo: private printing, 1896], pp. 2-3). To these may be added the works of the see ibid., pp. 28-29, 182, and idem, "Preliminary Results from a Study of Two Madhyamika philosopher Nagarjuna, as well as numerous other works attributed Rasayana Systems in Indo-Tibetan Esoterism," in Tibetan Studies in Honor of Hugh to "Nāgārjuna" by other authors. Richardson, ed. Michael Aris and Aung San Sun Kyi (Warminster, Eng .: Aris and 97. Bhaisajya-Nāgārjuna or Bhadant Nāgārjuna, the Keralan author of a medi- Phillips, 1979), pp. 319-24. cal work entitled Rasavaisesika Sūtra (Renou and Filliozat, l'Inde classique, vol. 2, 91. Sharma, Ayurved, p. 470. Tārā was introduced into Tibet by the Nepalese p. 168). princess Tr'itsün, the daughter of Amsuvarman and wife of the first great Tibetan 98. Śrī Năgārjuna Muni, who, according to the ca. 12th c. Rasopanisat (16.10), king Songtsen gampo (A.D. 617-50), who brought a sandalwood image of the god- discovered a method for whitening copper without sulfur. dess to her husband's palace when she married him: Beyer, Tārā, p. 5. 99. Albert Grünwedel, "Die Geschichten der vierundachtzig Zauberer (Mahā- 92. The Tibetan Buddhists nevertheless had and continue to have their own siddhas) aus dem tibetischen übersetzt," Baessler-Archiv 5 (1916), pp. 137-228, cited alchemical tradition. In addition to the alchemical traditions of the Nyingma in Toni Schmidt, "Fünfundachtzig Mahäsiddhas," Ethnos 2-3 (1955), p. 120. schools, the early 11th c. Kalacakra Tantra contains alchemical data. See below, 100. Max Walleser, "The Life of Nagarjuna from Tibetian and Chinese nn. 122-27. Sources," in Bruno Schindler, ed., Asia Major, Hirth Anniversary Volume (London: 93. Sharma, Indian Medicine, pp. 38-42; Sharma's commentary to the RRS, 1923), p. 424. p. 207; Needham, Science and Civilisation, vol. 5, part 5 (1983), pp. 187, 217; D. C. 101. Buddhist legend knows of two Nagarjunas who would have preceded the Sircar, Cosmography and Geograpby in Early Indian Literature (Calcutta: Indian Stud- historical figure I am now discussing. One is Menander's ca. 140 B.c. interlocutor ies: Past and Present, 1967), pp. 74, 89, 106, 108, 128; André Wink, Al-bind: The in the Milinda Panha, the "Questions of King Milinda"; the other is a bodhisattva Making of the Indo-Islamic World, vol. 1: Early Medieval India and the Expansion of who incarnated himself as the king of Kashmir in about the same period, according Islam, Seventh to Eleventh Centuries (Delhi: Oxford University Press, 1990), p. to the 12th c. Rājataramginī of Kalhana (1.173, 177). 232; and Watt, Dictionary, pp. 232-33. On mentions of the countries of Parada- 102. According to the 5th c. Chinese pilgrim to India Fa-hsien, Nāgarjuna also desa and Hingula-desa in the age of Panini, see Sharma's commentary to RRS received the Avatamsaka (or Paramartha) Sutras from the Nagas. On the Chinese 1.68-75 (p. 12). Kautilya, who hailed from present-day Pakistan, knew of hingula: traditions, beginning with Kumārajīva's 5th c. biography of Nāgārjuna, see K. Ven- Arthaśāstra 2.13.23. kata Ramanan, Nagārjuna's Philosopby: As Presented in the Mabā-Prajnāparamitā-

382 383

Notes to Chapter Three Notes to Chapter Three

Śāstra (Delhi: Motilal Banarsidass, 1975), p. 27; and Cordier, Nāgārjuna et l'Uttara- juna may also have been the eye doctor mentioned in Chinese sources as early as tantra, p. 2. The term mabānāga has an esoteric meaning: Tucci, "Animadversiones the 8th c., and reported by Vrnda to have inscribed the instructions for an eye Indicae," p. 140. medicine on a pillar at the Buddhist center of Pâtaliputra (Patna): see below, n. 146. 103. The source of the Tibetan legend is the A.D. 1322 Chos bbyung (History 112. Indeed, Alex Wayman refers to the tantric commentarial tradition (on the of Buddhism) of Bu-ston: Obermiller, History, vol. 2, p. 124. Cf. Bhattacharyya Gubyasamājatantra), founded by Nāgārjuna, who authored the Pindīkrta-sādbana, History, p. 239. The Chos bbyung also relates the legend of the gold-making elixir the Pancakrama, and the Astādasa-patala-vistara-vyākbya, as the "Arya School." He obtained by Nāgārjuna from the intermediate continent. See above, n. 82. dates this tantric Nāgärjuna, the guru of Aryadeva, to the latter half of the 8th c .: 104. The Satavahana king whom these sources call Satavahana or Sadvaha-raja Alex Wayman, Yoga of the Gubyasamajatantra: Arcane Lore of the Forty Verses (New was named Gautamaputra Sātakarņi. Nāgārjuna's "Epistle to a Dear Friend" York: Samuel Weiser, 1980), pp. 84, 91, 96. (Subrllebika) was written to this figure: Venkata Ramanan, Nagarjuna's Philosopby, 113. Bhattacharya, Sādbanamālā, vol. 2, p. xiv; Dutt, "Notes," p. 638. pp. 27-28. Cf. Krishna Gairola, "Les conditions sociales et religieuses à l'époque 114. Tucci, "Animadversiones Indicae," p. 145. de Sātavāhana dans l'Inde (Ie siècle av. J .- C .- IIe siècle ap. J .- C.)," Journal Asiatique 115. This vague reference may have been inserted to square with another tradi- 243 (1955), pp. 281-95. tion which held that Nagarjuna was first initiated into the tantric or "diamond" 105. Harsacarita 8.283-84. Banabhatta also states in two places (Harsacarita 415 path (vajrayana) by the bodhisattva Vajrasattva, at a place called the Iron Tower in and Kadambari 393) that the tribal peoples of the Vindhya mountains painted their south India: Dasgupta, Obscure, p. 17. swords with mercury (pārada): Sharma, Indian Medicine, p. 217. 116. The earliest such prophecy is found in the Sanskrit Mahāmegha Sūtra, 106. Needham, Science and Civilisation, vol. 5, pt. 3 (1976), "Spagyrical Discov- which was first translated into the Chinese between A.D. 414 and 421 (a relatively ery and Invention: Historical Survey from Cinnabar Elixirs to Synthetic Insulin," early Tibetan translation of this passage, much less confused than those which fol- p. 162. low, is found in the Kanjur [8.2]). An expanded version is found in the Lankāvatāra 107. But Fa-hsien merely reports an account he heard secondhand of the Bhra- Sütra appendix known as the "Sagathakam," which first appears in Chinese transla- maragiri monastery, which he never himself saw: Watters, On Yuan-Chwang's Trav- tion in A.D. 513. The final chapter of the Mañjusrimülakalpa, which is styled after els, 2:208. Cf. Gairola, "Conditions sociales," pp. 289-90. From the 2d through 6th the Sagathakam prophecy, is datable to about the 11th c. A.D. This is followed by centuries, the Andhra Vaitulyakas popularized the cult of Prajnāparamitā- the Tibetan prophecies found in Bu-ston's Chos bbyung (1322) and Taranatha's so closely associated with the philosopher Nägārjuna-in this region: Bhattachar- Dam-pa'i-chos-rin-po-che (1608). This last source further conflates all that had yya, History, p. 222. preceded it with both a Sanskrit work entitled the Mabasiddha-vrttanta and the 108. The meeting occurred near a city called Narasimha, on the banks of the Grub thob. On these sources, see especially Tucci, "Animadversiones Indicae," pp. Chenab River: Watters, On Yuan-Chwang's Travels, vol. I, pp. 286-87. 138-48; Dutt, "Notes," pp. 636, 639; Ariane MacDonald, Le mandala du Mañjuśri- 109. The sole extant manuscript of this work is its Chinese translation, the Ta mūlakalpa (Paris: Maisonneuve, 1962), pp. 2-3, 20; Watters, On Yuan-Chwang's chib Tu Lun, completed by Kumārajiva in A.D. 406. See Etienne Lamotte, ed. and Travels, vol. 2, pp. 200-4; Obermuller, History, vol. 2, pp. 110, 120-28; and Taranā- tr., Traité de la grande vertu de sagesse de Nāgārjuna (Mabāprajnāparamitā šāstra), 4 tha's History of Buddbism in India, ed. Debiprasad Chattopadhyaya, tr. Lama Chimpa vols. (Louvain: Institut Orientaliste, 1949-76), vol. I, pp. 382-83; and Needham, Alaka Chattopadhyaya (Simla: Indian Institute of Advanced Study, 1970), pp. Science and Civilisation, vol. 5, pt. 3 (1976), pp. 161, 164. I02-10. 110. Mabāyāna Samgraba Bhāsya, 221-22. The Abbidbarma Mabāvibhāsa re- chap. 4, sec. I. 117. Caryāpada 22, cited in Dasgupta, Obscure, p. 39. On these songs, see below, counts a myth of gold making by the 3d c. B.c. Candragupta Maurya and his minis- ter "Chanakya," i.e., Kauțilya, the author of the Arthasastra !: Needham, Science and 118. For a list of these works, see Bhattacharya, ed., Sādhanamālā, vol. 2, p. cvi. Civilisation, vol. 5, pt. 3 (1976), p. 164- On the dating of the translation of medical and magical works into Tibetan and 111. Bhattacharya, Sādbanamālā, vol. 2, pp. xliii-xlvi. On Nalanda as a pur- their incorporation into the Tanjur, see Palmyr Cordier, Vägbhata et l'Astangabrday- ported Buddhist seat of alchemical learning, see below, chap. 4, n. 120. This Nāgr- asambitā (Besançon: private printing, 1896), p. 11. One of the works of Nāgārjuna

384 385 Notes to Chapter Three Notes to Chapter Three

found in the Tanjur (Derge 1609; Peking 2480) is entitled Kaksaputa: this work, (5.213-15) description of an alchemical mandala to that of the RA (2.49-82); and however, is distinct from a ca. 13th c. Hindu work by the same title, which does in the Vimalaprabha (5.202b) on a corrosive mineral water called kartarī with a discus- fact contain some alchemical data. See below, chap. 5, secs. 5h and 5j. sion of the same in RA 12.205b-206. The Vimalaprabbā (2.108-111) discussion of 119. Chattopadhyaya, Tāranātba's History, pp. 106, 11O, 243; Dasgupta, Intro- the dynamics of pranāyāma closely resembles early Hindu batha yoga. Cf. Fenner, duction, p. 54. Rasāyan Siddhi, p. 77. If anything, the Vimalaprabba data is slightly more archaic 120. On this, see David Seyfort Ruegg, "Sur les rapports entre le Bouddhisme than that of the RA: it lists only seven mabarasas and six uparasas in contradistinc- et le 'substrat religieux' indien et tibetain," Journal Asiatique 252 (1964), pp. 82-88. tion to the two eightfold categories of the latter Hindu work. Apart from Harsa, the religious syncretism and tolerance of the Chandela kings of 125. Walter, Role of Alchemy, p. 28. Another 14th c. Tibetan work, brought to Khajuraho, as well as the Vijayanagara kings, is well documented: for references, my attention by David Germano, is the Tshig Don mDzod (Precious words and see Roșu, "A la recherche," p. 48. meanings) of kLong Chen Rab 'Byams Pa (1308-63). This work, citing the Thal 121. James B. Robinson, Buddha's Lions: The Lives of the Eighty-Four Siddhas 'Gyur (Direct consequence), discusses bcud len in terms similar to those found (Berkeley: Dharma Publishing, 1979), pp. 88-89, 206, 256-57. Vyādi's Rasasiddbi- in Bo dong-i.e., in the form of both "inner" and "outer" alchemy-in II.3.i .- šāstra (Derge 4313) and Dhātuvādasāstra are found together with the Sarvesvarara- b.2.i.a.2.i, "The meditative session which eliminates attachment to food." säyana in the Åyurveda section of the Tanjur; a work entitled Dhātuvāda is included 126. Bo dong's discussion of "outer" and "inner" alchemy is found in his 14th in the Tantra section: Edward Todd Fenner, Rasayan Siddbi: Medicine and Alchemy c. bCud len gyi man ngag bshad pa in Encyclopedia Tibetica: The Collected Works of Bo in Buddbist Tantras (Ph.D. thesis, University of Wisconsin, 1979), p. 23; and Misra, don Pan-chen phyogs-las-ram-rgyal ed. S. T. Kazi (New Delhi: Tibet House, 1970), Ayurvedīya, p. 21. vol. 2, pp. 507-601. Yet another variety of rasayana is a Tibetan Buddhist practice 122. The Kalacakra Tantra was composed in north India by a native of Java in known by the Mongolian name of tuiurgnikji, which involves the preparation of the late 1oth to early 11th c., after which it was carried to Tibet. Internal evidence holy water: J. Deniker, introduction to Gods of Northern Buddhism, by Alice Getty, indicates that the Vimalaprabba commentary was composed by the same person or p. xi, cited in Dasgupta, Obscure, p. 39. people as compiled the Kalacakra Tantra, in either 1012 or 1027: John Newman, "A 127. See also Robert Thurman, tr., The Tibetan Book of the Dead: Liberation Brief History of the Kalacakra," in The Wheel of Time: The Kalacakra in Context tbrough Understanding in the Between (New York: Bantam Books, 1994), p. 146, on (Madison, Wis .: Deer Park Books, 1985), n. 13 to p. 85, and personal communica- the eighty-four Mahasiddhas as persons who transmuted "demon blood and other tion, Charlottesville, Virginia, February 1995. The only other alchemical system bodily substances" into the elixir of immortality. Certain elements of "outer al- that we may qualify as Buddhist is the Burmese zawgyi or weikza system, but this, chemy" have persisted in Tibet, down to the present day. In a discussion of the although practiced by Buddhist monks, has a decidedly Taoist stamp to it: see extraction of the essence (chu len: i.e., bcud len), the present-day Tibetan Buddhist above, chap. I, n. 6; and above, n. I to the present chap. teacher Nankhai Norbu indicates that a specially prepared mercury pill is the 123. Vimalaprabbā 5.230, and Fenner, Rasāyan Siddbi, pp. 80, 183-84. Mundane "medicine" to be used for overcoming hunger in the "Dharmakāya-style" chu-len mercurial alchemy, whose goals are gold making and the accumulation of power, of Dzog-chen practice: The Teachings of Namkbai Norbu, 1982 (typescript), pp. 89, remains acceptable to the Tibetans, but because the practitioner is dependent upon 93. I am grateful to David Germano for sharing this source with me. external elements rather than his own contemplative procedure, it is deemed a 128. Light metaphysics is absolutely essential to the early Buddhist system of lower form of practice: Beyer, Tara, p. 261. In chap. nine of his 14th c. Rasasāra the Gubyasamaja Tantra. On this, see Wayman, Yoga, pp. 77-83 and throughout. the Hindu alchemist Govindācārya speaks of his indebtedness to the Tibetan Bud- Cf. Walter, Role of Alchemy, pp. 28-29, who notes the parallels between Nying-ma dhists, who he says taught him the art of rangākrsti or rangadruti: Jan G. Meu- and Näth Siddha practice. lenbeld, History of Indian Medical Literature (Leiden: Brill, 1996) [draft of 10 May 129. This is the case only in the very important and influential philosophical 1994], p. 879. summa of Abhinavagupta: on this see Alexis Sanderson, "Meaning in Tantric Rit- 124. Compare the Vimalaprabha (5.206) aphorism on the powers of bound and ual," in Anne-Marie Blondeau, ed., Le Rituel, vol. 3. (Paris: Ecole Pratique des killed mercury to that of RA 1.19; Vimalaprabba 5.204 on testing mercury on met- Hautes Etudes, 5è section, 1996), pp. 15-95. als before consuming it to the classic RA (17.265) statement; the Vimalaprabba 130. Here, I am using the Weberian-rather than a theological-reading of

386 387 Notes to Chapter Three Notes to Chapter Three

the term soteriology: for a discussion see Gellner, Monk, Housebolder, and Tantric tyābnikatilaka, in Newari script, is dated to A.D. 1197 by Shastri (Catalogue, 1:lxiv, Priest, p. 6. 11I-12) and to A.D. 1395 by Bagchi (KJiN, p. 67). Further evidence for the an- 131. Ancient pilgrimage sites, located at points of contact with China, in eastern tiquity of the southwestern Indian origins of this cult is found in an inscription, and (perhaps) northwestern India, the four pithas are mentioned in the Sadbana- dated to A.D. 1030: Saletore, "Kanaphața Jogis," pp. 18-19. mālā (ed. Bhattacharya, vol. 2, p. 453). They are quickly internalized to become 139. The Grub thob brgyad cu rtsa bzhi lo rgyus is found in vol. 86, pt. I of the part of the yogic landscape of the subtle body in Buddhism (Caryāpada 2.4; 4.3, in Tanjur: Palmyr Cordier, Catalogue du fonds tibétain de la Bibliothèque Nationale, vol. Per Kværne, An Anthology of Buddbist Tantric Songs: A Study of the Caryagiti [Oslo: 3, Index du Bstan-hgyur (Tibétain 180-332) (Paris: Bibliothèque Nationale, 1915), Universitetsforlaget, 1977; reprint Bangkok: White Orchid, 1986], pp. 76) and p. 247. Cf. Tucci, "Animadversiones Indicae," p. 142; and idem., Tibetan Painted later Hinduism (SSS, 1.55-61; 4.118-32, in J. A. Schoterman, The Satsabasrasam- Scrolls, vol. I, pp. 226-27. On the mid-14th c. dating of the Tibetan translation, bitā, Chapters 1-5 [Leiden: Brill, 1982], pp. 64, 141). see Simon Digby, "To Ride a Tiger or a Wall? Strategies of Prestige in Indian 132. In particular, the Vidyāpītha doctrines and practice of the Hindu Kāpālikas Sufi Legend," in Winand Caellewaert and Rupert Snell, eds., According to Tradition: were closely paralleled by the Highest Yoga (yoganuttara) Tantras and the Kāpālika Hagiographical Writing in India (Wiesbaden: Harrassowitz, 1994), p. 105. Yoginī cult of esoteric Buddhism: Alexis Sanderson, "Saivism and the Tantric Tra- 140. Keith Dowman (Masters of Mabämūdra: Songs and Histories of the Eighty- dition" in The World's Religions, ed. Stewart Sutherland et al. (London: Routledge Four Buddhist Siddbas [Albany, N.Y .: SUNY Press, 1985], p. 386), maintains that and Kegan Paul, 1988), pp. 678-79. the "Acts" were narrated, and not authored by Abhayadatta; they were translated 133. This reading of rasāyana as the "coming forth (i.e., the extraction) of the into Tibetan by sMon grub shes rab. The Tanjur also contains lists of the works of elixir" corresponds quite exactly to the Tibetan bcud len. I am grateful to Chris the eighty-four Siddhas. Variants of Kubjika's name are Kucipa, Cubji, Kusūlī, and Wilkinson for this suggestion. See below, chap. 5, n. 84. Kubjipa: Dowman, Masters, p. 206. 134. See below, chap. 5, n. 83. 141. The Hindu and Jain Vidyadharas had their Buddhist homologues in the 135. Schoterman, Satsäbasrasambitā, p. 6; Dyczkowski, Canon, p. 48. The refer- Vidyārājas: Ruegg, "Sur les rapports," p. 83. We have noted Nāgārjuna's connec- ences to the Kubjikamata are too vague to be verified. For an extended discussion, tion with Tara: as the leader of the Mahāvidya goddesses, she is called "Vidyārajni": see below, chap. 5, nn. 85-92. Bhattacharyya, History, p. 225. The Vidyādharas (rig 'dzin) serve as intermediaries 136. The Siddhas and Siddhaugha are worshipped in the Siddha Cakra which for the preparation of the elixir in Tibetan sources as well: Walter, Role of Alchemy, dates from well before the 11th c .: Dyczkowski, Canon, p. 81. See above, chap. I, pp. 36-37 and n. 77 n. 13. 142. My use of Max Weber's terminology is intentional here. After the death of 137. Haraprasad Shastri, A Catalogue of Palm-Leaf and Selected Paper Manuscripts the founder of a given social or religious group, a founder whose personal charisma Belonging to the Durbar Library, Nepal, 2 vols. (Calcutta: n.p., 1905, 1915), vol. I, held the group together, that group's structural cohesiveness must be routinized, pp. lxi, lxiv, lxx, 57, 99-100, 111, 258; vol. 2, p. 80; and Schoterman, Satsābasrasam- i.e., reordered along pragmatic lines, if the group is to survive (unless, of course, a bitā, pp. 5-6, 11-12, 36-39, on the major works of this tradition, all of which date charismatic figure replaces the charismatic founder, which is a rare scenario). from the 14th c. or before: the Kubjikāmata, Cincinimata, Manthānabhairava Tantra, 143. Two complementary processes may have been operative here. On the one Kubjikānityābnikatilaka, Śrīmatottara Tantra, and the ȘSS. Closely identified with hand, such names as "Matsyendra" may have been titles bestowed upon persons the latter two works are the figures of Manthanabhairava and Goraksanātha; men- who had reached a sublime level of initiation (Tucci, "Animadversiones Indi- tioned in the Kubjikānityābnikatilaka (NNA MSS no. 3-383; cited in Bagchi, KJnN, cae," pp. 133-34; cf. John K. Locke, Karunamaya: The Cult of Avalokitesvara- p. 68) is a certain Matsyendranätha. Matsyendra appears to identify himself with Matsyendranath in the Valley of Nepal [Kathmandu: Sahayogi Press, 1980], pp. Siddhanātha in the Jnānakārika (2.1; in Bagchi, K7nN, p. 116); and in the Nārada 422-27 and notes). On the other, post-13th c. ascetics may simply have claimed to Purāņa (Uttara khanda 69.25, cited in NSC, p. 23), where it is said that Matsyendra actually have been the "original" Gorakhnäth, etc. Simon Digby terms this "charis- resides in Kāmarūpa as Siddhanātha: Kāmarūpa is sometimes identified with the matic impersonation" and makes a cogent argument for its prevalence in what site of Candradvīpa (Bagchi, introduction to KJiN, pp. 29-32). may be the most widely circulated unpublished manuscript in the field of South 138. Schoterman, Satsābasrasambitā, pp. 36-39. A manuscript of the Kubjikāni- Asian studies: "Encounters with Jogis in Indian Sufi Hagiography," typescript of

388 389 Notes to Chapter Four Notes to Chapter Four

paper read at the School of Oriental and African Studies, London, 27 January 1970, as their founder. On this, see Daniel Gold, The Lord as Guru (New York: Oxford p. 32. University Press, 1987), pp. 18, 93-110. 144. Yogaśataka: texte médicale attribué à Nagārjuna, ed. and tr. Jean Filliozat 2. Sanderson, "Saivism," p. 668. The Siddha lists "front-loaded" certain ele- (Pondicherry: Institut Français d'Indologie, 1979), pp. vi, 105; idem., Doctrine ments of the old Siddha cults into their idealized genealogies: Dyczkowski, Canon, classique, pp. 10-11; and Palmyr Cordier, "Récentes découvertes de mss. médicaux pp. 79-82. See above, chapter I, n. 13, and chap. 3, n. 136. sanscrits dans l'Inde (1898-1902)," Le Muséon, n.s. 4 (1903), pp. 336-39. If one 3. HTSL, p. 40. This is particularly the case with the amnaya classification of accepts that the Yogaataka is mentioned in all but name by the late 7th c. Chinese Kaula traditions, as set forth in the Nepali Cincinimatasārasamucchaya: Dyczkowski, pilgrim I-Ching (Sharma, Ayurved, p. 195), then this is a 7th c. work. More tenable Canon, pp. 66-68. is the hypothesis that this was the same Nagarjuna who, in about the 9th c., edited 4. Wendy Doniger O'Flaherty, Dreams, Illusion and Other Realities (Chicago: the Susruta Sambita and added to it the final "Uttara" section. A Tibetan translation University of Chicago Press, 1984), p. 203, evoking comments made by Diana Eck of this work is found in the Tanjur. in a panel discussion of the American Academy of Religion, New York, 21 Decem- 145. A rare reference to mercury is found in Susruta Sambita 6.35.7. Palmyr ber 1983. Cf. Douglas Brooks's discussion (Secret, p. 63) of the structure and func- Cordier offers a compendium of convincing textual references to support the case tion of the śricakra or śriyantra, the greatest of all tantric cosmographs. that this editor was indeed named Nāgarjuna: Nagarjuna et l'Uttaratantra, pp. I-6; 5. The texts of the Vidya Pitha are divided into the yamalas, the "Union Tan- and "Récentes découvertes," pp. 332-33. Cf. Sharma, Ayurved, pp. 76-77, for other tras" and the saktitantras, the "Power Tantras"; this latter group forms the textual "alchemical" or "tantric" Suśruta passages attributable to Nāgārjuna. base for the Trika synthesis: Sanderson, "Saivism," pp. 668-72. The four literary 146. Siddhayoga 16.149: nāgārjuneņa likhitā stambhe pātaliputrake. Prafulla pīthas are also identified with the four primordial pithas of early Hindu and Bud- Chandra Ray (A History of Hindu Chemistry from the Earliest Times to the Middle of dhist tantrism (Jālandhara, Uddiyāna, Pūrņagiri, and Kāmarūpa, mentioned in the 16th Century A.D., 2 vols. [Calcutta: Prithwis Chandra Ray, 1904-09], vol. I, pp. K7iN 8.20-22, which substitutes Arbuda for Purnagiri), with four cakras of the xxxii, liii, xcii) asserts that kajjali was the active ingredient in this preparation. Cf. subtle body, four colors, etc .: Dyczkowski, Canon, pp. 49-55. Sharma, Ayurved, pp. 195, 330. Chinese references to an eye specialist named Na- 6. Abhinavagupta makes no mention of the āmnāya system; it is, however, men- gārjuna begin to appear in the latter half of the 8th c .: Needham, Science and Civi- tioned in the Kubjikāmata (KM), the root tantra of the Western Transmission (Dyc- lisation, vol. 5, pt. 3 (1976), p. 163. zkowski, Canon, p. 66) and the 1oth-14th c. Kulārnava Tantra (3.7-10). See below, 147. Sharma, Ayurved, pp. 229, 290-91, 479. Cakrapānidatta, like Vrnda, also chap. 5, n. 93 quotes extensively from the Yogaśataka: Cordier, "Récentes découvertes," p. 338. 7. It is also along the lines of the amnaya system that Alexis Sanderson organizes 148. Sharma, Ayurved, p. 291. Following Cakrapānidatta, Vangasena would, in his excellent survey, "Saivism and the Tantric Traditions" (especially pp. 680-90). his Cikitsāsārasamgraba, introduce further innovations in mercury-based rasa šāstra; But see Dyczkowski (Canon, p. 87) and Goudriaan (HTSL, p. 14), who characterize but by this time, the tantric alchemical tradition was well established: ibid., p. 295. this classification as "empty": it is, however, to the artificiality of assigning of geo- 149. Ayurvedic works attributed to Nāgārjuna include the Jivasūtra, Yogasāra, graphical coordinates to the four schools that these authors are referring. A fifth Sarasamgraba, Bhesajakalpa, and the Arogyamanjari: Bhattacharyya, History, p. 20; ("upper") amnaya would later be added: HTSL, p. 17. Ray, History, vol. 1, p. xcii; Jean Filliozat, "Liste des manuscrits de la Collection 8. Scholars have been attempting for decades to make guru-paramparās out of Palmyr Cordier conserves à la Bibliothèque Nationale," Journal Asiatique 224 these lists; such may be possible for certain figures: see, for example Sankrtyayana, (1934), pp. 155-73. The Jivasutra and the Bhesajakalpa are preserved in Tibetan "Recherches Bouddhiques," interleaf following p. 218; Bhattacharya, Sadhanamālā, translation only: Sharma, Ayurved, p. 195. vol. 2, pp. xlii-cix; and Tucci, "Animadversiones Indicae," p. 142. 150. See below, chap. 5, n. 4. 9. Tucci, Tibetan Painted Scrolls, vol. I, p. 226. 1o. Hazariprasad Dvivedi (Nath Sampraday, 3d ed. [Allahabad: Lokabharati Pra- Chapter Four kasan, 1982], pp. 26-35) reproduces a number of these lists, noting that a total of I. There is an important distinction between "clan" and "lineage": members of 148 distinct figures can be generated from them. the former claim descent from a divine ancestor; those of the latter claim a human I1. See above, chap. 3, n. 139. A similar list, dating from A.D. 1513, is found in

390 391 Notes to Chapter Four Notes to Chapter Four

the Bhaktapur (Nepal) Museum; another has been found in Java, Indonesia: Locke, unique to Pandeya's unspecified source. I have not been able to locate an edition Karunamaya, p. 422. of the RRS (there are several) in which such a list appears. 12. Sankrtyayana, "Recherches Bouddhiques," p. 220; Dharmavir Bharati, Siddb 21. One modern author emends the first names on this list into an alchemical Sabity, 2d ed. (Allahabad: Kitab Mahal, 1955, 1968), p. 31. On the history of this guru-disciple lineage (raseśvara paramparā), with certain variations: (1) Ādināth; (2) work, see A. I. Vostrikov, Tibetan Historical Literature, tr. from the Russian by Har- Candrasena; (3) Nityānanda; (4) Gorakhnāth; (5) Kapāli; (6) Bhāluki; (7) and Māņ- ish Chandra Gupta (Calcutta: R. D. Press, 1970), p. 66. This text also lists the caste davya: Gananath Sen, Ayurveda Paricaya (Shantiniketan: Visva Vidya Samgraha, and country of the Siddhas. 1942), pp. 12-13, 174. 13. The doctrines exposed in these texts correspond to those of the late Vajra- 22. This and the three italicized names that follow correspond to a Tibetan yāna traditions of the Yoginī-tantras or Yoganuttaratantras: personal communica- guru-paramparā, found in the Grub thob, in which (17) Nagarjuna is the guru of (19) tion from Alexis Sanderson, 6 July 1992. See above, chap. 3, n. 132. Nāgabodhi as well as of (16) Vyādi, Kubjika, and Pankaja. Vyādi is, in turn, the 14. On the dating of this collection, see Dušan Zbavitel, Bengali Literature, His- guru of (15) Kambala: Tucci, "Animadversiones Indicae," p. 142. tory of Indian Literature, vol. 9, fasc. 3 (Leiden: Brill, 1976), pp. 124-25, who also 23. This is the sole Rasa Siddha on whose name the RRA and the RRS 6.51-53 notes that these poems, while tantric, are neither specifically Buddhist nor Hindu. do not agree. Kværne, Anthology, has produced a historical study and quadrilingual (proto- 24. Meulenbeld, History (draft of 10 May 1994), p. 835. The only names listed Bengali, Sanskrit, Tibetan, English) edition of this fascinating collection. by Caturbhujamiśra that do not figure in the HYP list are: Gaja, Khandin (but see 15. P. C. Bagchi ("The Cult of the Buddhist Siddhacaryas," Cultural Heritage of Khanda, no. 22 in the RM/RRS [a] list), and Nāgadeva. India, vol. 4, p. 275) indicates that the enlarged Tibetan translation of these songs 25. The numbering of Siddhas 28 through 33 in RRS (a) 1.5-7 is identical to is found in volumes 47-48 of the Tanjur. In addition to these, the Tanjur contains that found in RRS (c) 1.5-10. Tibetan translations, from the Sanskrit, of a number of other works attributed to 26. The list is found in the seventh chapter of the Varnaratnākara, ed. S. K. these figures. Chatterji and Babua Misra, Bibliotheca Indica, no. 262 (Calcutta: Bibliotheca In- 16. The RA (14.40) evokes, without naming them, twenty-four [Rasa] Siddhas, dica, 1940), pp. 57-58. The Varnaratnakara is the oldest known text written in the which I take to be an indication of its relative antiquity. See below, chap. 5, n. 162. Maithili language. It was composed by Kaviśekharācārya Jyotīvara, the court poet 17. Bibliothèque Nationale (Paris), Fonds Cordier, Sanscrit MSS no. 1222, fol. to King Harisimha Deva of Simraungadh (Mithila, a kingdom whose ancient 3oa.9-30b.2. Meulenbeld (History, draft of 10 May 1994, p. 971) maintains that borders would fall between modern-day Bihar and Nepal), who reigned from A.D. these verses, forming as they do a portion of the tippana (commentary) on the work 1300 to 1321: Dasgupta, Obscure, p. 202, n. I. See below, n. 50. (which begins at fol. 30a.3 of the present manuscript), are posterior to the RM 27. An edited version of this work exists: Anandakanda, ed. Sri S. V. Radha- proper, and borrowed by the commentator, from the RRS. krishna Sastri. Tanjore Sarasvati Mahal Series, 15 (Tanjore: TMSSM Library, 18. This author is so named in order to distinguish him from the Vagbhatta the 1952). The same work constitutes MSS no. 830 (Äyurveda, no. 2) of the Śrī Bhu- Elder and Vägbhatta the Younger, who authored the Ayurvedic Astangasamgraba vaneśvarī Pīth Ausadhāśrama (Gondal, Gujarat) collection, now held at Gujarat and Astangabrdaya, respectively: Sharma, Ayurved, pp. 481-82 (who calls the author Ayurved University, Jamnagar. of the RRS "Rasa-Vagbhatta"). 28. Under the patronage of King Man Singh, the Nath Siddhas of Jodhpur, 19. RRA 3.1.66-69, in Śrīpārvatīputranityanāthasiddbaviracita Rasaratnākarāntar- drawing on Maharashtran sources, also generated lists of eighty-four Siddhas: P. N. gataścaturtho [sic] Rddbikbandah-Vadikhandab, ed. Jivaram Kalidas Shastri, Rasashala Joshi, Mabārastra ke Nātbpanthīya Kaviyom kā Hindī Kāvya (Mathura: Jawahar Pus- Granthamala, no. 9 (Gondal: Rasashala Aushadhashram, 1940). takalaya, 1976), pp. 216-22, cited in Pandey, Gorakhnāth, pp. 104-5. Gorakhnāth 20. Dharmananda Sharma's edition of the RRS lists fifteen additional Rasa Sid- heads an Oriyan list of sixty-four Siddhas: ibid., p. 105. dhas in 1.2-7; whereas Janardana Pandeya, the editor of the Goraksa Samhita (intro- 29. "Sidh Vandanam" is the first poem in the anthology Nāth siddhom kī bāniyām, duction to vol. 2, p. cha), presents a list of sixty-seven Rasa Siddhas which he claims ed. with an introduction by Hazariprasad Dvivedi, 2d ed. (Benares: Kasi Nagari- to be from the RRS, but of which only the first thirty-three are identical to those pracarini Sabha, 1978), pp. 1-2. One also finds evocations of the nine Näths to- listed in Sharma's edition of the RRS (1.2-6a), with the last thirty-four being gether with the eighty-four Siddhas in Muhammad Jayasi's 16th c. Padmavat (Bhar-

392 393 Notes to Chapter Four Notes to Chapter Four

ati, Siddh Sabity, p. 28) and in a modern work by the great Nepali Nath Siddha 39. The nine-plus-sixteen configuration is a commonplace of such Western scholar Narharinath, entitled Nav Nāth Caurāsi Siddh (Benares: Rastriya Press, Transmission sources as the Manthanabbairava Tantra: HTSL, p. 56. The number 1968). Worship of the nine Naths, the eighty-four Siddhas and the Srīvidyā god- sixteen here is linked to the sixteen aksaras of the Nine Näths, who are one with dess Balāsundarī concludes the Nath Siddha initiation ceremony: George Weston the sixteen nityā divinities (sodašanityātmā): Tantrarāja Tantra 2.16. Chapters 42 and Briggs, Gorakhnäth and the Kānphata Yogis (Calcutta: Y.M.C.A. Press, 1938; reprint 43 of the SSS are entitled "Srī Ādināthavaktrāvatāra" and "Śrīmūlanāthāvatāra," Delhi: Motilal Banarsidass, 1982), p. 33. respectively: Schoterman, Satsābasrasambitā, p. 15. 30. Caurangi is the ninth Nath and Caurangi the first Siddha in the AK list. 40. Bharadvaja Sambita, cited without reference in Pandey, Gorakbnath, p. 109. 31. Saletore, "Kanaphața Jogis," p. 19. The KJiN (8.18) names a Konkaņāī-pāda. See below, n. 54. 32. Revana is listed with eight other familiar Naths in the YSA (ed. Sivanath 41. On the Ciñcinimatasārasamucchaya, see Shastri, Catalogue, vol. I, p. lxi. Cf. Yogi, tr. Candranath Yogi [Ahmedabad: Sivanath Yogi, 1924]: on this work, see Dyczkowski, Canon, pp. 67, 69, 90, 138 n. 23, and 191 n. 222; and Schoterman, below, n. 183), pp. 116-35. He figures in modern-day Maharashtran religious post- Şatsābasrasambitā, p. 39. ers of the Nine Naths. On the Vīrasaiva Revana Siddha, see Sūnyasampādane, 42. The Kubjikānityähnikatilaka presents, in addition to the given names of its ed. and tr. S. C. Nandimath, L. M. A. Menezes, R. C. Hiremath, et al. 5 vols. teachers, their caryā- name, their pūjā- name, their gupta- name, their kīrti- name, (Dharwar: Karnataka University Press, 1965-72), vol. 5, p. 374; and Jan Gonda, as well as additional names based on fanciful etymologies. Here, Matsyendra is one Medieval, p. 227. of the seven names of the fifth (in a series of nine, followed by a series of sixteen) 33. HTSL, p. 12. teachers, whose given name is Visnuśarman: Bagchi, KJñN, p. 68. 34. A bibliography of Tamil medical-alchemical works, prepared in 1826 by 43. Sanderson, "Purity and Power," p. 214 n. 110. See below, chap. 5, n. 93. Whitelaw Ainslie, was published by D. V. Subba Reddy in Madras Medical Journal 44. TA 1.7 (vol. I, p. 25). For editions, see bibliography. On the identification 2:3 (July 1972), pp. 150-55. of Macchanda Vibhu with Matsyendra, see Pandey, Gorakbnäth, p. 21. In his com- 35. These names are culled from lists found in Madhavan, Siddha Medical Manu- mentary to this verse, Jayaratha states that "he [Macchanda] is also renowned for scripts, pp. 20-26. I have Sanskritized the Tamil orthography of these names. On having brought down the entire kula-sāstra." this last group, see Ramana Shastri, "Doctrinal Culture," p. 305, on the eighteen 45. Sanderson, "Trika Saivism," in Encyclopedia of Religion, vol. 13, p. 15. "Maheśvara Siddhas of the Suddha Marga." 46. Dyczkowski, Canon, pp. 69, 79-82; HTSL, p. 54. On the name of the "north- 36. This list is taken from Shastri's Sanskrit introduction to the AK (pp. 8-9), ern" Matsyendra's consort, Konkaņa, see Sanderson, "Saivism," p. 681. to which the editor refers as "those renowned as the 18 Siddhas, of which Agastya 47. Dyczkowski, Canon, p. 89. was the first, among the Dravidas." 48. These are summarized in Locke, Karunamaya, pp. 291-300. 37. See above, chap. 3, n. 59. 49. Another geographical explanation for this toponym may be found in the 38. His name is rendered Satnath or Satyanath: he is one of the Siddhas referred Goan site of Candranäth Hill in the Salsete taluka of Goa, on the Konkan coast of to in the "Siddhamata-nirvāhanam" chapter (49) of the 14th-15th c. Sankaravijaya India, upon whose summit the Saiva temples of Candreśvara and Bhūtanāth are of Ānandagiri (Śrīankaravijaya of Anantānandagiri, ed. N. Veezhinathan, with an situated. The custodians of these temples were Näth Siddhas-according to their introduction by T. M. P. Mahadevan [Madras: University of Madras, 1971]). On current pūjari-until a recent date. On these "lunar" toponyms, see above, chap. the Satyanathī (or Satnathi) panth, see Dvivedi, Nath Sampraday, pp. 12, 178-80. 3, nn. 137-38; and below, chap. 5, n. 81. On his identification with Brahma, ibid., and Gorakh Upanisad (in Mallik, SSP, p. 50. In the course of the 12th c., Jaitugi (1191-1210), the Yadava king of Deva- 73). On his alchemical works, and the spread of the Satnathi panth to Uttar Pradesh giri; the western Cālukya king Someśvara III Bhūloka Malla (1127-38), to whom (Srinagar, Pauri Garhwal district) and Karnataka (Dharwar), see NSC, p. 219. The the Manasollasa, which contains long passages on dhātuvāda and rasāyana, is attrib- Nath monastery at Mrgasthali, above the Pasupatinath temple in Kathmandu, uted; and the 12th c. Kalacuriya king Tribhuvanamalla Bijjana, who had close con- belongs to this suborder, as does the monastery at Puri (Orissa): Pandey, Gorakb- nections with Basava, the founder of Virasaivism, all claimed to have reduced Ne- nāth, p. 65. pal to vassal status. The Mallas of Bhaktapur imported brahmins from Bengal to

394 395 Notes to Chapter Four Notes to Chapter Four

serve as temple priests: Sylvain Lévi, Le Népal, 3 vols. (Paris: Ernest Leroux, 1905; 69). Adinātha's role here closely parallels that of Ādi Buddha, who is identified with reprint Paris: Toit du Monde & Errance, 1985), vol. 1, p. 364; Sharma, Ayurved, Vajrasattva in Buddhist tantra: Dasgupta, Obscure, p. 195 p. 311; Bagchi, "Further Notes," Studies, p. 19. Harisimhadeva, the Maithili king 57. The Tantra Mabarnava passage is cited in Dasgupta, Obscure, p. 206. Cf. who commissioned the Varnaratnakara (see above n. 26) in the early 14th c., was Schoterman, SSS, p. 15; and Dyczkowski, "Kundalinī," p. 13. obliged to seek refuge in Nepal when his kingdom was destroyed by Muslims. His 58. Such is explicitly stated in the opening verses of the Sodasanityātantra widow and court found refuge in Bhaktapur with Rudra Mälla, where they exerted (1.7b-8a: reproduced in Shastri, Catalogue, vol. 2, p. 148); and Tārā Tantra 1.24b- a profound influence on the political and cultural history of the Kathmandu Valley 26, where they are called the Kaula gurus. The names of the Nathas in this and in the late Mälla period. Harisimhadeva traced his dynastic roots back to Karna- other lists often end in -ananda, rather than -natha: this is peculiar to the Western taka: Locke, Karunamaya, p. 422 n. 42. Transmission, whose initiates received names with just such endings: Dyczkowski, 51. In terms of mythic geography, the linga at Pasupatinath, the national temple "Kuņdalinī," p. I. of Nepal, is said to have the same origin as that of Gokarnesvara, located on the 59. Schoterman, SSS, pp. 6, 32, 35-38. On this, see also Dyczkowski, Canon, p. coast of northern Karnataka; the Godavari River of Maharashtra state is said to 87; HTSL, p. 37; and Bagchi, KJnN, p. 21. On Tamil traditions of nine Sittars, see have a branch in Nepal; the Deccan forest of Slesmantaka is transferred to the Madhavan, Siddba Medical Manuscripts, p. 22. Kathmandu Valley. In terms of religious institutions, south Indian Saivite brahmins 60. Gorakh Upanisad reproduced in Mallik, SSP, pp. 72-75. A similar list is found and tancrikas controlled Pasupatinath throughout much of the Malla period. Lévi in the Śrīnāthakatbāsāra of Dwarkanāth (ed. Narharinath [Benares: n.p., 1951]), p. surmises that it was the Pasupata "Yogis" who originally transformed Pasupatināth 4; and Shastri, Catalogue, vol. 2, p. 149. from an indigenous deity into a form of Siva: Lévi, Le Népal, vol. 1, pp. 364-65. 61. In his commentary to TA 1.6, Jayaratha quotes an unnamed earlier source Cf. Axel Michaels, "On 12th=13th Century Relations between Nepal and South which in fact makes this identification (vol. 1, p. 24). For an expanded discussion India," Journal of the Nepal Research Center 7 (1985), pp. 69-72. of this complex matter, see below, n. 143; and chap. 8, nn. 33-37. 52. On the Caryā songs, see Locke, Karunamaya, p. 426; on Kubjikā, Mark Dyc- 62. A number of these variant lists are found in Pandey, Gorakhnath, pp. 105-8; zkowski, "Kundalinī, the Erotic Goddess: Sexual Potency, Transformation and Re- Dasgupta, Obscure, pp. 206-7. Cf. Dvivedi's valiant attempts (Nath Sampradāy, p. versal in the Heterodox Theophanies of the Kubjika Tantras," American Academy 169; Nath Siddhom, p. 14) to generate a single genealogy from these numerous lists. of Religion Annual Meeting, San Francisco, 23 Nov. 1992, p. 2 n. 3. I am grateful Kalyani Mallik has attempted to correlate Siddhacarya and Nath Siddha lineages to Dyczkowski for providing me with a typescript of his lecture. in the same fashion: Pandey, Gorakbnath, p. 104. See below, n. 160. 53. This passage is found in the fourteenth chapter of the Kubjikāmata (London, 63. The Bengali drama, variously known as the Goraksa Vijaya or Mīn Cetan, Wellcome MSS no. g501, fol. 69a lines 4-7; fol. 69b, lines 2-3). gives this data: Dasgupta, Obscure, p. 209. No manuscript of this Bengali work, 54. Manthanabbairava Tantra, cited in HTSL, p. 55; Schoterman, ȘSS, pp. 32, variously attributed to Sheikh Faysullā, Kabīndradās, Śyāmdās Sen, and Bhīmsena 37-38. Rãy, predates the 17th c., although it is clearly the product of earlier folk traditions: 55. The nine Näths (divided into divya-, siddha-, and mānava-oghas) are identi- Zbavitel, Bengali Literature, pp. 189-90. The earliest literary work by this name is fied with the bodily orifices in the Tantraraja Tantra 2.2-5 (in Woodroffe, Tantra- the Sanskrit-Maithili play composed between 1412 and 1428 by Vidyapati (Ha- rāja, p. 19), which also (2.58-72) lists nine types of consecrations (abbisekas), in rimohan Misra, ed., Kavikokil Vidyāpati-krt Goraksavijay [Patna: Bihar-Rashtrab- which nine gemstones, which correspond to the eight bodily dhatus, are placed in hasha-Parishad, 1974]) at the behest of Sivasimha, king of Mithila and descendant medicated water: ibid., p. 20. The RC (12.1) identifies the nine gemstones with the of Harisimha Deva, the king for whom the Varnaratnakara was composed early in nine heavenly bodies. See below, chap. 8, n. 234. the 14th c .: Mataprasad Gupta, Prācīn Bhāsā nātak Sangrab (Agra: Agra University 56. The Tantra Mabarnava is quoted without reference by the 17th-18th c. Gor- Press, 1970), pp. 29-37. Nepali popular dramas treating similar themes also pre- akşa Siddhānta Samgraba (ed. Janardana Pandeya, Sarasvatibhavana Granthamala, date the Bengali works: Bagchi, introduction to K7iN, p. 13. An edition of the no. 110 [Benares: Varanaseya Sanskrit Visvavidyalaya Press, 1973], pp. 44-45). I Bengali drama (Gorkba-bijay of Bhīmsen Ray) is by Pancanand Mandal (Calcutta: base my dating of this anonymous work on a citation in it (p. 10) of the Saktisan- Visvabharati Granthalaya, 1949). gama Tantra, which has been dated to the 16th-17th c. by Goudriaan (HTSL, p. 64. YSÃ, p. 14. This work is attributed to Jñāneśvara: see below, n. 183. Cf.

396 397 Notes to Chapter Four Notes to Chapter Four

Dasgupta, Obscure, p. 207 n. 3; and Pandey, Gorakbnāth, p. 106. Unique to this nerī or Kanerī-nāth is so named because he originally hailed from (Vijayanagara Maharashtran tradition is an identification of these nine Naths with "nine Nārā- in) Karnataka or Kanara: Dvivedi, Nāth Sampradāy, p. 78. yaņas" of the Avadhūta sampradāya founded by Dattātreya. Jñāneśvara's commen- 73. See above, chap. 3, n. 138. tary on the Bhagavad Gita, the Jnanesvari, in which identical data is found in 74. Bhasker Anand Saletore, Ancient Karnataka, 2 vols. (Poona: Oriental Books 18.1733-42, is dated to A.D. 1290. Agency, 1936), vol. I, pp. 94, 377-84; and idem, "Kanaphata Jogis," p. 18 and n. 3. 65. Navanāthacaritra of Gauraņa, ed. K. Ramakrishnaiah (Madras: Madras Uni- A similar identification, of a Nath Siddha (Matsyendra) with Avalokiteśvara, is versity Press, 1970), cited in Rou, "Alchemy and Sacred Geography," p. 154; and found in the cult of Lokeśvara-Rato Macchendranath in the Kathmandu Valley. Ashok Prabhakar Kamat, Mabarāstra ke Nātbpanthīya Kaviyom kā Hindi Kāvya (Ma- 75. Pietro della Valle, Travels, vol. 2, pp. 345-46, quoted in Bhasker Anand thura: Jawahar Pustakalaya, 1976), p. 26. A similar list is found in the Bharadvāja Saletore, Social and Political Life in the Vijayanagara Empire (A.D. 1346-A.D. 1646), 2 Sambitā: Pandey, Gorakhnāth, p. 109. vols. (Madras: B. G. Paul & Co., 1934), p. 53. 66. Rose, Glossary, vol. 2, p. 393, reproduced with modifications in Briggs, 76. Jnānesvarī 18.1733-42. The succession he traces is: Matsyendra-Gorakh- Gorakbnatb, p. 77, as chart D. This lineage account is found in the Tabgiqat-i-Cisti: Gahaņīnāth-Nivrttināth-Jñāneśvara. Jñāneśvara's grandfather, Tryambaka Pant, Raya Bahadur Munshi Haradayal Singh, Riport Mardumasumārī Rāja Mārvād Bābat who purportedly encountered Dattätreya on Girnar, was a contemporary of Gora- san|vat] 1891 Isvī, Tīsarābissā, 2 vols. (Jodhpur: Vidyasal, 1895), vol. 2, p. 231. kşa: Pandey, Gorakbnātb, p. 20. 67. Additional partial or impressionistic lists of the Nine Naths are also found 77. Pandey, Gorakbnāth, p. 20, and Kamat, Mabārāstra, p. 288. in the Agni and Brabmavaivarta Puranas, as well as a work entitled the Kedarakbanda 78. Listed are Jālandhara, Govindacandra, Kānha, Nāgārjuna, Kaņerī, Goda- (not to be confused with the opening section of the Skanda Purāna): Pandey, cülī, and Luī-pā: Pandey, Gorakbnāth, p. 20. Gorakbnäth, p. 109. 79. Saletore, "Kanaphața Jogis," pp. 20-21. 68. Särngadhara is the name of the author[s] of the 14th c. Sarngadhara Paddhati 80. Charlotte Vaudeville, Kabir Granthavali (Doba), Publications de l'Institut and Sarngadhara Sambita, both of which contain alchemical data, and of which the Français d'Indologie, no. 12.(Pondicherry: Institut Français d'Indologie, 1957), former also contains data on hatha yoga (and mentions Goraksa in v. 4372). For p. v; Bharati, Siddb Sabity, pp. 333-34. On the ca. 14th c. "canonization" and "an- editions, see bibliography. Sarnganātha is, together with Gorakh and Matsyendra, thologization" of a corpus of vernacular poetry attributed to Gorakh and many of considered to be a founding guru of the old Näth monastery at Kadri, in Karnataka: the other historical nine Naths, see below, chap. 5, sec. 2. see below, n. 74. 81. Śarngadhara Paddhati, vv. 4372-4419 (pp. 662-69 of Peterson's edition). Cf. 69. See especially Briggs, Gorakbnath, pp. 62-77, and chart A (interleaf between Sharma, Ayurved, pp. 197-98. pp- 74-75), which presents twelve alternative listings of the twelve subdivisions of 82. Nowotny, Goraksaśataka, p. 19. the sect. Cf. Singh, Riport, p. 240; and Dvivedi, Nath Sampraday, pp. 12-15; 165-68. 83. Locke, Karunamaya, p. 433. The A.D. 1382 inscription is from Itum Baha; 70. Briggs, Gorakbnäth, p. 65, who notes that Bhartrhari's guru is considered by and the 1391 inscription from the "Gorakhnath" cave in Pharphing, fifteen miles some to have been Jālandharī-pā. south of Kathmandu. This latter inscription commemorates Acintanäth's establish- 71. Padu Kala is located on the main road between Pushkar and Meerta City, ment of an image of Gorakh's feet, for worship in the cave. The feet and inscription approximately 40 km northwest of Pushkar. Narayan Nath's leadership is contested are currently found outside the cave, inside of which an image of the Tibetan lama by the two Naths presently living at the Rathadunda monastery, 15 km to the south, Rinpoche now presides. Acintanath's name curiously resembles that of Goraksa's who claim the rightful succession. Inside the temple containing ten of the samadhis and Caurangi's guru in the Grub thob: Mina-pa is alternately referred to as Acin- in question was also a moldering stack of papers, which Narayan Nāth maintained ta-pã: Dowman, Masters, p. 81; Bagchi, KJñN, p. 23. contained the records of the Vairag Panth, a claim I was unable to verify. According 84. The inscription commemorates the installation of five lingas, together with to Briggs (Gorakhnäth, pp. 122-24), the seat of this suborder is at Pae dhuni, in images of "Gorakşaka," Bhairava, Añjaneya (Hanumān), Sarasvatī, and Siddhi Vi- Bombay. nāyaka (Gaņeśa): Briggs, Gorakbnāth, p. 247. 72. Saletore, "Kanaphața Jogis," pp. 17-19. Cf. Briggs, Gorakhnāth, p. 58. Kar- 85. Pāśupata-Sütras (1.2-38), with the commentary of Kaundinya, ed. R. Anan-

398 399

Notes to Chapter Four Notes to Chapter Four

thakrishna Sastri (Trivandrum: 1940), cited in Gonda, Medieval, p. 218. The Pāśu- tic orders and lineages, certain of its member groups have, at different periods in pata use of the term yoga is glossed by Kaundinya as ātmeśvarasariyoga: ibid., p. 217 history, been highly successful in organizing themselves into economic, political,

n. 16. and military powerhouses, using their itinerancy to develop wide-ranging networks 86. Gonda, Medieval, pp. 165-66. of trading and fighting men who were at times capable of removing kings from 87. The scriptures of the Saiva Siddhantins, which were compiled in north their thrones and replacing them with princes of their own choosing. See below, (Madhya Pradesh, Kashmir) and south (Tamil Nadu) India between the 7th and chap. 11, sec. 3. 1oth centuries, are called the Agamas, which traditionally dealt with four subjects: 95. In addition to the lineage found in the A.D. 1030 Siddhesvara inscription jñāna ("knowledge"), yoga (internalized worship), kriyā ("ritual," external worship), noted above, oral and written traditions concerning the Jūna Ākhāda establishment and carya ("conduct, behavior"). In fact, most of the content of the agamas deals at Ayodhya known as "Siddhagiri's monastery" indicate a lineage comprising ab- with the last two (and the definition of kriya has in fact been altered from "magic, bots whose names end in -nath prior to A.D. 1300, after which their names end in sorcery" to its present meaning). The Saiva Agamas are subdivided into four classes -giri, the traditional Dasnāmi suffix: Peter Van der Veer, Gods on Earth: The Man- (Śaiva, Pāśupata, Soma, and Lākula), with the Saiva group being further subdivided agement of Religious Experience and Identity in a North Indian Pilgrimage Centre, Lon- into left-hand, right-hand, and Siddhānta. The pre-A.D. 924 Kiranāgama does, don School of Economics Monographs on Social Anthropology, 59 (London: Ath- however, contain a section on "sixfold" yoga that appears to anticipate teachings lone Press, 1988), p. 146. of Gorakhnāth: Gonda, Medieval, pp. 164-67, 180-81, 189. 96. According to Pitambaradatta Barthwal (cited in Bharati, Siddb Sabity, p. 88. Dvivedi, Näth Sampradäy, p. 11. Briggs (Gorakhnāth, p. 63) gives a figure of 323), Nath was a title, meaning guru or lama in Bengali Vajrayāna or "Vajranāthī" eighteen, rather than twelve, original Saivite sects. Buddhist traditions, whence allusions to nath in Saraha-pa and Kānha-pā as "one 89. Sabara Tantra, quoted in Goraksa Siddbānta Samgraba, p. 14. Cf. Pandey, whose mind is unwavering" (Dohakośa, pp. 31, 44, 46). Gorakhnath, p. 107. 97. Bharati, Siddb Sābity, p. 325. See below, n. 213. 90. Briggs, Gorakhnāth, p. 63; Dvivedi, Nath Sampraday, pp. 11, 180. 98. The Nimnathis and Parasnathis, both Jain subdivisions, claim descent from 91. It is significant that all but four of the twelve celestial images (jyotirlingas) two sons of Matsyendra. A number of subdivisions are now in the hands of Muslims of Siva are located in northwestern India. The exceptions are Mallikārjuna, at Sri- (Zāfir Pīrs, Rāwals or Nāgnāthis) or foreigners (Dhajjannāthis). The Sepalas, Sa- sailam; Viśvanātha at Benares; Vaidyanātha in eastern Bihar; and Rameśvara pelas, or Samperas are (generally low-caste) snake charmers. Together with the [-nātha] at Ramesvaram. The zd c. B.c .- 2d c. A.D. Kushanas who, based in the Ka- Bāmārg ("left-handed," possibly an offshoot of the Kāpālikas) subdivision, they bul Valley, ruled over much of northwestern India, including the Punjab and west- claim descent from Kāņipā: Briggs, Gorakhnāth, pp. 63, 66, 69, 71, 72, 74. Cf. ern Rajasthan, were the earliest imperial devotees of Siva, to whose name they ad- Dvivedi, Nath Sampradãy, pp. 163-64, 178. See below, chap. 8, n. 193, on the rela- joined the epithet Mahadeva or Maheśvara. tionship between the Nath Siddhas and the Dasnamī Nagas, and chap. I1, sec. 4, 92. This modern list, enumerated by a disciple of Gambhīrnäth, a former abbot for my encounter with a Sapela. of the Gorakhpur monastery, is found in Dvivedi, Nath Sampraday, p. 12. Seven 99. On this, see, for example, Pitambaradatta Barthwal, Yog-Pravāb (Benares: alternative modern lists are given by Pradip Kumar Bandyopadhyay, Natha Cult and Kashi Vidyapith, 1946), p. 56, on the subject of Jñāneśvara's guru, Gahaņīnāth. Mabanad (Delhi: B. R. Publishing Company, 1992), pp. 71-73. Cf. Briggs (Gorakh- Dvivedi (Nath Siddhom, p. 15) notes that the so-called "Saiva" lineages have gener- nāth, pp. 62-75), who attempts a historical reconstruction of the twelve panths (of ally been more tantric in their orientation than the "Gorakhnāthis," whose practices which there are actually at least thirty). reflect their founder's hathayogic legacy. 93. Briggs (Gorakbnāth, pp. 34-38) indicates that the organization of the Nath 100. See Briggs (Gorakhnäth, chart A, interleaf following p. 74) for twelve vari- sampradaya is ideally quite centralized, but that such is not the case in practice. ant lists of the twelve panths; ibid., pp. 62-77, on their tangled lineages; and Avedyanāth, the current abbot of the Gorakhpur monastery, is a primus inter pares, Dvivedi, Nath Sampraday, pp. 166-67, who indicates the four different orders who both on institutional grounds and by virtue of his dynamism and powerful person- claim descent from Hetunäth and the five orders who claim descent from Colinath ality. See below, chap. 11, sec. 3. (Ghoda Coli). The nature of the confusion and competition that arises from too 94. Even if the Näth sampradaya has always been a loose confederation of monas- many groups and too few "slots" is illustrated by the reaction of the Nath Siddhas

400 401 Notes to Chapter Four Notes to Chapter Four

of the Rathadunda monastery when I told them I had just come from the residence guru-disciple model of instruction. Such south Indian traditions as those of the of Narayan Nath at Padu Kala, Rāthadunda's rival for the seat of the Vairag panth: Sittars of Tamil Nadu and the astavaidya brahmins of Kerala (on which see Francis "He's nobody." Zimmerman's excellent Le discours des remèdes au pays des épices: Enquête sur la méde- IOI. In the middle of the 11th c., Alberuni describes "Jogis" with pierced or cine hindoue [Paris: Payot, 1989]) are medical rather than alchemical. bored ears and ochre robes: Singh, Riport, vol. 2, p. 235. Under the heading of IIO. RA 2.2, 91. Cf. RRĂ 1.1.21a, 22a. Kāyastha Cāmunda, the 15th c. author "Jogi," Singh lists six different groups for the case of 19th c. Rajasthan: 1) Näths or of the Rasasanketakalikā, twice mentions the sampradāya of his guru (4.85, 91). Kānphațas; 2) Masāniya Jogīs; 3) Kālavelīs; 4) Aughars; 5) Aghoris; and 6) Rāwals 111. RA 1.27a: na garbhah sampradāyātha raso garbho vidhīyate. (ibid., p. 241). 112. So, for example, the RA (2.39) advises the alchemist to establish his labora- 102. Some hold that Jogi and, in Bengal, Kapalika are the names of scheduled tory in a country where the people are Māheśvaras, i.e., followers of Śiva. castes precisely because they descend from yogins, perhaps Nath Siddhas, who 113. An alchemical author patronized by Bukku I was Visnudeva, author of the broke their vows of celibacy. Indeed, this applies to the majority of 20th c. Nath Rasasindhu and Rasarājalaksmi: Meulenbeld, History (draft of 10 May 1994), p. 943. society: i.e., most Näths are noncelibate householders. For (dated) census figures, Neither he nor his works are mentioned by Mādhava. see Briggs, Gorakbnath, pp. 4-6, 53, where he notes that the 1881 census counted 114. See, for example RU 1.5 and 15.50 and RHT 1.15-17, 20-26, 31-32. See 350,000 "Jugīs" in Bengal. The 1895 census Riport (Singh, vol. 2, p. 235) of Marwar below, chap. 5, nn. 146 and 196. counts 30,213 Jogis, of which 16,427 are male and 13,786 are female. 115. Dasgupta, Obscure, pp. 192-93, 254-55. It is related in the "Prabhulin- 103. Briggs, Gorakbnāth, pp. 63, 67, and passim; Dvivedi, Nath Sampradāy, pp. galīla," of the Bhavisya Purana (cited in Gopinath Kaviraj, Tantrik Vanmay mem 163-64, 178. Certain Näths who trace their lineage back to Kanha-pa wear their Šākta Drsti [Patna: Bihar Rastrabhasa Parisad, 1963], p. 272). kundalas through the lobes, rather than the thick cartilage of their ears: ibid., p. 5. 116. Śūnyasampādane 21.1-9 (vol. 5, pp. 388-95). The same source (p. 374) re- Occasionally one can, by looking at a list, discern changes within a lineage or a lates a similar hostile encounter between Gorakh and Revana, in which the latter's monastery. So, for example, the pothiratan lineage shows a shift from Kānphata to siddhis prove to be stronger than those of the former. See below, n. 167 and chap. Aughar leadership in the succession Macchandar-Goraksa-Ratannāth-Dhar- 8, nn. 110-12. madas (after which all gurus in the lineage have names with the Aughar suffix of 117. Raman Sastri's analysis ("Doctrinal Culture," pp. 300-8) totally confuses -däs): Pandey, Gorakbnāth, p. 107, citing Rangeya Raghava, Gorakhnāth aur unkā matters by attempting to identify the "Mahesvara Siddhas of the Suddha Marga" Yug, pp. 18-19. with the Tamil Sittars. Far more useful discussions are Dasgupta, Obscure, pp. 192- 104. These include the periodical Yog Vani, as well as special editions of the 93, 254-55; and the editor's introduction to the RU (with Hindi translation and same publication, on Gorakhnāth, the vernacular poetry (banīs) of Gorakh, Nāth commentary of Badrinarayan Sarma, 2 vols. [Ajmer: Krsna Gopal Ayurved Bhavan, Siddha hagiography, etc. See bibliography under Srivastav. 1959], vol. I, p. 12). 105. Yogabīja 8, 81, 189 and Gorakşasiddbānta Samgraba, p. 12. For other refer- 118. Roşu ("Alchemy and Sacred Geography," p. 153), who notes that "ac- ences, see Dvivedi, Nath Sampraday, pp. I-3. cording to certain archaeologists, these sanctuaries may have been founded by sid- 106. Aughars suffix their names with -das. Members of the Pa panths, who trace dha ascetics who were practitioners of alchemy (rasavada)." In the balance of this their descent back to Jalandharī-pa, suffix their names with -pa and are for the most article, however, Roșu tends to discount such claims. part Aughars (but do not use the suffix -das). Members of the Kanthadi subsect 119. Visnudbarmottara Purāna 3.40.1-9, translated in Shiva Sheikhar Misra, Fine suffix their names with -kanthad: Briggs, Gorakhnāth, pp. 33, 67; Locke, Karuna- Arts and Technical Sciences in Ancient India with Special Reference to Somesvara's Mana- maya p. 431. sollāsa (Benares: Krishnadas Academy, 1982), p. 1O1. I have been told that such 107. Briggs, Gorakbnath, pp. 44-61, and Nowotny, introduction to Goraksaśa- preparations are still in use, in Rajasthan, where they are called candralepa ("moon- taka, p. 53. coating"), and in Karnataka, where the carving style is called bemarapanth. 108. RA 1.4; KJN 16.48. 120. Sharma, Ayurved (pp. 471 [photo], 473), maintains that there was an al- 109. Daulat Ram Shastri, foreword to Siddhinandan Misra's edition of the RHT, chemical laboratory at Nalanda, noting the remains of an oven found in the ruins p. 11. Modern university-level instruction in rasa sāstra does not fit the traditional there as evidence. However, Satya Prakash (Prācīn Bhārat mem Rasāyan kā Vikās

402 403

Notes to Chapter Four Notes to Chapter Four

[Allahabad: Prayag Visvavidyalaya, 1960], pp. 835-37), who analyzed the chemical 132. Sachau, Alberuni's India, vol. 1, p. 22. That Alberuni did not consider al- content of a number of metal objects found at Nalanda, found no trace of mercury chemy to be a science, but rather a popular superstition is made abundantly clear:

in any of them. ibid., vol. I, p. 193. 121. Personal communication from Dr. Surya Kumar Yogi, Menal, Rajasthan, 133. As I have argued (see above, chap. 3, nn. 122-24), the early 11th c. Vimala-

March 1985. prabba commentary to the Kalacakra Tantra contains alchemical data that is quite 122. Treloar, "Use of Mercury," pp. 232-33. The ruins of the Chandi Bukit identical to that of the Hindu RA, which belongs to the same century.

Batu Pahat temple are located in Kedah, Malaysia. 134. See below, chap. 5, nn. 22-26. 123. Ray, History, vol. 2, pp. xxxviii-xliii. It should be noted that in 1903, shortly 135. Sharma, Ayurved, pp. 310-11, 402-3. after the publication of Ray's first volume, Palmyr Cordier ("Récentes découvertes," 136. Sārngadhara Samhitā, ed. and tr. K. R. Srikanta Murty (Benares: Chau- pp. 347-48) described this text under its proper name of Rasendramangala, noting khambha, 1984), pp. xiii-xiv. that it was "quite identical to the Rasaratnakara recently used by Prof. P. C. Ray." 137. Carl Ernst, "The Arabic Version of 'The Pool of the Water of Life' (Amr- 124. Lévi, "Kanişka," p. 106; Filliozat, Doctrine classique, p. 1O; Eliade, Yoga: takunda)," has identified forty-five manuscripts of the Arabic translation of this

Immortality and Freedom, p. 416. work and notes that there exist two Persian translations, two Turkish translations, 125. Dominik Wujastyk, "An Alchemical Ghost: The Rasaratnākara by Nāgār- and one Urdu translation of this work. One manuscript makes reference to histor-

juna." Ambix 31:2 (July 1984), pp. 70-83. ical events transpiring in the early 13th c .: personal communication from Carl

126. See below, chap. 5, sec. 4j. Ernst, Chapel Hill, N.C. November 1994. See below, n. 153. 127. See above, chap. 3, secs. 2 and 3, for the historical background to the 138. Needham, Science and Civilisation, vol. 5, part 5 (1983), p. 285; Dvivedi,

RM's dramatization. Näth Sampraday, p. 22. For other cases of Hindu-Muslim mystic syncretism in Ben- 128. Ray, History, vol. 2, pp. xli-xlii. He further claims (ibid., p. liii), on the basis gal, see Dasgupta, Obscure, pp. 157-87, on the Bauls of Bengal. See Zbavitel, Bengali of a manuscript colophon, that Govinda, the author of the RHT, "was evidently a Literature, pp. 189-90 on the dating of this literary work in Bengali. Buddhist" in spite of the fact that this text, too, is thoroughly Hindu. 139. Krsna Kumar Bali, Tilla Goraksanath (Haridwar: Pir Kala Nath Trust, 129. Ibid., pp. xlii-xlvi. Ray's note, in which he speaks of his indebtedness "for 1983), p. 17 and throughout. Cf. Bharati, Siddh Sabity, pp. 325-26. the passages cited here" to Shastri, "who has wended his way through the bulky 140. Thus Bandyopadhyay (Nath Cult, p. 43) states (misquoting Sukumar Sen, MS," does not state where in the Kubjikamata Shastri allegedly found these pas- History of Bengali Literature, with a foreword by Jawaharlal Nehru [New Delhi: Sah- sages: they are certainly not in his Catalogue (vol. 1, pp. lxxviii-lxxx), which Ray itya Academy, 1960], p. 42) that "there is no doubt that the Nath cult originated in also cites. Ray's attribution of the description of six-times killed mercury to the Eastern India, probably in Bengal, long before the fourteenth century when we first get Kubjikāmata is repeated by Mircea Eliade (Forgerons et alchimistes, revised ed. [Paris: a reference." For an informed discussion of such theories and their pitfalls, see Bhar- Flammarion, 1977], pp. 112-13) and White ("Why Gurus Are Heavy," Numen, 33 ati, Siddb Sābity, pp. 323-24. Cf. Pandey, Gorakbnāth, pp. 18-19. [1984], p. 52). His misconception, that many of the early alchemical tantras were 141. Bharati, Siddh Sabity, pp. 335-38. See Vaudeville (Kabir Granthavali (Doba), Buddhist, is followed by O. P. Jaggi, Yogic and Tantric Medicine (Delhi: Atma Ram pp. viii-ix; xviii-xx) for the specific terms šūnya and sabaja.

and Sons, 1973), pp. 133-34. 142. Chattopadhyaya, ed., Tāranātha's History, p. vii. 130. Here, it is worthwhile to note that a later Indian historian of alchemy, 143. See above, chap. 3, n. 143, and below, chap. 8, nn. 33-37. Luī-pa is the Bhudeb Mookerji, states in the introduction to volume one of his five-volume Rasa- author of six Buddhist works preserved in the Tanjur: Chattopadhyaya, Taranatha's jala-nidbi, that "it is earnestly hoped that Dr. Sir P. C. Ray will live to bring out a History, p. 393. revised edition of his book, which contains so many misinterpretations ... due, no 144. Chattopadhyaya, Taranātha's History, p. 320. Tāranātha also maintains that doubt, to a hasty and superficial study of the subject." Goraksa was born as a Buddhist by the name of Anangavajra, and was the son of 131. Here, Sharma (Ayurved, p. 474) miscites pp. 125-26 of the 1956 revised the Pāla king Gopāla. one-volume edition of Ray's History, which are a discussion of Tamil Sittar alchemy. 145. See above, n. 63. Khilji's sustained invasion of Bengal began in A.D. 1202. 146. He calls Jalandharī-pa his guru in song no. 36; he refers to kundalas (which

404 405 Notes to Chapter Four Notes to Chapter Four

he identifies as the sun and moon, i.e., the two peripheral channels of the subtle phenomena, in the colophon of which the author identifies himself as Ācārya body) and the unstruck sound in song no. 11: Kværne, Buddhist Tantric Songs, pp. Ghorakha. Curiously, the text that follows (Derge 2378) bears the same title as this 119, 215. one and is attributed to Caurangi, that Siddha with whom Goraksa's legend be- 147. Dvivedi, Nāth Sampradāy, p. 7; Rose, Glossary, vol. 2, p. 390; Moti Singh, comes confused in Tibetan hagiography. I am grateful to Steven Weinberger for Nirgun Sabity: Sāmskrtik Prsthabhūmi (Benares: Nagaripracarini Sabha, 1963), translating this text from the Tibetan. On the carya songs attributed to Ghoraksa p. 52. in the Tanjur, see above, n. 15. 148. Dowman, Masters, pp. 81-90. Taranātha's is a case of mistaken identity: 153. On this date, see Digby, "Encounters with Jogīs," p. 34. Nowotny, Goraksaśataka, pp. 24-25; and Chattopadhyaya, Taranātha's History, p. 154. This is a role Goraksa continues to play in rural Bengal: for a study, see 227. Caurangi gives the same account of himself in verses 1-2 of the ca. 14th c. Sarat Chandra Misra, "On the Cult of Goraksanath in Eastern Bengal," Journal of Prān Sānkalī (in Dvivedi, Nāth Siddhom, p. 19); but the Grub thob myth is clearly the Department of Letters (University of Calcutta) 14 (1927), pp. 1-41. the source of later Hindu legends of Puran Bhagat of the Punjab: ibid., p. 227; and 155. Pandey, Gorakbnāth, p. 6o. See above, n. 74, for a possible parallel develop- Nowotny, Goraksaśataka, pp. 24-25. While the Tibetan sources make Caurangi and ment at Kadri, in Karnataka. Goraksa Pāla princes of Bengal, the vernacular Indian traditions make Caurangi 156. Ernst Bruce Alexander, Statistical Description and Historical Account of tbe the son of king Salivāhana of Sialkot (Jhelum dist., Punjab, Pakistan). The accuracy North-Western Provinces of India [Gazetteer, Northwest Provinces, vol. 6] (Allah- of this tradition is attested by a reference Gahanīnāth (13th c.) makes to Gorakh as abad: Northwest Provinces and Oudh Government Press, 1881), pp. 371, 436. Cf. the guru of Pipa: in the 19th c. Punjabi songs of Püran Bhagat, Pipa is the adoptive Pandey, Gorakbnath, p. 61, citing the Citravali of Usman Kavi without citation. The father of Lona, the "Potiphar's wife" of this story: Dasgupta, Obscure, p. 388; same author states (ibid.) that "Allah ud-Din Khilaji destroyed the famous temple of Temple, Legends, vol. 2, pp. 388-89. See below, chap. 9, sec. 7, for an extended dis- Gorakhnäth in the 13th century," which is chronologically impossible. Its destruc- cussion of this legend. tion by Aurangzeb in the 17th c. (ibid.) is more likely based in historical fact. 149. This is Rahul Sankrtyayana's theory ("Recherches Bouddhiques," p. 228). 157. The ca. 13th c. MBbT, a Sakta Tantra and the sole text with any significant On the Jogi and Kāpālika subcastes, see Dasgupta, Obscure, pp. 368-69; and Briggs, alchemical content to have come out of eastern India, makes no mention of any of Gorakhnäth, pp. 4-6, 53. As far as conversion to Islam from Hinduism is con- the Siddhas. cerned, it has been hypothesized that renouncers who were dispossessed by their 158. Edward C. Dimock, Jr., The Place of the Hidden Moon: Erotic Mysticism in the religious orders for having broken their vows of celibacy were eventually driven to Vaisnava-sabajiyā Cult of Bengal (Chicago: University of Chicago Press, 1966. Re- embrace Islam. Here, their status changed from dispossessed Hindu (called "Jogi" print Delhi: Motilal Banarsidass, 1991), pp. 249-70. in this case) to Islamicized julaba (Kabīr belonged to this group, which straddled 159. A ground-breaking survey of the often sympathetic relations between the Hinduism and Islam). These Islamicized subcastes were then converted, en masse, Nath Siddhas and Sufis in medieval India is Digby, "Encounters with Jogīs." See by the Nath Siddhas, which would thus account for the large numbers of house- also Gold, Lord as Guru, pp. 207-8; and Dominique-Sila Khan, "L'Origine ismaé- holder Näths: personal communication from Nagendranath Upadhyaya, Benares, lienne du culte hindou de Ramdeo Pir," Revue de l'Histoire des Religions 210 (1993), March 1985. pp. 27-47; and idem, "Ramdeo Pir and the Kamadiya Panth," in Folk, Faith, and 150. If at times their work appears to have Hindu overtones, this is because they Feudalism, ed. N. K. Singhi and Rajendra Joshi (Jaipur: Rawat Publications, 1995), cribbed much of their material from such Hindu tantric texts as the "Yoginīsam- PP. 295-327. cāra" of the Jayadrathayamalatantra, the Picumata-Brabmayamalatantra and the Tan- 160. Many vain attempts have been made to generate biographies of the major trasadbbäva. On this and other resemblances between the esoteric Buddhism of the Näths out of the mass of legend on their subjects: Dasgupta, Obscure, pp. 191-92, Highest Yoga Tantras (yoganuttara tantra) and the Yoginī cults of the Kāpālika Vid- 218, 367-98; Dvivedi, Nath Sampraday, pp. 149-80; idem, Nāth Siddbom, pp. 4-21; yāpītha, see Sanderson, "Saivism," pp. 678-79. and Briggs, Gorakhnath, pp. 228-50. 151. Chattopadhyaya, History, p. 227. 161. The entire Gangetic Plain, from Bengal in the east to Jalandhara (Punjab) 152. Vāyutattvam Bhāvanopadeśa of Ghoraksa (Tanjur; Derge 2377; Peking 3219) in the northwest, was politically unified under the Pala kings of Bengal between is less than one folio in length. It is a poetical description of a number of hathayogic the 8th and 12th centuries: Joseph E. Schwartzberg, ed. A Historical Atlas of South

406 407 Notes to Chapter Four Notes to Chapter Four

Asia (Chicago: University of Chicago Press, 1978; second impression, with addi- āya II (A.D. 1419-47): Sūnyasampādane, foreword to vol. 1, p. ii. See also Dasgupta, tional material, New York: Oxford University Press, 1992), p. 146, pl. xiv. Obscure, p. 255. 162. Vāsavadattā, verse 87; and Roșu, "Alchemy and Sacred Geography," p. 153. 168. Kardaswamy, History, p. 354. The linga at Pasupatinath, the "national The RRA (4.8.1-185: Śrīnityanāthasiddbaviracita (Rasaratnākarāntargataś caturtha) shrine" of Kathmandu and Nepal, is also called paraspati, transmutational. Tradi- Rasāyanakbandab, ed. Yadavji Trikamji Acharya [Benares: Chaukhamba Sanskrit tion holds that a man wearing lead bracelets once purposely embraced the linga. Pustakalaya, 1939]); and the AK (1.12.1-200) contain long descriptions of Srisai- The bracelets were turned to gold, but the man, who had come to worship out of lam. See also Arion Roșu, "Mantra et Yantra dans la médecine et l'alchimie indi- desire rather than devotion, was physically unable to leave the temple sanctum until ennes," Journal Asiatique, 274:3-4 (1986), pp. 251, 255. he had removed the bracelets: personal communication from Purna Giri, Hardwar, 163. The Kapalikas, who had important ties with Srisailam, were also especially March 1984. strong in the south, following the 1oth c. Prior to that, they may have been cen- 169. Kardaswamy, History, p. 330. Korakkar's south Indian samadhi is located at tered in the Himalayan regions, where their practices of ritualized sex strongly in- Poiyur (Thanjavur district), a village near Nagapattinam on the Tamil coast: ibid., fluenced the nascent tantric tradition, as documented in the ca. 9th c. Jayadrathayā- p. 354. A Näth Siddha establishment is located at the same site: personal communi- mala Tantra: Gonda, Medieval, pp. 165, 216; Lorenzen, s.v. "Saivism: Pāśupatas," in cation from N. Sethu Raghunathan, Madurai, January 1985. Encyclopedia of Religion; idem, "New Data," in Hiltebeitel, ed., Criminal Gods, p. 231; 170. Korakkar Malai Vagatam (Madurai: G. Ramasamkykone Booksellers, 1968), and idem, Kāpālikas, pp. 106-8. cited in N. Sethu Raghunathan, "Contribution of Yoogi Munivar for Siddha Sys- 164. The oldest existing edifice on the peak of Srisailam itself is the Mallikār- tem of Medicine," Heritage of the Tamils: Siddba Medicine, ed. S. V. Subrahmanian juna temple, which houses the jyotirlinga of the same name. This temple dates from and V. R. Madhavan (Madras: International Institute of Tamil Studies, 1983), p. the 13th-14th c .: Roșu, "A la recherche," p. 34. A Mallikārjuna is located at Śrīpar- 616 vata in the 6th c. Vāsavadattā: Hall, Vāsavadattā, p. II. 171. Roşu, "Alchemy and Sacred Geography," p. 153. On Nityanātha's probable 165. Malati-Madhava (Bhavabhūti's Malatī-Madbava, with the Commentary of Ja- membership in the Näth order, see below, chap. 5, n. 41 and sec. 4h. gaddbara, 3d ed., M. R. Kale, ed. [Delhi: Motilal Banarsidass, 1967]), act 5, vv. I-2. 172. B. Rama Rao and M. V. Reddy, "A Note on Goraksanatha and his Work See also Lorenzen, Kāpālikas, p. 95, and below, chap. 5, n. 84. Yogadīpika," Bulletin of the Indian Institute of the History of Medicine (Hyderabad) 12 166. Sastri, "Doctrinal Culture," p. 304; Bhattacharyya, History, p. 284. The (1982), pp. 34-35. Following this episode, Gaurana goes on to relate the battle Tungabhadra, a branch of the Krishna or Kistna River, flows past Alampur (Kur- between the diamond-bodied Gorakh and the divine-bodied Allama Prabhu: see nool district, Andhra Pradesh), the western "gateway" to Srisailam (Roşu, "Al- above, n. 115. The so-called Yogadīpikā, as these authors describe it, is clearly an chemy and Sacred Geography," p. 155), and on to the modern Hampi (Bellary Old Kannada version of the Hathayogapradīpikā of Svātmarāma, with added mate- district, Karnataka), i.e., the capital of Vijayanagara, whose kings were, in the 14th- rial on the transmutation of base metals into gold: ibid., p. 37. See below, chap. 5, 17th centuries, great patrons of the Virasaivas. Kadri, the great "Yogi" monastery, n. 117. was also located within the borders of this empire: Saletore, Social and Political Life, 173. Roşu, "Alchemy and Sacred Geography," p. 154. Another Hindu version vol. 2, p. 53. In more common usage, the term Antarvedi referred to the Doab, the of this legend is found in the Katbāsaritsāgara (41.57): Ocean, vol. 3, pp. 252-56. region lying between the Ganges and Yamuna rivers in modern-day south central Cf. Watters, Yuan-Chwang, vol. 2, pp. 201-6. See above, chap. 3, nn. 106-107. Uttar Pradesh: D. C. Sircar, Studies in the Geography of Ancient and Medieval India 174. Watt, Dictionary, vol. 5, pp. 232-34. Cf. Roşu, "Alchemy and Sacred Geog- (Delhi: Motilal Banarsidass, 1990), pp. 303-7- raphy," pp. 155-56. 167. This is not the same text as that portion of the Bhavisya Purana cited above 175. The Vijayanagara kings (A.D. 1336-1565), although based at Hampi, were in n. 116. In the Sūnyasampādane, a Vīrasaiva source, Allama Prabhu condemns devotees of Siva Mallikārjuna at Srisailam: Konduri Sarojini Devi, Religion in the Gorakh's alchemy in no uncertain terms: Ramanujan, Speaking of Siva, p. 147. Vijayanagara Empire (New Delhi: Sterling, 1990), pp. 88-89, 176, 180. While Allama Prabhu lived in the 12th c., this work, which casts him as a contem- 176. Roşu, "A la recherche," pp. 44-46, 52. See above, chap. 3, n. 47. porary of Gorakh, was compiled during the reign of the Vijayanagara king Devar- 177. NSC, pp. 215, 217; idem. "Gorakb" Viśesānk, pp. 266-68. There are prob-

408 409 Notes to Chapter Four Notes to Chapter Four

lems of chronology here. Satyanäth is associated with the 16th c. Garhwali king 186. Kaviraj, Bhāratīy, vol. I, pp. 151, 191-92, 197, 203; Briggs, Gorakbnāth, Ajaypal; Ānandagiri, if he is the same hagiographer of Sankara as the Ānandagiri p. I2I. who authored the Sankaravijaya, is a 14th-15th c. author; and Sankaracarya himself 187. Srivastav, "Gorakb" Visesank, p. 119. The 7th c. Bāna mentions that plantain is an 8th century reformer of Hinduism. flourishes along the shore of the Godavari: Sharma, Indian Medicine, p. 145. 178. KJnN 16.5-7; KM 2.5-6; KPT 15.2 and 16.9 (pp. 356 and 367 of Jivananda 188. Chinese and Persian reports of exceptional Indian waters date back to the Bhattacarya, ed. Indrajālavidyāsamgraba (Calcutta: V. V. Mukherji, 1925) Cf. Pan- middle of the 9th c. Like mercury and sulfur, sal ammoniac occurs naturally in dey, Gorakbnātb, pp. 66-68. regions of geothermic activity: Needham, Science and Civilisation, vol. 5, pt. 4 179. Srivastav, ed. "Gorakb" Visesank, p. 119. (1980), pp. 197-98, 435-37. 180. Like the Calukyas of Kalyānī (10th-12th c.) before them, the Yādavas 189. RA 12.236-39. Cf. yet another reference to a Srisailam, where "rock" or of Devagiri (A.D. 1175-1318) ruled an area comprising Maharashtra, Karnataka, "peak water"-śailodaka-is found, as well as a long list of sources of low-grade and the western Deccan (encompassing the region of the Andhran Srisailam): "rock water," most of which are in southwestern India (RA 12.282-87). Schwartzberg, Historical Atlas, pp. 147-48, pl. xiv.3). 190. Kadalīmañjunātha Māhātmya (p. 33), quoted in Pandey, Gorakbnāth, p. 66. 181. While Bhaskara's is an extremely common name, it may well be that the 191. The reference is found in RRS 16.126. See above, nn. 134-35, and chap. same Bhäskara who was eulogized by Siddha Nagarjuna in the opening verse of the 5, n. 269. One could, however, read RA 12.236 ("on the southern path, in the direc- 12th-13th c. Yogaratnamala, was also the author of a lost alchemical work entitled tion of Death") as an indication that the author was from somewhere further to the the Rasendra Bhaskara: Priyavrat Sharma, Nagarjuna's Yogaratnamāla with the Com- north. However, such is a commonplace way of speaking of India south of the mentary of Svetāmbara Bhiksu Guņākara (Benares: Chaukhambha Orientalia, 1977), Vindhyas (Yama, Death, is the regent of the southern quarter). Prior to the Yāda- pp. 9-II; and idem, Ayurved, p. 230. vas, the Cālukyas of Kalyānī ruled this region. While the Mānasollāsa (A.D. 1131), 182. Sharma, Ayurved, pp. 310-12, 402-7, 474, 482, 697. A precious document written by or for the Calukya monarch Someśvara III ("Bhūlokamalla"), contains which records the royal patronage of Ayurvedic physicians and alchemists by the long passages on medicine (1.135-307), sorcery for locating buried treasure early Yādavas is a passage from the Sangītaratnākara (1.2-13), a treatise on music (2.332-61), transmutational alchemy (2.377-94) and the testing of gemstones by Sarngadeva, the grandson of Bhaskara: the passage is reproduced in Sharma, (2.378-536), it shows no direct influence from the RA, which dates from a century Åyurved, p. 311. earlier: Mānasollāsa of King Bhūlokamalla Someśvara, 3 vols, ed. G. K. Srigondekar, 183. This is the Yogisampradāyaviskrti (YSA). In an introduction to the YSA, (p. Gaekwad's Oriental Series, no. 28, 2d ed. (Baroda: Oriental Institute, 1967). ca), Candranath Yogi states that his is an abridged Hindi translation of Jnanesvara's 192. Sharma, Yogaratnamālā, pp. 8-10, 13; idem, Ayurved, pp. 310-11, 402. See translation (into "Maharashtran") of a Bengali original written shortly before his below, chap. 5, nn. 6o and 251. time. He gives no explanation for this "Bengali original," which I find implausible, 193. Sharma, Yogaratnamālā, pp. 14-17. given that the names of the Nath Siddhas treated in the work correspond to those 194. Phyllis Granoff, "Jain Biographies of Nagarjuna: Notes on the Composing found in western Marathi rather than Bengali hagiographic traditions. At this of a Biography in Medieval India," in Phyllis Granoff and Koichi Shinohara, eds., point, I am unable to confirm Candranath Yogi's assertion that this is in fact a Monks and Magicians: Religious Biographies in Asia (Oakville, Ontario: Mosaic Press, work (or translation) originally composed by Jñānesvara; however, apart from some 1988), pp. 48-49 mentioned in a number of appendices, none of the persons figuring in the work 195. The "Satavāhana" king is an obvious carryover from Buddhist traditions. appears to be later than Jnanesvara's late 13th c. dates. The manuscript used by Palittanakapura is to be identified with both the modern Palitana and Mount Sa- Candranath Yogi is dated to A.D. 1773-74- truñjaya, as well as Vallabhi and Pattan, and by extension, Mount Dhank. 184. Pandey, Gorakhnath, p. 65. 196. Summarized in Granoff, "Jain Biographies," pp. 48, 50-51. The Prabandha- 185. The Tantra Mabarnava is a work which locates the Nine Naths at eight cintāmani calls the elixir "ten-million transmutation mercury" (kotavedhi-rasa); the of the nine cardinal directions. Ngarjuna's Godavari location corresponds to the "grinding of mercury" (pārada-mardana) is the second alchemical samskāra: southern quadrant: Dasgupta, Obscure, p. 206. See below, chap. 5, n. 82. Dvivedi, Nath Sampradāy, p. 153.

410 411 Notes to Chapter Four Notes to Chapter Four

197. See below, chap. 7, n. 69. 207. Nāgārjuna's toponym may correspond to Patan, a city in northern Kathia- 198. Granoff, "Jain Biographies," p. 52. Dhank is located sixteen miles (by road) war that has been continuously inhabited since the period in question and near to northwest of Upaleta (Rajkot district, Gujarat). Dominik Wujastyk also describes which is a mountain (Osam) that has long been a site connected to the practice of an image of Nagarjuna-together with that of a figure named "Barot" (who, he Ayurvedic medicine: Kalyān Sakti Ank, pp. 661, 664. The Pattan that Dharamnāth was told, wrote Nagarjuna's biography)-which, located in a private dwelling in the brought down was located in southern Kacch, over two hundred miles to the south- village of Dhank itself, is receiving active worship: letter dated 24 February 1993. west. Both toponyms are, however, within Gujarat, as are a number of other top- 199. According to Shantilal Pranjivan Joshi, who wrote the forward to the 1959 onyms ending in -patan. Ajmer edition of the RU (p. 11), it was with the alchemical gold produced at Valabhī 208. Granoff "Jain Biographies," p. 60, citing Jyoti Prasad Jain, "Jain Authors that the Jains financed the building of the 1oth-14th c. Dilwara temples of Mount and Their Works," Jaina Antiquary 33 [1980], p. 27. Granoff does not give the title Abu, considered by many to be the most beautiful examples of Jain temple architec- of this work. ture in India. 209. Pāduba-doba no. 126, cited in Srivastav, ed., "Gorakh" Visesānk, p. 50. Cf. 200. Valabhī had been the capital of the Maitraka kings, whose charters date Dasgupta, Obscure, pp. 58-60. from A.D. 502 to 766. Popular and bardic traditions, which maintain that Valabhī 210. Elizabeth Sharpe, An Eight-Hundred Year Old Book of Indian Medicine and fell in A.D. 524, are chronologically impossible: D. C. Sircar, The Guhilas of Kiskin- Formulas, Translated from the Original Very Old Hindi into Gujarati Character and dbã (Calcutta: Sanskrit College, 1965), pp. 24-25. On Amru ibn Jamāl, see Histori- Thence into English (London: Luzac & Co, 1937; reprint New Delhi: Asian Educa- cal and Cultural Chronology of Gujarat, ed. M. R. Majumdar et al., 2 vols. (Baroda: tional Services, 1979), p. 6. A 12th-14th c. alchemical text, entitled Rasaratnasa- University of Baroda, 1960), vol. I, p. 226. muccaya, was the work of the Jain monk Mānikyadeva Sūri: J. C. Sikdar, ed. and 201. For a complete rendering of this account, see below, chap. 9, sec. 8. tr., The Rasa-ratna-samuccaya of Manikyadeva Süri (Jaipur: Prakrta Bharati Academy, 202. Sachau, Alberuni's India, vol. I, p. 193. 1986), p. i. It in no way resembles Vagbhatta II's alchemical work of the same title. 203. Personal communication from Dominik Wujastyk, London, May 1992. 211. RC 2.1; RPS 13.15. See below, chap. 5, n. 225. 204. Briggs, Gorakbnath, pp. 116-17. This leads into another legend concerning 212. Watt, Dictionary, vol. 5, p. 233. The Jamnagar alchemical experiment was Dharamnath and Dhinodar hill, the site of an important Nath monastery to the described to me by Siddhinandan Misra: personal communication, Benares, north of Bhuj, the capital of Kacch: Dalpatram Pranjivan Khakhar, "History of the March 1985. Kānphatas of Kacch, Indian Antiquary 7 (1878), pp. 47-53. These accounts further 213. John Cort, "Twelve Chapters from The Guidebook to Various Pilgrimage resemble the myth of the fall of Krsna's capital, Dvaraka, which was swallowed up Places, the Vividhatirthakalpa of Jīnaprabhasūri," in Phyllis Granoff, ed., The Clever by the ocean shortly after the conclusion of the great Mababbarata war: Dvaraka is Adulteress and the Hungry Monk: A Treasury of Jain Literature (New York, Mosaic located across the Gulf of Kacch from Mandavi. The fact that the Kathiawar penin- Press, 1990), pp. 251-58; and Skanda Purāna 7.2.6.8. Girinārāyana, the Sanskrit sula and Kacch are both low-lying regions with swampy, shifting coastlines is no name for this site, reflects a link with the cult of Dattatreya as an incarnation of doubt also a source for accounts of "fallen" cities of Pattan. See the maps of Gujarat the god Visnu. In the YSA, the nine Näths are identified with nine Nārāyanas, and in Schwartzberg, Atlas, pp. 8-32, which indicate that the Kacch region has been led by Dattātreya. The names of the nine Narayanas are found in the Bhavisya periodically submerged by the ocean over the past millennia. Purāna (cited in Rajesh Dikshit, Śrī Navanath Caritra Sagar [Delhi: Dehati Pustak 205. Walter Ruben, Eisenschmiede und Dämonen in Indien, Internationales Archiv Bhandar, 1969], pp. 13-14); the nine Näths mentioned there are those currently für Ethnographie, 37, suppl. (Leiden: Brill, 1939), p. 205, citing Linga Purāna revered in Maharashtrau traditions. I.97.I-42. 214. Watters, Yuan-Chwang's Travels, vol. 2, pp. 248-49; Bānabhatta, Harsacarita 206. KPT 2.40: asya pūrvamevāyutam japtvā tato 'nena mantreņa pāsānam sap- 8.226. The YSA states that Matsyendra ended his career at Girnar: Mallik, SSP, p. tābhimantritam krtvā pattane vā grāme ksipet vā tena pāšānena vrksam tādayet/ 10. The early 19th c. Nätha Caritra (ASL MSS no. 1573/1645) of Mãn Singh, king grāma madhye aprārthitam sukhabhogam prāpnoti. Cf. KPT 1.33 (in Bhattacarya, of Marwar, devotes no fewer than fourteen chapters (1.2; 2.1-13) to Girnar; most ed, Indrajālavidyāsamgraba, pp. 267, 280). of these concern the travels of Gopicand to the site and the meetings he had with

412 413 Notes to Chapter Four Notes to Chapter Four

Dattätreya, Matsyendra, and other Nath Siddhas there. Man Singh was married to about halfway between Amba Bhavani and Abu, at about twelve miles' distance a Gujarati princess. See also Kaviraj, Bhāratīy, vol. I, p. 197, on the Aghori Bābā from both sites: Alexander Kinloch Forbes, Rās-māla: Hindu Annals of Western India, Kinnaram, who had a vision (darsan) of Dattatreya on Girnar in 1724. with Particular Reference to Gujarat (London: Richardson, 1878; reprint New Delhi: 215. Briggs, Gorakbnāth, pp. 110-19; Pandey, Gorakhnāth, pp. 63-64. Heritage Publishers, 1973), p. 210. 216. Personal communication, Kathmandu, October 1984. See below, chap. 11, 224. Tod, Annals and Antiquities, vol. 1, p. 259; and Sircar, Gubilas, pp. 12-13. A sec. 4, for the fruit of my searches. The geographical remoteness of the sites of number of legends connect the aboriginal Bhils of Rajasthan and Gujarat with the Girnar and Abu may also account for their perennial frequentation by Näth Sid- alchemical touchstone. The founder of the Mandalgarh branch of the Calukya race dhas and members of other tantric orders whose practices were viewed as abomi- raised the walls of his fortress (Mandalgarh) with the gold he made from a pāras nable by mainstream Hindu society. On the terrible practices of the Aghoris at patthar ("touchstone") found by a certain Mandoo, a Bhil in his service: Tod, Annals these two sites, see H. W. Barrow, "On Aghoris and Aghorapanthis," Anthropological and Antiquities, vol. 2, p. 545. A similar scenario involving a Bhil blacksmith's ap- Society of Bombay 3:4 (1893), pp. 199, 211-14, 246. prentice is found in the Katbāsaritsāgara (Ocean, vol. 3, p. 161 n. 1). 217. A number of major Nath sites are called "[Gorakh] Țilla," an alloform of 225. Tod, Annals and Antiquities, vol. I, pp. 255, 259, 265-70; and Sircar, Gubilas, "Țileti": the most important of these is found above Jhelum (Jhelum district, Pun- pp. 26-27. jab, Pakistan): Briggs, Gorakhnäth, pp. 101-3. Formerly, the Nath Siddhas had an 226. Here, Bappa's legend parallels that of Guha. Bappa's mother is also from important monastery called Gorakh Țilla in Benares: ibid., p. 84. the kingdom of Candravatī. 218. At Mount Abu, "Gopicand's cave" is located near the top of the central 227. Ekalinga Mabātmyam 19.35-68 (Premlata Sharma, Ekalingamābātmyam peak (above it is a temple to the goddess Camunda) of Acalgarh; in the pass be- (Ekling Mandir kā Sthalapurāņ evam Mewād ke Rāj-Vams kā Itibās [Delhi: Motilal tween this and the eastern peak is a shrine to Gorakh. At Girnar, it is Bhartrhari Banarsidass, 1976], pp. 85-88). This source does not provide the important account whose cave (an image of Gopicand is there as well) is located in a dip between the of spitting, which is given only in Tod (Annals and Antiquities, vol. I,pp. 264-65), first and second peaks and Gorakh whose dhüni is found atop the central peak. who also indicates that the initiatory thread with which Harita Rasi invested Bappā 219. See below, chap. 1o, nn. 125-30. Rāwal was called janeo, which is a specifically Nath Siddha term. 220. Cannabis, called vijaya "victory" in a number of tantric texts is, according 228. Gaurishankar H. Ojha, "A Gold Coin of Bappa Rawal," Journal of the Asiatic to the Tara Tantra (3.1-11), essential to ecstasy. The 7th c. south Indian Tirumūlar Society of Bengal, new series no. 6, vol. 23, Numismatic Supplement no. 40 (1926- sings the praises of marijuana in his Tirumantiram (Roșu, "Yoga et alchimie," p. 27), pp. 14-19. 370); and the kañja plant (gañja, marijuana) is called "Gorakh's root," korakkar-muli, 229. On Hārīta Rāsi's name, see Sircar, Gubila, p. 25. Rāśi was also a common in Sittar traditions (Kardaswamy, History, p. 354) suffix to names of Kalamukhas, an early sectarian evolute of the Pāsupatas: A. S. 221. Briggs, Gorakbnāth, p. 72. See above, n. 67. Altekar, "The Rashtrakutas," in G. Yazdani, ed., The Early History of the Deccan, 2 222. This and a number of other Rajasthani legends that follow were related vols. (London: Oxford University Press, 1960-61), vol. 2, pp. 705, 707. According to Colonel James Tod, author of Annals and Antiquities of Rajasthan, ed. with an to Singh's 19th c. Riport (vol. 2, p. 241), the forerunners of the Kanphata Jogis in introduction by William Crooke, 3 vols. (London: H. Milford, 1920; reprint Delhi: Rajasthan were called "Rāsi." Low Cost Publication, 1990). Tod (vol. 1, pp. 258-60) presents the following ac- 230. Briggs, Gorakbnath, p. 247; and Henri Stern, "Le temple d'Eklingji et le count as history; Sircar (Gubilas, pp. 12-13) demonstrates that it is a legend. Wink royaume de Mewar (Rajasthan) (rapport au divin, royauté et territoire: sources (Al-Hind, vol. I, p. 294) maintains that the Guhilots who would become the lords d'une maîtrise)," in L'Espace du Temple [Purusartha 10] (1986), p. 30 n. 19. of Mewar originated in Gujarat, noting the existence of a forest named Guhila in 231. Dvivedi, Nāth Sampradāy, pp. 174-75. that province. 232. Briggs, Gorakhnath, p. 66; idem., The Chamars (Oxford: Oxford University 223. The Acalgarh fort at Abu was the old stronghold of the Paramaras of Can- Press, 1920), p. 76; Rose, Glossary, vol. 2, pp. 407-8; NSC, pp. 191-92; Dvivedi, dravatī and Abu: B. N. Dhoundiyal, Rajasthan District Gazateers: Sirobi (Jaipur: Nath Sampraday, p. 179. The Pagalpanthis may be connected, through the persons Government Central Press, 1967), p. 430. The ruins of Candravati are situated of Cauranginäth-Pūran Bhagat and Rasālu, with an 11th c. king named Gaj, whose

414 415 Notes to Chapter Five Notes to Chapter Five

capital, originally called Gajapurī, had its name changed to Rāwalpindi (the "Rā- suggests that other data are available in the Rasaratnäkara (there are three distinct wal's body"?): ibid., p. 178. There is also an Indian jati called Rawal, which probably works by this title) and the Rasakaumudi (of which there are also three), as well as descends from these western Yogis: Briggs, Gorakbnāth, pp. 49, 53, 54- the Paradasambitā and other 20th c. works on rasa šāstra. 233. Gorakb Bānī, Pad 31. Cf. Briggs, Gorakhnātb, p. 66; Dvivedi, Nāth Sampra- 3. The names Ādinātha, Rudra, Visņu, and Brahmā figure in the Kulārnava Tan- dãy, p. 165; and NSC, p. 191-92, 226. There is a Nāgnāth temple located at Dwar- tra (6.63b-64) list of the twelve deities of the divyaugha. hat, north of Ranikhet, in the Almora district of western Uttar Pradesh: Kanti Pra- 4. Personal communication from N. Sethu Raghunathan, Madurai, January sad Nautiyal, The Archaeology of Kumaon (Varanasi: Chowkhambha, 1969), p. 219. 1985, referring to an appendix to the 14th c. Sarngadhara Sambita. I am more in- 234. NSC, pp. 191, 224; idem., ed. "Gorakb" Visesānk, p. 370; Rose, Glossary, clined to attribute these lacunae to the fact that much of what south Indian histori- vol. 2, p. 396; and Tucci, "Animadversiones Indicae," p. 137. See below, chap. 5, ography treats as historical fact (names of authors and texts, etc.) is in fact legend. nn. 15-16. So, for example, A. Shanmugavelan (Siddhar Science, p. 16) maintains that the Vedic 235. On Jvālamukhī-Bādava, the "seven-tongued flame," in the Mabābhārata sage Agastya (Agathiar) learned the Tamil language from the (Buddhist bodhi- (3.80.106), see Surinder Mohan Bhardwaj, Hindu Places of Pilgrimage in India sattva) Avalogithar in north India, before coming south to found the discipline of (Berkeley: University of California Press, 1973), pp. 44, 47-48. On the bādava of Sittar alchemy; the same Agastya would have taught medicine to the elder Vag- the subtle body, see below, chap. 8, sec. 2c. bhatta! 236. The Nath Siddhas associate the site with Gorakh and Nagarjuna: Kathleen 5. Sharma, Ayurved, p. 491; Ray, History, vol. 2, p. xcvi. M. Erndl, Victory to the Mother: The Hindu Goddess of Northwest India in Myth, Ritual, 6. Rāvaņa is mentioned by name in RM: Paris MSS no. 1222, fol. 27b, lines and Symbol (New York: Oxford University Press, 1993), pp. 44-48. 9-10. In the Rāmāyana, Candrasena (2) was the name of Ravana's father-in-law (i.e., 237. On the role of the Nagnathis on the Hinglaj pilgrimage, see Devadatt the father of Ravana's wife Mandodari), a king of Simhala-dvīpa (generally identi- Sastri, Agneyatīrtha Hinglāj (Bombay: Lokaloka Prakashan, 1978) pp. 4, 6-8. The fied with Sri Lanka): Singh, Nirgun Sabity, p. 63. A Sittar alchemical work names Rawal "caste" traditionally worships Hinglaj Mata, together with Siva Mahadev, Rävana's son Indrajit as an alchemist: see above, chap. 3, n. 34. Another of Rāvana's Bhairava, Gorakhnāth, and Macchandranāth: R. E. Enthoven, The Tribes and Castes sons, Meghanāda, is named as a Nāth Siddha in a ca. 1400 A.D. Telugu text, the of Bombay, 3 vols. (Bombay: Government Central Press, 1923), vol. 3, pp. 306-8. Navanāthacaritra of Gaurana: see above, chap. 4, n. 65. On the Muslim Jogis of the Sind who continue to go to Hinglaj on pilgrimage, see 7. RM, fol. 25a, lines 6-10. Cf. Sharma, Ayurved, pp. 185, 493. Mujtaba and Shah, "Taming," p. 83. 8. RM, fols. 24b.I, 26b.9. Cf. Cordier, "Récentes découvertes," p. 347. 238. Yogvānī 2:3 (March 1977), pp. 44-46, cited in Pandey, Gorakbnātb, p. 64. 9. Roşu, "Yoga et alchimie," p. 376; idem, "Mantra," p. 252. 239. Tod, Annals and Antiquities, vol. 2, p. 1196. 10. RC 15.35. Earlier than the Bhaskara who was Nāgarjuna's guru (see above, 240. I have summarized Tod's rendering of this account. The name of Tod's chap. 4, n. 181), another Bhäskara would have authored a commentary on the Su- "Rita" may be related to the Rītabiknäth "Yogi caste" of the Central Provinces who, śruta Sambitā, entitled the Suśruta Panjika, in the 12th century: Sharma, Yogaratna- Briggs reports (Gorakhnäth, p. 50), prepare and sell soap-nut (an ingredient used in māla, pp. 9-11; and idem., Ayurved, p. 230, 310-I1, 402-5. Somadeva also men- an alchemical experiment conducted in the 1940s: see below, chap. II, sec. I). tions Dineśvara (15.34), Dandī, and the sage Brahmajyoti, who are mentioned 241. See below, chap. Io, nn. 19-21. nowhere else. 11. Sharma, Ayurved, pp. 489-94. Naravāhana is a title of Kubera, the Hindu Chapter Five god of wealth; Naravāhanadatta is a hero of Somadeva's Katbāsaritsāgara who be- I. Marcel Detienne, L'invention de la mythologie (Paris: Gallimard, 1981), pp. comes the emperor of the Vidyādharas: Ocean, vol. 8, pp. 21-132, and vol. 9, p. 119. 131-43. 12. These works are the Rasasiddbisastra (no. 4313 in the Derge Tanjur) and the 2. The majority of these solitary citations are culled from a list found in Sharma Rasāyana Sāstroddhrti. See above, chap. 3, n. 121. (Ayurved, pp. 489-94), who inventories alchemical titles together with their alleged 13. See above, chap. 4, n. 235. authors, without any other descriptive material. He notes that he obtained this list 14. The Carpat Näth ji ke slok is found in Dvivedi, Näth Siddhom, p. 18. A frag- from Siddhinandan Misra, author of Ayurvediya Rasasastra and other works, and ment (5 folios, 72 verses) of a vernacular work entitled Carpat Rasayan is held by

416 417 Notes to Chapter Five Notes to Chapter Five

the MSL (Hindi MSS no. 669). His lost Svargavaidyakapalika is mentioned in the thored (by Virabhadra, Sukra, Indra, Vināyaka, etc.) alchemical works mentioned 13th c. RRA (1.1.17). A Carpati Siddbanta is mentioned in Priyanaranjan Ray, His- in no other source. tory, p. 128. The Nath Siddhas also attribute two other nonalchemical works-the 21. Misra, Āyurvedīya, pp. 39-40; Sharma, Ayurved, p. 490. Two procedures, Carpatamanjarī (of which a passage is quoted in Barthwal, Yog-Pravāb, p. 71) and called the "Kapalika method" and the "sequential Kāpāli method," are described in the Caturbbavabhivasanakrama-to Carpati: NSC, p. 199. Three Buddhist works the RA (6.84; 14.85; 16.34, 43) and later works as techniques for the coloration attributed to him-none of which are alchemical-are preserved in the Tanjur: or liquification of gemstones, mercury, and other minerals. A work entitled the Tucci, "Animadversiones Indicae," p. 137; Bagchi, K7#N, p. 28; Chattopadhyaya, Svargavaidya-kāpālika, attributed to Carpați by Nityanātha, author of the RRĀ, is ed., Taranātha's History, pp. 153-54. On Carpați and Vyādi, see Dowman, Masters, also a possible source for the name of the Rasa Siddha "Kāpālika." See above, n. 13. p. 382. 22. Sharma, Ayurved, p. 482. Cf. Siddhinandan Misra's foreword to Tripathi's 15. On the Buddhist Siddhacārya Carpati's alchemical legend, see above, chap. edition of the RA (Rasarnavam nama Rasatantram), p. 13. 3, n. 121. A Telegu myth involving "Sarpati" is found in the Bhojarājīyamu of 23. Iti īśvarasamvādam rasavādam sudurlabbam/ rasārņavam mabāšāstram ... ma- Ananta: Roşu, "Alchemy and Sacred Geography," pp. 154-55. On the Prān Sangalī, bāsiddhi yogānandena bhāsitam: ASL MSS 4256 (dated samvat 1732, i.e., A.D. 1675- see NSC, p. 194. 76) and erroneously catalogued under the title Rasarājaśankara, fol. 105a; ASL MSS 16. According to Saiva Siddhanta metaphysics, Srīkantha is the seventh of the 4273, fol. 109 (dated samvat 1684 i.e., A.D. 1627-28); and ASL MSS 4274, fol. 149. eight Vidyeśvaras, emanates or devolutes of Sadāśiva. 24. Bhairavānanda Yogin is the purported author of the Dhātukriyā or Dbātu- 17. Sharma, Ayurved, p. 697. Chapter colophons to the Anandakanda read: iti śrī mañjarī, a post-16th c. work: Satyaprakash, Prācin Bhärat, p. 625. However, see bhairavokte ānandakande. Dvivedi (Nāth Sampradāy, p. 190) names Manthānabhair- below, n. 53. ava as the author of the lost Rasaratna. A certain Ma[n]thana-simha, physician to 25. iyam hi poțalī proktā singhaņena mahībhrtā ... singhaņasya vinirdișțā the king of Malava, is the author of the ca. 14th c. Rasanaksatramālikā: Ray, History, bhairavānandayoginā lokanāthoktapoțalyă upacārā iha smrtāh. vol. 2, p. lx. 26. See above, chap. 4, nn. 180, 191. 18. Sanderson, "Saivism," pp. 671, 699. Manthāna (the "Churner") Bhairava, 27. The mantras and iconography of this alchemical pair are described in RA which is the name of a wild ascetic form of Siva whose cult belongs to the old Vidyā Pītha, later replaces the thirteenth and culminating Kāli in the krama of Kaula tra- 2.52-53, 62-72. 28. Like a number of alchemical works with -kalpa in their title, the Rasakalpa dition. The northern transmission also had the same tradition of four "Yuga Na- claims to be a portion of the Rudrayamala Tantra: Ray, History, vol. 2, p. lvii-lviii. thas" as did Abhinavagupta. It, however, called these the gurukrama, the succession See below, nn. 153, 160. of teachers: HTSL, pp. 76-77. 29. BbP I.I11b-112b. See below, n. 202. 19. For manuscripts of this work, see bibliography. This work is cast as a dia- 30. BhP 9.136b: nāthagoraksa samprokta. See below, n. 203. logue, in which the alchemical gnosis is revealed to the goddess Kākacandesvarī by 31. Svacchandabhairava is the deity of the Svacchanda Bhairava or Svaccbanda a form of Siva who is variously called Iśvara, Sadāsiva, or Bhairava. Another work, Tantra, the principal Tantra of the mantra pītha. Like Manthānabhairava, Svacchan- entitled the Kakacandisvara Kalpatantra, ed. Kailashpati Pandey, Kashi Sanskrit dabhairava has an alchemical preparation named after him: the RM (fol. 6b.4) and Granthamala, 73 (Benares: Chowhkambha Sanskrit Series Office, 1963) is also an RPS (8.132-33) describe a svacchandabbairava rasa. Elsewhere, one also finds min- alchemical work. This text, substantially shorter than the Kākacandesvarimata, may be identical to the Mabrasāyanavidhi, referred to by Shastri in his Catalogue (vol. eral preparations called ānandabbairava rasa, sannipātabbairava rasa, and tripura- bhairava rasa. I, p. lviii), 32. Letter from Alexis Sanderson dated 6 July 1992, citing Svacchandatantra 20. RU 1.9, 46-58 (which summarizes the chapter contents of the Mahodadbi). 2.88c-94b, and who also notes that the thirty-two-syllable mantra of Rasabhairava The Bhūtiprakarana (1.114b) states that the Mabodadhi (but not the Rasarajamabo- (RA 2.68) is a variant of the thirty-two-syllable agbora mantra that is the sakala dadhi) is a work in twelve thousand verses, while the Rasopanisat is a work in six form of Svacchandabhairava: (SvT 1.41-43). See below, chap. 6, nn. 59, 62. thousand verses. This work goes on (1.115-21) to list a number of divinely au- 33. Sharma, Ayurved, p. 493.

418 419 Notes to Chapter Five Notes to Chapter Five

34. This work is quoted extensively by Ksemarāja (fl. A.D. 1000-50) in his com- 44. Discussion in Misra, Ayurvedīya, p. 39. mentary on chapter nineteen of the Netra Tantra .: Dyczkowski, Canon, p. 42. A 45. The latter of these two works is ASL MSS no. 4267. On these texts and the palm-leaf manuscript of this work held in the NNA (MSS no. 3-392 [Saivatantra dating of their author to the reign of Bukka I of Vijayanagara, see Meulenbeld, 67]) is dated Nepal samvat 304 (= A.D. 1184). See Bagchi, KJiN (p. 2), and Gonda, History (draft of 1o May 1994), pp. 799-8o0, 943. History (p. 216), on Srikantha-guru as the great preceptor of the Pāsupatas. 46. ASL MSS no. 4260. For a historical discussion, see Misra, Ayurvedīya, p. 42. 35. Sharma, Ayurved, p. 492. Dvivedi (Nāth Sampradāy, p. 190) names Manthā- 47. ASL MSS no. 4222. For a historical discussion, see Sharma, Ayurved, p. 409. nabhairava as the author of this work. See below, nn. 170-73. 48. Ed. Jivaram Kalidas (Bombay: Venkatesvara Steam Press, 1967). 36. A single alchemical manuscript (no. 1.14.ii.19) entitled Ghoda Coli is held in 49. Ed. Manirama Sastri (Bikaner: Maniramasarma, 1934). the private collection of the Srī Ramacarana Prāya Vidyapīth in Jaipur; three one- 50. Ed. with notes by Yadavji Trikamji Acarya and Hindi commentary by In- folio manuscripts of the Ghodacoli Vakya (MSL MSS nos. 1481, 1482 [a], and 1483 dradeo Tripathi, Chaukhamba Ayurveda Granthamala, 10 (Benares: Chaukhamba [b]) are held in the Maharaja Mān Singh Library in Jodhpur. Ghoda Colī is listed Amarabharati Prakashan, 1984). For a discussion, see Sharma, Åyurved, pp. 336, as a Rasa Siddha in the AK and as a Mahasiddha in the HYP. 484. 37. Barthwal, Yog Pravab, pp. 68-69; Dvivedi, Nath Sampraday, p. 167. 51. ASL MSS no. 4348. 38. Kardaswamy, History, p. 354. See above, chap. 3, n. 54- 52. ASL MSS no. 4258. 39. See below, n. 235. 53. A portion of the Dbātukriyā is reproduced in Priyanaranjan Ray, History, pp. 40. Pandey, Gorakbnatb, p. 118. Another work on tantric sorcery, the Kāmaratna, 414-42: it appears to be the same text as the Dbātukalpa, attributed to Nāgārjuna is alternatively attributed to Nityanātha Siddha, "son of Pārvatī," or to rīnātha, in manuscript colophons, as are the Paradakalpa and Gandbakakalpa: see bibliogra- Siddhanātha (a name sometimes identified with Matsyendranāth), or Nāga Bhatta: phy and Ray, History, vol. 2, pp. lvii-lviii. HTSL, p. 122. 54. Sharma, Ayurved, p. 485. 41. There is confusion over this attribution. The Goraksa Siddbanta Samgraba 55. ASL MSS no. 4280. On the dating of this text, see Meulenbeld, History knows of two texts entitled SSP, of which one was authored by Goraksanātha and (draft of ro May 1994), p. 805. the other by Nityanātha. Nowotny (Goraksaśataka, pp. 23-24) indicates that Fitz- 56. Tantric sorcery is the subject of three classes of tantras, according to the Edward Hall found a copy of the SSP attributed to Nityanatha, but notes that Netra Tantra's classificatory scheme of the srotas: these are the Vama Tantras (on Rajendralal Mitra (The Yogic Apborisms of Patañjali [Calcutta: n.p., 1883, appendix, the gaining of siddhis), Garuda Tantras (on poisons), and Bhūta Tantras (on posses- "A Descriptive List of Works Extant on the Yoga System of Philosophy," S.S. 1181) sion and exorcism): HTSL, pp. 16-17. All such works are placed within the fourth, disagrees. Kaviraj (Bhāratīy, vol. 2, p. 275 n. 1) agrees with Hall; while Briggs (Gora- lower śrota of an alternative classificatory schema, found in the Brabmayāmala Tan- kbnatb, p. 255), citing Mitra, calls the author Nityananda Siddha. This last name tra: Dyczkowski, Canon, pp. 39, 42. corresponds to that of one of the Siddhas enumerated in chapter 49 of the Sankara- 57. NNA MSS nos. 3-392 (dated A.D. 1184) and 5-4947 (reel nos. B25/32 and vijaya of Änandagiri, which constitutes a Siddha charter of sorts. On this, see David A149/2); Kaiser Library (Kathmandu) MSS no. 297 (reel no. C30/16). Gordon White, "The Nath Siddhas," forthcoming in Constantina Rhodes-Bailly, 58. ASL MSS no. 4184. Douglas Renfrew Brooks, William K. Mahony, and Paul Muller-Ortega, eds., Sid- 59. ASL MSS no. 3967. dha Yoga: Continuity and Freedom in a Contemporary Spiritual Tradition, vol. 2 (nn. 60. The KPT has been edited and constitutes the last of the six works antholo- 29-32). gized in the Indrajālavidyāsamgraba, ed. Bhattacarya, pp. 264-390. The Yogarat- 42. Acārya Yasodbara krta Rasaprakāśasudbākara, with a Hindi translation by Sid- namāl, a much shorter work, has been edited with an excellent historical intro- dhinandan Misra, Jaikrishnadas Ayurveda Series, no. 54 (Benares: Chaukhambha duction by Priyavrata Sharma. Sharma notes parallel verses on p. 13 of his Orientalia, 1984). introduction. 43. Edited, together with a 12th c. work, the Lobasarvasvam of Sureśvara, by 61. The Dattätreya Tantra has been published as an independent work by Rajesh Yadavji Trikamji Acarya (Bombay: Nirnay Sagar Press, 1925). Dikshit (Delhi: Dehati Pustak Bhandar, n.d.). It is also included in Bhattacarya's

420 421 Notes to Chapter Five Notes to Chapter Five

edition of the Indrajālavidyāsamgraba (pp. 132-79). Compare 1.5a of this work (ud- titled the Datta Patala was in the possession of an old Delhi vaidya named Sriram dīse merutantre ca kālacandeśvare tathā) with RRĀ 5.1.7 and KPT 1.9ab (uddīse vātule Sharma Shastri. When I met him, in January of 1985, he was attempting to produce tantre ucchiste siddhisāvare kinkiņī merutantre ca kālacandesvare mate). alchemical gold by following its instructions. 62. The Matrkabbeda Tantra has been edited by Chintamani Bhattacharya, Cal- 73. The Rasadarpana is cited by the 16th c. Todarananda: Sharma, Ayurved, pp. cutta Sanskrit Series, 7 (Calcutta: Metropolitan, 1933, 1958); and translated by Mi- 486, 491, 493. chael Magee [Loknath Maharaj] (Delhi: Indological Book House, 1989). 74. The colophon to the NNA MSS no. 1-1473 of the KM states that it was 63. See above, chap. 4, n. 191. adyavatare; the same is said of a list of sixty-four yoginis contained in the Kubjikā- 64. Peter Peterson, ed., The Paddhati of Särngadbara, A Sanskrit Anthology (Bom- pūjā-paddbati (fol. 52): Shastri, Catalogue, vol. I, p. 58; and vol. 2, p. 80. Ādinātha bay: Central Book Depot, 1888), pp. 659-703. For a historical discussion, see is the author of the Mabakala Sambita: Pandey, Gorakbnath, p. 112. The Tāra Sharma, Ayurved, pp. 197-98; and K. R. Srikanta Murthy's introduction to his edi- Khanda of the Saktisangama Tantra is cast as a dialogue between the goddess Sakti tion and English translation of the Sarngadbara Sambita, which states (p. v) that the and (Śiva-)Ādinātha: Dvivedi, Nāth Sampradāy, p. 4 n. I. alchemical portions of the Paddhati were taken from a lost text entitled Yogara- 75. Ädinätha is invoked as a divinity in the opening (1.1) of the SSP of Gorakh- sāyana. nath, and is later (5.5-12) quoted at some length in the same work. 65. Kaviraj, Bharatīy, vol. 2, p. 292, note 1. On the role of the Yogasikhopanisad in 76. Five of these are edited together by Bagchi, in his Kaulajñananirnaya. They later 108 Upanișad traditions, see Christian Bouy, Les Natha-Yogin et les Upanisads, are: Kaulajñānanirnaya; two distinct works entitled Kulavira Tantra (A and B); Kulā- Publications de l'Institut de Civilisation Indienne, no. 62 (Paris: De Boccard, nanda Tantra; and Jñānakārika. A short vernacular work, entitled Matsyendranāthji 1994), pp. 112-13. ka pad, has been edited in Dvivedi, Nāth Siddhom, pp. 64-65; and in Mallik, SSP, 66. Mānasollāsa 1.2.10-51. p. 76. The Yoga Visaya, a Sanskrit work on hatha yoga attributed to Matsyendra, is 67. Pitambaradatta Barthwal, Gorakb Bānī (Allahabad: Hindi Sahitya Sammelan, edited ibid., pp. 45-47. A relatively early work, also attributed to Matsyendra, is 1942), who offers tentative Hindi translations. Another edition of this collection, the Candravalocana: it is quoted in the HYP (4.16, 54): Bouy, Natha-Yogin, p. 82 n. which includes a long and sometimes useful Hindi commentary, was published by 345. A very late hathayogic source, entitled the Matsyendra Sambita, is found in the Gorakhnath Mandir in Gorakhpur in 1979: Ramlal Srivastav, ed., "Gorakh manuscript form in the MSL and the NNA. Cf. Pandeya, introduction to Goraksa Bāni” Višesānk (Yog Vānī, special issue no. I for 1979) (Gorakhpur: Gorakhnath Sambitā, vol. I, p. gha. Mandir, 1979). 77. RA 1.4b: kulakaulamahākaulasiddhakaulādi śāsanam (here I emend nāśanam 68. Dvivedi, Nath Siddhom, introduction, pp. I-2, on the contents of Gorakb of Ray's and Tripathi's editions to sasanam, which is the reading found in the three Banī (cf. Barthwal's introduction to Gorakb Bāni, pp. 11-20), and pp. 2-4, 16, 21, ASL manuscripts of the work, and which makes more sense). RA 18.228a: tasmi- on the contents and dating of the works of the other Näth Siddhas. Cf. Vaudeville, nekārņave ghore nastasthāvarajangame. Cf. Matsyendra's Kulānanda Tantra (in Kabir Granthavali (Doba), p. v; Ronald Stuart McGregor, Hindi Literature from Its Bagchi, KJiN, p. 107, whose opening verse is repeated nearly verbatim by the Beginnings to the Nineteenth Century (Wiesbaden: Otto Harrassowitz, 1984), pp. 21- KCM [1.1]). 22, 132; and Dasgupta, Obscure, p. 373. 78. KJiN 16.47-49. Cf. Bagchi's interpretation, on p. 35. 69. Dvivedi, Nath Siddbom, p. 56 ("Bhartharī jī kī sabadī," v. 6). 79. Mrgendrāgama, Caryāpāda 1.36b-37, 40b-41a (Hélène Brunner-Lachaux, 70. Reproduced in Mallik, SSP, pp. 82-87. See above, chap. 3, nn. 3, 121. Mrgendrägama, Section des Rites et Section du Comportement avec la Vrtti de Bhattanārā- 71. One of the two is edited in Dvivedi, Nath Siddhom, p. 39. yanakantha, Publications de l'Institut Française d'Indologie, no. 69 [Pondicherry, 72. Chap. 13 of the Dattatreya Tantra is entitled "Rasayanam" and contains four Institut Français d'Indologie, 1985], pp. 364, 366). In a list of auspicious dream magical recipes that make use of mercury and sulfur, as well as other mineral, visions, the same source (Kriyapāda 8.14a-15b) names "seeing white mica ... the herbal, and animal (herpal) ingredients. Other chapters contain descriptions of Siddhas ... [and] coming upon the siddha-fluid." This last element is glossed by the "magical" uses of minerals, as for example chap. 8, which explains how to use kajjali commentator as rasāyana. See above, chap. 3, n. 133. on one's eyes to bring a woman under one's power (strīvasikarana is, moreover, the 80. Tarapada Mukherji, ed., Gopicandra Nāțaka (Calcutta: University of Cal- title of this chapter). See Sharma, Ayurved, p. 492. The manuscript of a work en- cutta, 1970), p. xliv; Bagchi, KJiN, p. 1. A ca. 14th c. palm-leaf manuscript in the

422 423 Notes to Chapter Five Notes to Chapter Five

Nepal National Archives, entitled Śrīkāmākbyāguhyasiddbi purports, in its colo- notes that this method of extraction is described in minute detail in chapter 3 of phons, to be have been "brought down by Srī Matsyendra-pāda": ibid., p. 6o. This the Jayadrathayamala (NNA MSS no. 5-1975, Saiva tantra 429), fol. 184, v. 5-8. It work is cited in the late-13th c. Kubjikānityābnikatilaka: Dyczkowski, Canon, pp. is unlikely, however, that this source is as old as Bhavabhūti. The process is also 163-64 (n. 23 to part 2). However, as Tucci notes ("Animadversiones," p. 134), described in Matsyendranath's KJAN 11.18. The RA (18.217-19) appears to de- Matsyendra may have been an initiatory name given to a number of persons. scribe a similar process, but on an external, alchemical register. See above, chap. 3, 81. See chapter colophons to the K7nN and KM, discussed below, chap. 8, sec. n. 133. 2b. Like Srisailam, which has been identified with at least three peaks, Candragiri 85. Sanderson, "Saivism," pp. 687-88. is also a ubiquitous mountain. An inscription from Karnataka locates a Candrapurī 86. This list is also found, with variations, in KJnN 5.25b. In 3.6-10, Matsy- in the western ocean (Saletore, "Kanaphata Jogis," pp. 20-21). Candragiri is also endra describes a set of nine cakras. These correspond precisely to standard enu- the name of a peak situated at the western end of the Kathmandu Valley. Candra- merations in later sources. dvīpa, Candragiri, Candrapurī, and the "western country" (paścima disi: Dvivedi, 87. KJiN 3.2-3 is adapted, nearly verbatim, by the HYP (4.14): Bagchi, KJN, Nath Siddbom, p. 9 = v. 105) are also mystic locations within the subtle body. p. vii. On matsyodara, see below, chap. 8, sec. 2a. 82. Cited in NSC, pp. 49, 217; Dvivedi, Nāth Sampradāy; and Pandey, Goraksa 88. The earliest extant manuscripts of both works are held in the Nepal Na- Sambitā, vol. 1, p. cha (who also notes that the Goraksasabasranāmastotra gives the tional Archives. The KM (NNA MSS no. I-285) is dated A.D. 1160; the KJiN southern toponym of Vadava as Gorakh's birthplace). On this, see also Bagchi, (NNA MSS no. 2-362 [H]) is dated to the mid-11th c. A.D. (Bagchi, KJâN, p. 5). KJiN, p. 64. 89. Abhinavagupta (apparently paraphrasing K7nN 22.8-9) speaks of his rever- 83. Such early Buddhist works as the Gubyasamaja Tantra (with the Sekoddesa ence for Macchandanath in TA 1.7: "May Macchandanātha be propitious to me, he and other commentaries) know of six-limbed (sadanga) hatha yoga, but only speak who tore apart the glowing net made of knots and holes, a batch of bits and pieces of the four specifically Buddhist cakras. In Buddhist systems, the Upanisadic tarka unfolding and spreading everywhere" (tr. in Silburn, Kundalinī, p. 121). Cf. TĂ ("analytical reasoning") is replaced with anusmrti, "recollection," and in Hindu ba- 29.32, and Jayaratha's commentary to TA 1.18; Sanderson, "Saivism," p. 681; and tha yoga with the āsanas ("postures"). One commentary on chapter 18 of the Gubya- Goudriaan, HTSL, p. 18. samāja Tantra, entitled "Sadanga Yoga," is the work of Nāgārjuna. It is extant in 90. He mentions the KM in his Parātrimsikāvivarana, p. 184 (cited in Dyczkow- Tibetan: Wayman, Yoga, pp. 163-73. The Mahasiddha authors of the pre-12th c. ski, Canon, p. 172, n. 79). See also ibid., pp. 48-49, 65, 84, and Schoterman, Satsā- Caryapadas were also well acquainted with a number of hathayogic techniques, but basrasambitā, p. 6. held to the older four-cakra system. See also Majjbimanikaya 12 and 36, as evidence 91. The TA 7.40 cites a work entitled Yoginikaula, on the subject of the distribu- that the Buddha himself was a teacher of batha yoga: Nowotny, Goraksaśataka, p. 14. tion of the śaktitattva into padas, mantras, aksaras, and cakras. A number of hathayogic techniques are vaguely described in Patañjali's Yoga Sütras 92. Dyczkowski, Canon, p. 65, n. 65. The KM also mentions the Siddha Kaula 2.46-50: Vyāsa, in his 5th c. commentary to the Yoga Sütras, lists a number of pos- (ibid., p. 171, n. 70) and the Siddha-märga (London, Wellcome MSS g501, fol. tures but gives no account of the cakras. The practice of diaphragmatic retention is 35.7-8). alluded to, by the term of stambha-vrtti, in YS 2.50. The ca. Ist . A.D. Mrcchakatikā 93. Sanderson, "Śaivism," p. 681; Dyczkowski, Canon, p. 163 (n. 23 to p. 62). mentions one yogic posture (1.I: palankāsana); none are mentioned in the Mabāb- The other three are Khaga (Bird), Tortoise (Kūrma), and Mesa (Ram). Abhinava- barata (E. W. Hopkins, "Yoga Techniques in the Epics," JAOS 22 [1901], pp. 345- gupta never employs the term amnaya in his writings; he seems however to assume 48, 369, 373). The MBb (5.42.33) compares the yogin who cultivates virya with "a it: ibid., p. 66. dog that eats its own vomit." 94. Dyczkowski, Canon, p. 81. 84. nityam nyastasadangacakranihitam hrtpadmamadhyoditam/ ... nādīnāmu- 95. The Western Transmission subscribed to this same mythic doctrine. On dayakrameņa jagataḥ pañcāmrtākarșaņādaprāptotpatana: Mālatī-Mādbava 5.2. On this, the sons of Macchanda (or Minanätha), and the ovalli system, see Dyczkowski, the importance of this passage for the history of hatha yoga, see Dvivedi, Nath Sam- Canon, pp. 62, 70, 163-64, 166, and 191 (nn. 23, 32, and 233 to part 2); and San- pradãy, pp. 82, 84; and Sanderson ("Purity and Power," p. 213 nn. 89, 108), who derson, "Śaivism," p. 681.

424 425 Notes to Chapter Five Notes to Chapter Five

96. For historical discussions, see Sanderson, "Saivism," p. 679; idem., "Trika 112. Important works include Harvey Alper, ed., Mantra (Albany: SUNY Press, Saivism," p. 15; Dyczkowski, Canon, pp. 65, 80; and HTSL, p. 50. See above, chap. 1989) (much of his volume is devoted to the methodological principle of treating 3, n. 132. tantric mantra as a "language game"); Brooks, Secret (mainly on the Śrīvidya tradi- 97. Sanderson, "Purity and Power," p. 214 n. 110, citing TA 13.301, 320-21b; tion); Muller-Ortega, Triadic Heart; and Padoux, Vac. On the parallel Vedic intel- and Mabāmnāyaprakāśa 1.30. lectualization of concrete sacrifice, see Frits Staal, Rules Without Meaning (New 98. Sanderson, "Trika Saivism," in Encyclopedia of Religion, vol. 13, p. 15. In spite York: Peter Lang, 1989), and Jan Heesterman, The Broken World of Sacrifice (Chi- of these reforms, and the broadening of the Trika base in 1oth-11th c. Kashmir, it cago: University of Chicago Press, 1993). was the dualist orthodox Saiva Siddhanta doctrines and the cult of Svacchanda- 113. Gopinath Kaviraj, Tantrik Sadhana aur Siddbant (Patna: Bihar Rashtra- bhairava that remained predominant in the valley: idem., "Kashmir Saivism," vol. bhasa Parisad, 1979), p. 392. Bindu-sādhana refers to practices involving "drops" 13, p. 16. (bindu) of sexual fluids. 99. On the collection and use of combined sexual emissions (called kundagolaka, 114. This is an abridgement of the Srimatottara Tantra, which is also closely etc.) in prereform ritual and worship, see KJiN 11.18, 32; 18.7-8; HTSL, p. 44; related to the Mantbanabbairava Tantra, a work that also refers to itself as Kādi text: Dyczkowski, Canon, p. 80; and Schoterman, Yonitantra, pp. 28-29, 32. see HTSL, p. 55, and below, nn. 202, 209. 100. See K7iN 18.6-23 on ritualized sex; 8.31-32 for an explicit discussion of 115. Elsewhere, Jalandhara has no alchemical works to his name, either in San- the use of eight types of vidya [goddesses]; 8.30-45 on means to gaining access to skrit or vernacular languages. His anthologized poems are on the subject of batha the sixty-four yoginis; and 11.1-43 on the offering and consumption of prohibited yoga. Like a number of other Nath Siddhas, Jalandhara appears to have been too foods and sexual fluids. See also 14.56 and 17.42. Cf. Sanderson, "Trika Saivism," busy living his yoga to have written about it. The rich body of legend surrounding Encyclopedia of Religion, vol. 13, p. 15 him contains a wide array of yogic and alchemical motifs. Jalandhara has seven 101. Dyczkowski, Canon, pp. 60, 62, 82. Buddhist works attributed to him in the Tanjur: Chattopadhyaya, Taranatha's His- 102. Sanderson, "Saivism," pp. 663, 679-80; idem, "Purity and Power," pp. tory, p. 417. He is also the author of a Sanskrit work, definitely from within the 191-92; idem, "Trika Saivism," p. 15. On historical clues for the geographical and Näth Siddha fold, the Siddhānta Väkya: Dvivedi, Nāth Sampraday, pp. 6-7. His ver- temporal parameters of these changes, see also Dyczkowski, Canon, pp. 34-35, 48 nacular poem, "Jalandhrī Pāv jī kī sabadī," is edited in idem., Nāth Siddhom, pp. 55, 86, 92, 189 n. 193. On the Vidyapītha as the most important subdivision within 30-31, and Mallik, SSP, pp. 90-91. the tantric classificatory system of the four pīthas, ibid., pp. 49-52. 116. This poem is translated and discussed below, chap. 9, sec. 8. Gorakh dispar- 103. TA 15.160-70b with Jayaratha's commentary, cited in Sanderson, "Man- ages alchemy and other tantric practices that afford siddhis inferior to those pro- dala," p. 176, n. 29. cured through yogic practice in Yogabija 174. 104. TA 29.127-32. This practice is prescribed for those whose goal is enjoy- 117. Rao and Reddy, "A Note on Goraksanatha and His Work Yogadīpikā," p. ment (bubhuksus) as a means to attaining siddhi; and for those who have yet to attain 37. See above, chap. 4, n. 172. self-realization (apraptavibodha). See below, chap. 1o, sec. 2. 118. The Gorakb Upanisad is edited in Mallik, SSP, pp. 72-75. 105. Dyczkowski, Canon, p. 80. 119. This work is often referred to simply as the Hatha Yoga. The critical edition 106. Jayaratha's commentary to TA 29.129a. I am grateful to Alexis Sanderson was established by Nowotny (Das Goraksaśataka) in 1976. for bringing this reference to my attention. 120. For a longer listing of these works, see Dvivedi, Näth Sampraday, pp. 98-99; 107. TA 3.261-64 and 5.151. Cf. idem., Parātrīsikā-laghuvrtti 11-16, with com- Briggs, Gorakhnäth, pp. 252-53; Pandey, Gorakbnāth, pp. 82-83; Nowotny, Gora- mentary, in Muller-Ortega, Triadic Heart, pp. 181, 189, and 195 (English transla- ksaśataka, pp. 23-24; and Gopinath Kaviraj's introduction to his edition of the Gor- tions) and 283, 286, and 288 (original Sanskrit). aksa Siddbānta Samgraba (Benares: Vidya Vilas Press, 1925), pp. ța-na. 108. TA 5.58a, translated by Muller-Ortega, Triadic Heart, p. 123. 121. Śārngadbara Paddhati 4372: dvidhā hathaḥ syādekastu goraksādis- 109. On this, see Silburn, Kundalinī, p. 174; and Dyczkowski, Canon, p. 82. usādhitah/ anyo mrkaņdaputrādyaih sādhitonisamudyataih. Dattātreya's exposition IIO. See below, chap. 7, n. 67. of yoga is found in Markandeya Purāna 39.1-65; 40.1-41; and 41.1-26. Cf. Mallik, 1II. See above, chap. 3, n. 129. SSP, p. 75.

426 427

Notes to Chapter Five Notes to Chapter Five

122. Pandey, Gorakbnātb, pp. 89-90. Elsewhere, a manuscript (MSL MSS no. 136. On the anthropic principle, see John D. Barrow and Frank J. Tipler, The 1557) of the Dattagoraksasamvāda claims to be a portion of the Tantra Mabārnava. Antbropic Cosmological Principle (New York: Oxford University Press, 1986). Cf. 123. The sole extant recension of the hathayogic Goraksa Sambita appears to be S. K. Ramachandra Rao, Śrī-Cakra (Delhi: Sri Satguru Publications, 1989), p. 9: a ca. 16th c. manuscript, written in Bengali characters, held in the Nepal National "The world is in reality a deliberate mechanism for the objectification of conscious- Archives, which was translated into Bengali by Prasannakumara Kaviratna in 1897. ness (prameya) while the individual is an equally deliberate mechanism for the sub- For a discussion of the relationships between the these works, see Tucci, "Anim- jectification of consciousness (pramāta)." adversiones Indicae," pp. 134-36; Bagchi, KJiN, p. 64; and Pandey, introduction 137. Summarized in Padoux, Vac, pp. 51, 79-83. to Goraksa Sambitā, vol. I, p. ța. Chapter 1 of the Goraksa Sambitā parallels chapters 138. In Trika Kaulism, an optical model is used to express the relationship be- 1-3 of the SSP; chapter 3 parallels SSP 4; and chapter 5 parallels chapter 6 of the tween Siva and Sakti: Siva is pure illuminating consciousness, prakāa; and the god- SSP, which describes the avadhūta. Cf. Goraksa Paddhati 1.4. dess reflective consciousness, vimarsa: Abhinavagupta, Parātrimšikā-vivāraņa, p. 90, 124. Meulenbeld (History, draft of 10 May 1994, p. 1011 n. 14010) attempts quoted in Muller-Ortega, Triadic Heart, p. 98. See also ibid., pp. 95-99. to make sense of this statement by dividing the work as follows: (1) Kādi-prakarana 139. Brooks, Secret, pp. 66, 88-89. (the Sākta work); (2) Bhūtiprakarana (the alchemical work); (3) Yoga-khanda; (4) 140. Brooks, Secret, p. 63; Rao, Śrī-Cakra, pp. 7-12. Nādījñānadīpikā; and (5) the Avadhūtagītā. 141. Brooks, Secret, pp. 49, 124; Rastogi, Krama, p. 38. 125. On alchemy, see Bhattacharya, Sādhanamālā, vol. 2, p. xxi. On hatba yoga, 142. Another instance of the fundamental cleavage between these two ap- see HT, p. 165. This source (p. 9) nevertheless includes alchemy and batha yoga in its proaches is to be found in the former's system of nine "alchemical" (rasa-sampra- list of the eighteen "constituents of tantra." The alchemical literature is discussed in dāya or kāpālika) samskāras, as opposed to the latter's eighteen "Ayurvedic" (bhagvān the chapter entitled "Tantras of Magic" in the survey of tantric literature produced śankara or saumya) samskāras: Misra, Ayurvedīya, pp. 214-15. by these same scholars: HTSL, pp. 117-26. 143. This work has been edited twice in this century, by Yadavji Trikamji Acarya 126. Brooks, Secret, pp. 52-53. (Lahore: Motilal Banarsidass, 1927), and by Daultarama Rasasastri, with an intro- 127. RRS 13.61; 16.7, 31. duction, commentary, and Hindi translation (Benares: Chaukhambha Orientalia, 128. The Rasarnava calls itself a "great Tantra" in 1.57. The full title of the 1989). Rasabrdaya is Rasabrdaya Tantra; the BhP calls itself the Bhuti Tantra (2.101), etc. 144. RM fol. 29b.10-30a.1; 30b.I. See below, n. 265. 129. The alchemical BbP, for example, calls itself the Hadibheda of the Goraksa 145. Vägbhatta II, author of the Rasaratnasamucchaya (1.33), copies this verse Sambitā. verbatim, as he does much of Govinda's first chapter. 130. A yantra is an apparatus which controls or subdues (yam) the volatile ele- 146. RHT 1.15-32. He mentions jīvanmukti in 1.33 ments combined in it (another use of yantra, as a synonym for mandala, diagram, 147. The eighteen samskäras are discussed in detail in chap. 9, sec. 2. has the same meanings, since by using it as a meditational support, the yogin is 148. The four elixir pills are: amarasundarī, mrtasañjīvanī, vajriņī, and kbecarī. able to concentrate and bring his mind under control). These and the results they produce are outlined in 19.61-76. 131. HT, pp. 78-89. 149. RHT 19.78-8o of Misra's edition. These concluding verses are not found 132. Brooks, Secret, p. 134; Gupta,"Pañcarātra," in Alper, Mantra, p. 235. in the NNA (MSS no. 3-118/ reel no. B165/1) manuscript of this work. On the 133. Sanderson, "Purity and Power," p. 192. basis of the word bhiksu in 19.80, Ray (History [1956], p. 149) concluded this was a 134. This description therefore excludes the important orthodox Saiva Sid- Buddhist work. dhänta, which was dualist, nonvibratory, nonbipolar, etc. 150. Sharma, Ayurved, p. 480; Misra, Ayurvedīya, pp. 34-35; and Romila Tha- 135. In opposition to Vedānta, which sees the phenomenal world as māyā, illu- par, "The Image of the Barbarian in Early India" Comparative Studies in Society and sion that veils ultimate reality: Navjivan Rastogi, The Krama Tantrism of Kasbmir History 13:3 (July 1971), p. 424. (Delhi: Motilal Banarsidass, 1979), vol. I, p. 39, citing the Mahārthamañjarī of 151. On the genealogies and history of the Chedi or Kalachuri kings, see Alex- Maheśvarānanda, p. 14. ander Cunningham, A Report of a Tour in the Central Provinces in 1873-74 and 1874-

428 429

Notes to Chapter Five Notes to Chapter Five

75, Archaeological Survey of India, vol. 9 (reprint ed. Benares: Indological Book atām kuryāt raso vāyuśca bhairavi. Cf. RM 1.8: baddhaḥ khecaratām padāmnayed House, 1966), pp. 77-113; R. D. Banerji, The Haibayas of Tripuri and Their Monu- dharate vyādhisamūho mūrchitah/ jaritabhavo janmabhavo vināśanaḥ kramito rañ- ments, Memoirs of the Archaeological Survey of India, no. 23 (Calcutta: Govern- jitaḥ bhuktimuktidaḥ. RA 1.19 is copied verbatim by HYP 4.27. See below,chap. 9, ment of India Central Publication Branch, 1931), pp. 1-116; and Babb, Divine Hi- nn. 54-55. erarchy, pp. 3-6. Wink (Al-Hind, vol. I, p. 287) surmises that they were driven out 162. RA: 14.40. "The Siddha [who uses] 'diamond-bound' mercury is quite in- of Malava in the 7th-8th centuries by the Pratīhāras. vincible to the gods and demons. He [becomes] the leader of the twenty-four Sidd- 152. Wink, Al-Hind, vol. I, pp. 258-59. Madana could not have been an has (caturvimśatisiddbānām nāyaka), a bestower of all the siddhis." The RM's list of important Kalachuri prince, however, since his name occurs nowhere in its royal twenty-seven Rasa Siddhas, which occurs in its commentary (tippana), may be bor-

inscriptions. rowed from the RRS: see below, n. 265. 153. A later work, entitled the Rasārnavakalpa (RAK) claims to be a portion of 163. See above, chap. 3, n. 133 and n. 84 to this chapter. the older Rudrayamala Tantra: HT, p. 11. This work has been edited and translated 164. A terse exposition of this theory and attendant practice is found in the into English by Mira Roy and B. V. Subbarayappa as Rasarnavakalpa (Manifold Pow- Jñānakārikā (2.2b=4) attributed to Matsyendranāth (in Bagchi, KJiN), p. 116. See ers of the Ocean of Rasa), (New Delhi: Indian National Science Academy, 1976). RAK also the metaphysics of Gorakhnäth, as presented in Dvivedi, Nath Sampraday, 78-207 is borrowed from RA 12.8-182; other portions of this text bear similarities pp. 102-13. to the KCKT; it has one verse in common with the RM: Meulenbeld, History (draft 165. See below, chap. 8, sec. 4.

of 10 May 1994), p. 873, note 12450; and p. 875. 166. RA 18.221-27. 154. In its two printed editions, RA 3.23a reads "dāmarākhyā mahāmantra dha- 167. RM fol. 27b.9-I0: pancāmrto mahāyogah hyukto manthānabhairave vā manīșu niyojayet." Certain manuscript versions (ASL MSS no. 4256, fol. 12a.10) rāksasarāvaņe vā punastatraiva bhāșitam. read dāmarākhyām mahātantram .... There exists a work on tantric sorcery, enti- 168. RM fol. 28a.3-8. tled the Uddamara Mabatantra, which I have not been able to date. A work entitled 169. Roșu, Liétard et Cordier, introduction, p. c. Dämara Tantra has been edited and translated by Ram Kumar Rai (Benares: Pra- 170. RM fol.7a.1-3 prescribes this rasa for ailments involving wind and bile. It

chya Prakashan, 1988). is composed of mica, orpiment, copper, iron, saurabbarāja, and mercury, as well as

155. On the dating of the Vimalaprabba, see above, chap. 3, n. 122. a number of botanicals. Manthanabhairavarasa is also described in the RPS (8.173- 156. Cf. the KCM, which singles out eight (of which six figure in the RA enu- 74) and the RRS (14.90-92), which uses more or less the same ingredients and meration) of the eighteen samskāras as astakarma, the "eight Works." See above, n. prescribes it for the treatment of coughs. See above, n. 31.

142, and below, n. 185. 171. Dyczkowski, Canon, pp. 97-98. Eagerly awaited is Mark Dyczkowski's criti- 157. Partial lists of the samskaras are found in RA 10.10 and 11.210-12. Ac- cal edition and translation of this monumental work, now under preparation. The cording to Siddhinandan Misra (Ayurvediya, p. 214), a number of other alchemical RC (6.1) cites the Manthanabbairava Mahagama as the source for its discussion of works follow the kapalika method for the purification of mercury. Other kāpālika the divine herbs (divyausadbi). See also above, n. 114. methods are mentioned in RA 6.43, 84; 14.85; 16.34; etc. See below, chap. 6, n. 16; 172. Kubjikā is mentioned in RA 3.8, 31, referring to a mudra and a mantra.

and above, n. 142. 173. This entire passage opens, in the RM (fol. 26b.10), with the statement: "I 158. Quite often, the sole difference between passages found in the two works will now tell ... what was done by Martandeya [or Markandeya]." The former would is in word order. Compare RHT 2.3 and RA 10.41; RHT 12.4 and RA 8.36; RHT refer either to a divine father of Yama (Martanda); the Vaisnava sage to whom was 17.3-5 and RA 17.78; and RHT 19.2, 9, 18a, 20 and RA 18.3, 8, 13a, 15ab. revealed an eponymous Purana; or to the Rja Martanda of King Bhoja (the narrati- 159. For a discussion, see Misra, Ayurvediya, pp. 19-20. vization of this operation casts Bhoja as the alchemist's royal assistant in both Al- 160. So too does the 12th c. RAK, which goes so far as to co-opt the RA's title beruni's and the Telugu author Ananta's versions: see above, chap. 3, n. 3). Evidence from the 12th-13th c. RC militates against an identification of Manthanabhairava even as it claims to be a portion of the Rudrayamala Tantra. 161. RA 1.19: mūrcchito harate vyādhim mrto jīvayati svayam/ baddhah khecar- with the Bhairava who revealed or authored the RA. This author evokes both a

430 431 Notes to Chapter Five Notes to Chapter Five

Bhairavācarya (whom he seems to identify as the author of the RA), and Mantha- bhūtas, vetālas, brabmarāksas, women of divine beauty, and the goddess Bhairavī. In nabhairava: RC 1.28, 6.1, 12.25, 16.55, 16.72. the RA, the emphasis is on flight and immortality. 174. RA 11.198-205, 209, 214, 218; 12.104-9, 348-51; 15.131; 14-3-20; 183. KCM fol. 6b.3-12b.7. 18.165-69, 174-77, 194, 201-3, and 211-21. The RM copies or paraphrases the RA 184. KCM fol. 12b.7-17a.5. in each of the four chapters (with tippana) contained in the two manuscripts I have 185. KCM fol. 17a.8-17b.3. See above, n. 156. consulted (Paris MSS no. 1222 and GAU MSS no. 862/34). The RA repeats itself 186. The KCM list of eight metals is quite idiosyncratic: tin, iron, gold (which in the following places: 18.165-70 = 18.201-3, 206-7, 218. it calls niska), lead, white copper (kāmsya), copper, copper or iron pyrites (māksika), 175. The KCKT (41.6) cites Nagarjuna and a work entitled Siddhayoga (KCKT and mercury. 8.7-15, referring either to Nāgārjuna's Āścaryayogamālā- Yogaratnamālā, or to 187. The uparasas are discussed in KCM fol. 20b.5-2 1a.9. Vrnda's Siddhiyoga). It is likely a 13th c. work, from western India. 188. The auto-commentary begins at fol. 20b.2 (cf. RM, Paris MSS no. 1222, 176. NNA MSS no. 5-3969 (reel no. A211/19). A partial edition (of selected fol. 30a.3). It is found in every one of the manuscripts of the KCM, held both in portions of the first six chapters) of the KCM is found in Priyanaranjan Ray, History, the ASL and NNA collections, that I have consulted: see below, n. 265. pp. 345-50. 189. KCM fol. 22b.4-23a.3. 177. Passages that are nearly identical are RA 6.34-38 and KCM 6.21-25, 29-30 190. KCM fol. 23a.3-4. On the varna-classification of mercury by color, see (numbering here is from Ray's partial edition, p. 350); RA 7.10 and KCM fol. below, chap. 7, n. 138. The RA (6.67b-68) classifies diamonds according to caste 19b.4-5; RA 10.10-12 and KCM fol. 17b.2-3; RA 12.280, 293b-294 and KCM fol. and gender. 13b.3-5; RA 12.311, 313-14 and KCM fol. 14a.3-4 and 14b.2-3; and RA 18.94-95 191. KCM fol. 24a.4-25a.1. Cf. RA 15.131-32, which is paraphrased by RM fol. and KCM fol. 5a.7-8. No manuscript of the KCM numbers its chapters or verses. 9b.7-8 and 17b.3. 178. Sanderson, "Saivism," pp. 682-85. An echo of these cults is found in the 192. KCM fol. 28b.8. KJnN 23.2-9. 193. KCM fol. 29b.3-7. A work entitled Sugrīva[va]-śamkaraņī-vidya (Spell 179. On the worship of Batukanātha, see KCM fol. 6a.5-6. The RA (2.119-21) Which Subjugates Sugrīva) is devoted to the exorcism of a host of evil spirits prescribes the worship of Batukeśvara, in the context of a rite intended to pacify through the intervention of this powerful figure: HTSL, p. 124. demons who might otherwise destroy one's Work. Mercury is referred to as rasa- 194. KCM fol. 22a.1. The numerous manuscripts of the KCM I have consulted bhairava in fol. 7b.I: the RA calls for the worship of Rasa (the "Mercurial") Bhair- are more or less identical, in spite of the geographical diversity of their provenance. ava. Sarvajña is likely a distortion of Sarvaja ("Arrow-Born"), a name of Rudra-Siva The sole dated manuscript I consulted is ASL MSS no. 3952, dated to A.D. 1679. used extensively in some manuscript versions of the BhP. The Rasasara (ASL MSS no. 4260) of Govindācārya (not the same as the RHT 180. The Kulananda Tantra passage is found in Bagchi, KJaN, p. 107; on the author) is a 14th-15th c. text. It is an alchemical tantra, whose content is similar to Lalitāvistara (v. 44 = KCM 1.1) see Snellgrove, Indo-Tibetan Buddhism, vol. I, p. that of the RA and the KCM. See above, n. 61. 302. 195. This work has been edited twice in this century: Rasopanisat, ed. Sambhas- 181. KCM 1.28, 29, 33 (numbering here is from Ray's partial edition, p. 348). iva Sastri (Trivandrum: Superintendent, Government Press, 1928); and Rasopanisat, Cf. BbP 1.1-12. The term dravya is employed in the RA to signify the sexual fluids 2 vols. ed. with commentary and Hindi translation by Badrinarayan Sarma (Ajmer: (2.121b) offered in tantric worship (siddhadravya), as well as all the "secret" fluids Krsna Gopal Ayurved Bhavan, 1959). (124b-125a), including mercury, used in alchemy. Dravya-guna is the term em- 196. A detailed description of a site in Kerala is found in RU 16.11-13. This ployed in modern Åyurveda for pharmacology. is one of a number of Keralan works on therapeutic alchemy. Others include the 182. The amarasundarī is described in RA 18.94-95, in KCM fol. 5a.7-5b.6, in Rasavaiśesika Sūtra of Bhadant Nāgārjuna, the Vaidyamanoramā, Dhārākalpa, and RM fol. 16a.5, and in RHT 19.65-66. The RA (18.174-75) also mentions a pill Sabasrayoga: Sharma, Åyurved, p. 697. The RU also details the location of another called vajrāngasundarī, the "Diamond-Limbed Beauty." The KCM follows its dis- south Indian site, named Rāsima. This source does, however, list sites in north cussion of these aphrodisiacs (fol. 3a.9-6a.1) with an account of the khecarī gutikā India: sites at which vaikrnta is found include Mewar, Swat, Nepal, Jammu, and (fol. 6a.2-6b.1), which mainly concerns secretly "sporting" in the sky with sakinīs, Kashmir: Meulenbeld, History (draft of 10 May 1994), p. 936.

432 433 Notes to Chapter Five Notes to Chapter Five

197. RU 1.9, 56; and in chapter colophons. likely the Tantra by the same name than a lost eponymous alchemical work. See 198. These are: the Prabbrtamata, Vatulamata, Vaisnavamata, Aindramata, Śank- above, n. 31. aramata, Brhaspatimata, and aukramata. 207. On umāyoni, see BhP 2.6; on śarvaja (changed to sarvajñā in a number of 199. RA 2.90; RU 1.90. On the ankusā mudrā as a stabilizing device to be used manuscripts), see ASL MSS no. 4401, chap. I, v. 10, and throughout. by the tantric sorcerer for the purposes of subjugation and attraction, see Vināsi- 208. BbP 1.77: varņisvaramahālaksmī kāmodayamahodayaiḥ/ khecarīcakrayo- khatantra 84-87 (ed. Teun Goudriaan [Delhi: Motilal Banarsidass, 1985], with geņa mahāmārīmakhena vā. Cf. 6.514. On the khecarīcakra, see below, chap. 8, n. Goudriaan's commentary, pp. 54, 56. 165 200. On the samskāras, RU 1.20-39; on kūrpa, RU 11.12. 209. BhP I.I5a: pravistāḥ sarvašāstresu siddhānte vāmadaksiņe. "Uttarottara 201. Bhüti here is a shortening of vibhūti, a synonym for the siddhis that are the Mātrtantra" is a possible reference to the Śrimatottara Tantra, of which the "Sakta" stated goal (along with bodily immortality) of this work: Pandeya's introduction to or "Kādi" division of the Goraksa Samhita is an abridgment: Pandeya, introduction Goraksa Sambitā, vol. 2, p. ka. to Goraksa Sambitā, vol. I, pp. ța-na. MSL Hindi MSS no. 649, entitled Goraksa 202. A more simplified colophon is found in the ASL manuscript (no. 4401) of Sambitā Bhāsa, calls the Goraksa Sambita a portion (khanda) of the Śrimatottara, in the work (entitled Svacchandasaktyavatara): "thus ends the Production of Supernat- its chapter colophons. ural Powers [which is contained within] the Svacchandasaktyavatāra ("the work 210. 4.108: kadalī vanajā bandhyā gudūcīkandajam rasam/ pāyasam caiva mīnā brought down to earth by Sakti") within the Satasabasr(yla Sambita ("the compen- nāma japāyasi sandhitam. The aghora mantra is evoked frequently, as is the dium of 100,000 verses"). The colophon to the BhP, i.e., the so-called Hadibheda, trailokya-damara (7.163) and a form of mantric recitation called svayambhu. See follows 9.141 of Pandeya's edition of the Goraksa Sambitā (vol. 2, p. 578). The colo- below, chap. 6, nn. 59, 62, 66. phon to ASL MSS 4401 occurs at the end of an additional tenth chapter. The 211. This evocation is found only in the closing verse of chapter nine of ASL colophon to the Kadi or "Sakta" Goraksa Sambitā resembles that of the Mantbāna- MSS no. 4401 (fol. 106b.4-5): ratnākarokti vikhyātam bhūtitantram mahodayam. bhairava Tantra, which also ascribes itself to the Kadi variety: HTSL, p. 55. 212. BhP 1.111-23. Cf. RU 1.14, 56. 203. BbP 9.135a, 136b, 137a: simhabhūpahitārthāya nāthena prakațīkrtah/ ... 213. Meulenbeld, History (draft of 10 May 1994), p. 941. See above, n. 202. nātha goraksaḥ samproktaḥ sarasam kālakūțavat pūrvam vijnāyate samyak sākșād 214. BbP 1.28; 5.178, 262; 6.222; 7.144, 152, 244; 8.63-65. amrtavadbhavet. These verses are missing from the conclusion of chapter nine in 215. BhP 1.82-89, 5.62. ASL MSS no. 4401 (entitled Svacchandasaktyavatāra). This manuscript, however, 216. BbP 1.5, 8, 11, 25 (in ASL MSS no. 4401). On the use of dravya in the contains a long tenth chapter, on tantric sorcery, whose opening verse (10.1) is KCM, see above, n. 181. Cf. RA 1.22, which states that mercury renders one's man- identical to RA 1.1. This chapter does not, however, belong to the rest of the work. tras effective. On alchemical uses of menstrual blood, see below, chap. 7, sec. 3. It is also missing from MSL MSS no. 1431. 217. The term khanepane is found in BhP 4-77; RM fol. 27a.9 (khanipana); and 204. Rasakāmadbenu, ed. by Yadavji Trikamji Acarya (Bombay, Nirnaysagar Dattātreya Tantra 9.13. Khānapāna is found in Akulavīra Tantra [B], v. 135b (in Bag- Press, 1925; reprint Benares: Chaukhambha Orientalia, 1988). chi, KJiN, p. 106). A general discussion of vernacular terms in the BbP is found in 205. BhP 1.78b-79: svacchandakaulave nātha varșāvidyāmukhena vā/ svāyam- Tripathi's foreword to Pandeya's edition of the Goraksa Sambita, vol. 2, pp. 7-8. bhukaulave nātha dīptādyena makhena tu/ ebhir etatkrameņaiva bhavajam pūjayet 218. BbP 7.114-19. The itinerant alchemist is also instructed to go outside of sadā. Cf. 8.98 and 9.129. Through the use of mercury, one becomes a second Svac- India in his search for divine herbs (7.125). chanda Siva (5.178; 6.480). Svacchanda (Bhairava) is to be worshipped together 219. BbP 4.118-32. See below, chap. 7, sec. 3; and chap. 8, sec. 5 on the RA with Sakti, "in accordance with the precepts mentioned in the Svacchanda" (7.108). descriptions. It is worthy of note that two of the qualities which the BhP assigns to The term kaulave is employed in the KJiN (11.7, 9; 16.24, 62) to connote an "ex- its laboratory assistants correspond to names of "mystery" Rasa Siddhas from the pert in the Kaula practices" (kaula-vit), by which the five tantric makāras are im- lists reviewed in the last chapter. Visārada ("expert") is the name of the fourth Rasa plied. Siddha in the early lists; while sūra ("Mighty") may lie at the root of the names 206. BhP I.111-12. The "Svacchanda" text mentioned in BhP 7.108 is more Śūrasena (no. 9), Šūrānanda, and Šūravida (no. 19).

434 435 Notes to Chapter Five Notes to Chapter Five

220. RC 6.1; 7.1; 11.88-90; 12.25; 15.32-35; 15.54; 16.72. It also mentions 235. Wujastyk, "Alchemical Ghost," p. 75. Wujastyk however errs in identifying Svacchandabhairava (14-52), a possible reference to the BbP. the RRA as an epitome of the KPT: this he does on the basis of chapter colophons 221. RC 2.1; 14.199. See also Siddhinandan Misra's introduction, pp. 12-13. to two of the manuscripts of the Siddha Khanda of the RRA held by the Wellcome The RC also mentions Srisailam (8.23), the Tapti River and Kanauj (10.129), the Institute, London: ibid., pp. 75, 82. Chapter colophons to the manuscript of the Vindhya mountains (14.91), and Saurashtra (14.174). The type of earth in question Siddha Khanda I consulted (NNA MSS no. 5-3092, Ayurveda 266) call it the work is potash alum, phitkari in Sanskrit. Whereas Mount Abu is presently located within of Nityanātha Siddha, son of Pārvatī." the state of Rajasthan, its cultural and political ties have historically been Gujarati. 236. RRÃ 3.11b-12a. See above, n. 10 to the preface. Only the 1987 Bombay edition of the Rasakhanda of the RRA contains a reference 237. RRS 6.1-60 = RRA 3.1.10-76. to the RC: Meulenbeld, History (draft of 1o May 1994), P. 954- 238. On the "laksanas" of the Tantras, see HTSL, p. 10. 222. RC 14-58; 16.60. 239. Listed are Benares, Kāmarūpa (Assam), Hardwar, Gaņdakī (a river in cen- 223. Curiously, he is called Yasodhana in the RRS [a] list. On the basis of the tral Nepal; nearby is the Saiva pilgrimage site of Muktīnāth), Vadarīka (Badarīnāth, RPS's use of the terms dvīpāntarotthā and rasakarpūra, Jan Meulenbeld (History, in the Garhwal Himalayas), and Gangāsāgara (in the Ganges delta, West Bengal). draft of 10 May 1994 p. 871) maintains that Yasodhara Bhatta was a 16th, and not The RHT of Govinda may also have been an eastern Indian work: see above, nn. a 13th c. author as previously supposed. His argument is, however, based upon a 151-52. later identification of the former preparation with cobacini, a plant of foreign origin 240. MBbT 1.5-16; 5.17-33; 8.30-34. See above, nn. 109-10, on the more or introduced into India in the 16th c. less synonymous sadbbāva, kulāmrta, and kaulāmrta. 224. RPS 8.32, 37; 13.1. Opium, imported to India from the west, is first men- 241. MBbT 3.7-8, 18-20; 14.13-14- tioned in the ca. A.D. 1175-1225 Gadanigraba of Sodhala. 242. MBbT 7.58-60; 8.18-24; 12.14-17. 225. RPS 13.15. See above, chap. 4, n. 211. 243. MBbT 7.36. 226. RPS 13.14. The Gaudiyas are so called because their lineage originated in 244. MBbT 1.7; 5-4; 7.2-5, 60; 8.18; 10.22; 14.21. Gauda, i.e., Bengal-Orissa. 245. André Padoux, "The Body in Tantric Ritual: The Case of the Mudras," 227. Meulenbeld, History (draft of 10 May 1994), p. 871. On the alchemical in Teun Goudriaan, ed., The Sanskrit Tradition and Tantrism (Leiden: Brill, 1990), usage of the term kavi, see above, chap. I, n. 49, and below, chap. 10, nn. 78-80. p. 66. 228. Rasarnava is the reading found in all of the manuscripts of the Rasakhanda 246. ASL MSS no. 4281 (of which Paris MSS no. 1222 is a copy, commissioned of the RRÅ I consulted in the Nepal National Archives (see bibliography); however by Palmyr Cordier, and containing a number of corrections on the Bikaner origi- Tripathi's edition of this work gives a reading of "rasāyane" for RRA 1.1.16a. nal); and GAU MSS no. 862. This last manuscript does not contain the commen- 229. The Svargavaidya-Kāpālika is not extant; however, Carpați figures as a Rasa tary found following chapter four in the Bikaner manuscript. The ASL manuscript, Siddha in alchemical legend and is the author of a number of extant alchemical from which the Paris manuscript was copied, continues for a few additional lines manuscripts: see above, chap. 3, n. 121. and ends with the colophon "thus ends the RM of Śrīman Nāgārjuna." All manu- 230. The RRA cites the RC in 1.10.51-61; and it names the KCM in 5.1.7. On scripts of the RM are four chapters in length: personal communication from Domi- the former citation, see above, n. 221. nik Wujastyk, May 1994. 231. Meulenbeld, History (draft of 10 May 1994), p. 955. 247. Paris MSS no. 1222, fol. 22a.6. Cf. GAU MSS no. 861, fol. 28a.9-10: vak- 232. Meulenbeld, History (draft of 1o May 1994), p. 954. sye sarvahitārtham kaksapuțam sarvasiddhikaram. 233. The RRA has never been edited in toto; each of its divisions has, however, 248. Meulenbeld, History (draft of ro May 1994), p. 986. been edited separately, although many are extremely difficult of access. See bibliog- 249. rasendramangalam śrutvā yaścikitsetcikitsakah/ tasya siddhir na sandeho raphy. svayam nāgārjuno 'bravīt. 234. Meulenbeld, History (draft of 10 May 1994), p. 947. According to this au- 250. RM fol. 25a.2-3: parvatāgrha prāsādā saśailavanakānanām kāñcanamayām thor, the Rasendrakhanda is largely in agreement with the Ayurvedic Madhavani- karisyāmi. On the Navanātha Carita, see Roșu, "Alchemy," p. 154, citing P. V. Para- dana in its etiology, symptomology, and treatment of diseases. brahma Sastry.

436 437 Notes to Chapter Five Notes to Chapter Six

251. See above, n. 60 of this chapter. The "Laghu Vivrti" commentary on the 270. See Radhakrishna Sastri's long Sanskrit introduction (pp. 1-14) to his 1952 YRM is dated A.D. 1239 edition of the AK; and Misra, Ayurvediya, p. 38. Three copies of the Tanjore manu- 252. RM fol. 1b.4; 6a.3. script are housed in the Gujarat Ayurved University's collections: GAU MSS nos. 253. RM fol. 1b.7-fol. 2a.1. Here too, it follows the RA (10.36-37). The RC 829, 830, and 831. (15.30) gives the same measures. 271. These parallel passages are presented in the form of a chart in Radha- 254. RM fol. 4a.1-4, borrowing from RA 6.81-84- krishna Sastri's introduction to the 1952 edition of the AK (p. 13). 255. RM fol. 6a.8: mrtāni lohāni rasi bhavanti. I am grateful to Damodar Joshi 272. See also AK 1.3.101-3, on the inner yogic production and consumption of for this interpretation of the term rasi, i.e., digestible in the same way as is rasa, the nectar of immortality. AK 1.20.22-34 also clearly borrows from RA 1.8-19. chyle, the first of the eight dhatus: personal communication, Benares, December 273. AK 1.3.9-56. Four other types of initiation are described. These are: 1984. sādbaka-, nirvāņa-, ācārya-, and siddba-dīksā (ĀK 1.3.57-109). 256. Names include ekalingeśvara-rasa; svacchandabbairava-rasa, manthāna- 274. Brahmananda, "Jyotsnā" commentary to HYP I.1. On Nārāyana, who cites bhairava-rasa, jvālamukbī-rasa, kumārī-rasa, varābī-rasa, vadavāmukba-rasa, etc. See this work in his 17th c. commentary on the Yoga Upanisads, see Bouy, Natha-Yogin, above, n. 31. pp. 68-74. I have not been able to locate an independent text entitled Mabākala- 257. See above, chap. 3, nn. 131, 146. Here, anjana is used in the sense of balm, yogašāstra. unguent, or ointment, rather than in the sense of a miraculous preparation (usually 275. All of these manuscripts are held in Jodhpur (MSL MSS 1468, 1469, 1470, with a black antimony-anjana-base) which affords invisibility and other super- and 1471, of which the last dates from samvat 1740 [A.D. 1683-84]) and Kathmandu natural powers. (reel nos. A-999/7; A-1289-9; C-86/617; E-1145/12; and M-23/10, of which the 258. Of the thirty-seven folios of the Paris manuscript of the RM, fols. 15a to first dates from samvat 1735, or A.D. 1678-79). 29b make up chapter four alone. 276. The Yogamartanda is reproduced in Mallik, SSP, pp. 56-71. On the Mahā- 259. RM fol. 16a.5. Cf. RA 18.95 and KCM fol. 3b.6-9. kala Sambita, see Pandey, Gorakbnath, p. 112. 260. RM fol. 17b.1-2. 277. Bouy, Natha-Yogin, p. 41. 261. RM fol. 19b.10-21a.10. Cf. KCM fol. 12a.3-12b.6. 278. This technique will be discussed at length in chap. 8, sec. 4. 262. RM fol. 21b.5-6. Cf. RA 11.70-73. 279. Bernier, Voyages, p. 245. 263. RM fol. 24b.6. 264. RM fol. 25b.3-4. Chapter Six 265. RM fol. 30a.3. 1. See K. R. Srikanta Murthy's introduction to his edition of the Sarngadbara 266. RM fol. 30b.6. See above, n. 188. Sambita (p. xiv). Cf. BbP 9.135, in which the author calls his royal patron "Simha"; 267. RRS 1.32-59 = RHT 1.3-33. and RHT 19.78, in which Govinda names his patron as "Madana ... Lord of the 268. Chaps. 2-5 and 7-8 of the RRS are mainly drawn from the RC (chaps. Kiratas." I have translated this chapter of the RA in its entirety: "The Ocean of 10-12, 14, 3-4, respectively); much of chap. 6 is from RRÃ 3.1.1-76. Chap. 9 is Mercury: An Eleventh Century Alchemical Text," in Donald Lopez, ed. Indian Reli- mainly inspired by the RA and RC; chap. 1o samples passages from chaps. 5-9 of gions in Practice (Princeton: Princeton University Press, 1995). the RC; chap. 11 draws on the RA and the RRA: Meulenbeld, History, draft of 10 2. Ray's and Tripathi's editions of the RA substitute nāsana ("destroying") for May 1994, pp. 979-80. Its two RRS Siddha lists (1.2-5 and 6.51-55) are borrowed śasaka in this half verse (1.4b), which makes little sense in the light of what follows. directly from the appendix to the RM and the RRA (3.166-71). Moreover, šāsaka (or sāsana) was the reading I found in the three manuscripts of 269. Meulenbeld, History (draft of 10 May 1994), p. 979 n. 13681. The Rasarat- the RA that I consulted while in India (ASL MSS nos. 4256, 4273, 4274). napradipika is ASL MSS no. 4222. Its author, Ramaraja, who also compiled the 3. KJnN 16.47-49. Cf. RC (1.6-7) which gives the "Kaula" etymology of pārada, Madanavinodanighantu (NNA MSS no. 4-2224, reel no. A1289/4), was a subject of the term for mercury. See above, chap. 5, sec. 5f. the Pala king Madanapala, who ruled in the latter part of the 14th c. On his dates, 4. An apparent play on words: whereas the first half verse (1oa) employs the see Sharma, Ayurved, p. 409. third person plural of the present indicative of the verb muc, "release, liberate"

438 439 Notes to Chapter Six Notes to Chapter Six

(muncanti, "they liberate"), this half verse (rob) employs the past passive participle 16. RA 6.84; 14.85; 16.34. (mukta), which must therefore be read in the sense "what ram or bull['s semen 17. RA 1.34; RRS 1.79. Rasa, while generally referring to mercury alone, can would?] not be released?" also mean, by extension, "essential compound," e.g., mercuric sulfide, cinnabar, 5. This is a reference to the use of the so-called pañcamrta, the "five nectars," mica, etc .: Roșu, "Yoga et alchimie," p. 363. which include the male and female sexual emissions, urine, excrement, and mar- 18. I am grateful to Alexis Sanderson for this insight, as well as for a great num- row: KJiN 11.11; and the Vimalaprabba commentary on Kālacakra Tantra 2.119, in ber of other comments contained in a very long letter (of 6 July 1992) in which he Fenner, Rasāyan Siddbi, p. 138. responded to an early draft of this chapter. While many of the details (mainly from 6. Elsewhere, the text describes as feeble-minded those unknowing persons the SuT) concerning tantric iconography and mantra found in this chapter come who, indulging in liquor and flesh and deluded by Siva's māya, prattle that "we have from that letter, I will, out of a concern for brevity, refrain from citing it repeatedly. gone to the world of Siva" (1.29) and sends to hell those persons who, knowing 19. This passage has a number of parallels in other tantric sources, among which mantra and Tantra, corrupt the rasa-yoga (1.45). These rhetorical verses are remi- may be counted a passage from Abhinavagupta's Parātrimsikāvivarana (p. 88, niscent of those found in the coeval Akulavira Tantra [B] (1.79-86), which mocks quoted [p. 276] and translated [p. 160] in Muller-Ortega, Triadic), in which that the trappings of yoga. author identifies the ether of the heart with the eternal absolute. Abhinavagupta is 7. RA 1.25. himself commenting on a passage from the Yogavasistha: Padoux, Vac, p. 81, n. 139. 8. RA 1.26. Cf. RC 1.8-10. Gorakhnāth refers to drinking the liquor of immor- Cf. Amanaska Yoga 1.23. tality in Gorakb Bānī Sabadī 137. 20. RA 1.8. "Sivahood" (sivatvam) in one's own body, a standard goal of tantrism, 9. Cf. RA 2.2, 8, in which the ideal alchemical guru is described as "one who is is also evoked in RA 1.31a; 12.337b; 17.25ab; 18.29b, 59ab. attached to the kula path" (kula-marga-rata) and the ideal pupil as "one who is 21. Although he does not call it so by name, he is singling out orthodox Saiva devoted to honoring the kula [divinities]" (kula-pūjā-rata). See also RA 18.101. Siddhänta here. See Sanderson's discussion of this school, which dominated Kash- 10. RA 18.48-55. The RA (1.27) refers to mercury itself as the "womb" of the miri religious theory and practice prior to the 9th century: "Saivism," pp. 690-93. alchemical lineage: for a discussion, see above, chap. 4, n. 111. 22. The term used is pindapata, literally, the "fall[ing away] of the body." 11. RA 2.121b; 18.115. Cf. RA 18.47-52, 132, 144. See above, chap. 5, nn. 23. RA 1.9, 13. Cf. RRS 1.54; RHT 1.28. Here, the author is denigrating the 109-10. Saiva Siddhānta view of liberation upon death in favor of the Kaula doctrine of 12. RA 2.17-27; 6.90; 15.92, 131; 18.169b. See below, chap. 7, sec. 3. Such an bodily immortality, which is itself inferior to the practical techniques of the alche- assistant woman is to view herself as the daughter of the goddess Candī (or Kālika) mists, whose methods and results are proven and tangible. and the god Śiva (Sadāśiva, Tripurāntaka): RA 2.31, 34-35. 24. Here I emend yathā dhyānam with samādhanam. 13. RA 2.124b-125a; 18.53-54- 25. RA 1.20-22. This notion, of a stabilizing the body (pindadhāranam: RA 14. HYP 3.48-49, with the commentary of Brahmananda. See below, chap. 8, 1.18a), or bodily stability ([pinda-]sthairya), the watchword of tantric and Siddha sec. 4, for an extended discussion of this practice. alchemy, appears to be approprated by Jayaratha, in his commentary on TA 29.127- 15. RA 1.18. A second, and more concrete interpretation, is one suggested to 32, in which the mouth-to-mouth exchange of sexual fluids practiced by adepts me by Professor Mukunda Raj Aryal of Tribhuvan University (personal commu- whose goals are siddhis: here, Jayaratha glosses siddbi with pindasthairyādi. nication, Kathmandu, June 1993). Referring to modern-day techniques of temple 26. RA 1.25. This descent is termed saktipat. But see the editor's comment (p. roof gilding, which he himself witnessed at Bhaktapur, one of the three royal capi- 6) on a variant reading, in the RC, which would make this a straightforward state- tals of the Kathamandu Valley, Aryal noted the use of mercury for the amalgam- ment on the reduction (pata) of volatile energy (sakti) in mercury. Cf. RC 10.54, ation of the gold used in such work. When the gilders are actually working with which states that the grace of Siva is necessary for the liquefaction of mica. the mercury-gold amalgam, they stuff their mouth's with chunks of raw [buffalo?] 27. RA 1.37-40; 18.193b. Cf. BhP 6.217; 7.108; 9.76. On tantric pūjā, see HT, meat; and they regularly wash their mouths out with alcohol. These would presum- pp. 121-62. See also below, sec. 3. ably be techniques for reducing, as much as possible, the likelihood of inhaling or 28. RA 1.59; RC 1.4, 2.21. Other clearly alchemical names of Siva mentioned in ingesting the mercury being used. other texts are Paradeśvara, "Lord of Mercury" (BbP 5.27); Raseśvara (mentioned

440 441 Notes to Chapter Six Notes to Chapter Six

together with Gauri: 5.289), etc. The RC (following 2.50) also singles out the five also mentions Svayambhu Kaulava, a common name of Siva, or of his phallic em- tantric (as opposed to the eight Puranic) Bhairavas, of which each corresponds to blem the linga in devotional Saivism. This term is formed by grafting to the term a syllable in the all-powerful aghora mantra. Nandikeśvara is singled out as a Saiva Kaula the ending -ve or -va: this is apparently a shortening of -vit, "knower, expert, alchemical divinity in a number of sources (e.g., RC 14.58); however, this may be a virtuoso," a term used extensively in the KJiN (7.30; 11.79; 14-49; 16.24, 62; 18.1; divinization of a Rasa Siddha named Nandi, who is cited throughout the alchemical 23.1). The BhP (7.108) also describes Svacchanda[-bhairava] in union with his con- literature, most especially by Somadeva, author of the RC (5.77; 11.89-90; 14.128; sort Sakti (śaktisamyuktab). 15.29, 35, 66; 16.64). Cf. BhP 9.134; RRS 1.5. 39. RA 2.8, 121a; BhP 6.229, 281, 297, 399, 431; Cf. KJiN 8.27-45. The BbP 29. RA 1.59; RC 1.5, 2.21. A male Rasānkuśa[-bhairava] is named as a Rasa Sid- (7.369) states that offerings to demonic female divinities (pisācīs, rāksasīs, vāma- dha in RRS 1.5. devīs), made with inauspicious practices, either Sväyambhuva, mantra-based, or 30. RA 1.37-44; RRS 1.23-28; 6.14-26; MBbT 8.1-10; RC 1.37-38. See also Šākta, never succeed. Treloar, "The Use of Mercury," especially pp. 237-38; and Siddhinandan Misra's 40. On the highly elaborate Trika cult of these goddesses, see Sanderson, "Sai- introduction to the RC (p. 5), in which he discusses the alchemical grounds for the vism," pp. 673-74, 681-82; and idem, "Mandala," pp. 188-89, 199. foundation and installation of the rasalinga. Cf. MBbT 7.58-69 and 12.14-18, with 41. RA 2.62. Cf. TA 15.331C-333b. In this role Malinī "encompasses" the three notes (pp. 38-39, 59) on the benefits of worshipping lingas made from the eight goddesses. More often, this role is played by either Kālasamkarșiņī or Mātrsad- metals, mercury, flowers of sulfur, etc. This construction is discussed extensively in bhava: Alexis Sanderson, "The Visualization of the Deities of the Trika," in André chap. 8, sec. 5. Padoux, ed., L'image divine: culte et meditation dans l'bindouisme (Paris: Editions du 31. See above, chap. 5, nn. 29-32, 205. CNRS, 1990), pp. 45-47. See below, n. 50. 32. RA 2.105-6, 126-30 Cf. RC 2.42; RRS 6.57. 42. RA 2.47-73, especially 2.51-63, 69-71. A nearly identical mandala is de- 33. RA 1.37-40, 43. Cf. RRS 6.19, in which the figure is raised to ten million. scribed in RC 2.23-35. This mandala is graphically reproduced in Roșu, "Mantra Other Saiva deities evoked in the RA (18.186) are Nandikesa and Kumāra. On the et Yantra" p. 252. Quite different is the mandala described in RRS 6.40-57, in which alchemical significance of the important Siva linga of Kedārnāth, see below, chap. it is the minerals themselves, as well as the twenty-eight founding Siddhas of the 8, n. 153. alchemical tradition, that are worshipped in sequence. 34. RA 1.48-51 43. RA 2.50-51, 72. 35. Roşu, "Renouveau," pp. 65-66; and Majumdar, "Medicine," in Bose, ed., 44. RA 2.52-53. The RRS (6.22) is more explicit than the RA on the meditative Concise History, p. 233. superimposition of these anthropomorphic images upon the rasalinga. 36. RA 2.110-18 describes an image of Rasa-Bhairava which the alchemist proj- 45. On this practice, of working from the outside in (laya-krama), see HT, p. ects upon on his own subtle body: this corresponds to similar configurations found 164. in a number of tantras (Padoux, Vac, pp. 120-21; Sanderson, "Mandala," pp. 186, 46. RA 2.56. Moving clockwise from the east, their names are Sakra (Indra); 190-92). "Masters of yoga" (yogindras) are to be worshipped with arghya, flesh, Skanda; Rudra; Pavana (Agni); Šiva; Pāvaka (Vāyu); Umā; and Vyāpaka ("the Em- fish, and yoginīmantras: RA 18.55. Cf. RRS 6.56. anator"). 37. RA 12.83; BbP 6.46. The five faces of Sadāsiva (mentioned in RA 2.110-11) 47. RA 2.57-58a. Their names are Lepika ("[Transmutation by] Smearing"); are supports for that great god, the supreme attributeless divinity of the Saiva Sid- Ksepikā ("[Transmutation by] Casting"); Kşārikā ("Corroding, Reducing [Metals] dhäntins. On this, see T. A. Gopinath Rao, Elements of Hindu Iconography, 2 vols. in to Ashes"); Rañjikā ("Tinting, Coloring" [of Metals]); Lohatī ("Metallica"); Ban- 4 tomes (Madras: Law Road Printing House, 1914-16), vol. 2, pp. 367-68, 374; dhakari ("Binder [of Mercury]"); Bhucara ("[The Power of] Moving over the appendix B, pp. 187-91. Earth" [at Will]"); Mrtyunasinī ("Destroyer of Death"); Vibhutī ("Superhuman 38. The BhP actually calls this divinity Svacchanda Siva (5.178), Svacchanda Powers"); and Khecarī ("Power of Flight"). In the Vidya Pitha, the dūtī is the con- Kaulava (8.98; 9.129: the Svacchanda of the Kaula virtuousi: kaulavit), "the slayer secrated consort with whom the tantric practitioner copulates in order to produce of Time" (kālaghna), or simply Svacchanda (7.108). The same text (8.98; 9.129) the mingled sexual fluids he offered to the gods: Sanderson, "Saivism," p. 671.

442 443 Notes to Chapter Six Notes to Chapter Six

48. RA 2.59. Their names are Māksika (copper pyrites); Vimala (iron pyrites); discussion of sakala and niskala mantras in nondualist Kashmir Saivism, see Padoux, Śaila (śilājatu, bitumen); Capala (selenium); Rasaka (calamine); Sasyaka (copper sul- "Body," pp. 72-73- fate); Gandhaka (sulfur); and Tāla (baratāla, orpiment). 60. Nityāsodašikārņava 144b, 145a; Devīrabasya, p. 100; Puraścaryārnava, p. 806. 49. RA 2.60-61. These are the Vidyeśvaras: see below, n. 56. 61. Sanderson, "Saivism in Kashmir," in Encyclopedia of Religion, vol. 13, pp. 50. RA 2.62. "In the lotus's pericarp [at the heart of the mandala], starting from 16-17. the east are the four saktis." Here, Aparā is described as vajrasakti and Parāpara as 62. RA 2.68: Om hrīm krīm raseśvarāya mahākālāya mahābalāyāghorabhaira- kanti: these epithets are taken up by Somadeva in RC 2.23-24, who names the four vāya vajravīra krodhakankāla kslauh kslah. The same mantra is found in SvT alchemical Goddesses (rasa-śaktis) as Mālinī, Vajriņī ("Diamond Maiden"), Kāntā 1.41-43. ("Magnetite Maiden"), and Abhrā ("Mica Maiden"). Parā and Aparā are also men- 63. RA 2.72a: pūjayedrasasiddhyārtham vidyayā pañcabījayā. Decoded, Rasān- tioned in this source. kusī's mantra is "aim hrīm srīm klīm sauh"-which is, like her iconography, mod- 51. RA 2.63a, 64b. eled after that of Balā Tripurasundarī (aim klīm sauh). It also resembles the five 52. RA 2.63b-64a, 65-66a. pranavas or seed mantras of the Kubjikā cult (aim hrīm śrīm phrem sauḥ). Cf. RA 53. RA 2.69-72. 2.84, where the five elements are identified with these same five seed-mantras of 54. The classic discussion of the tantric superenthronements is Sanderson, the rasānkusī mudrā; and in which the rasānkusa mantra is praised (2.89-90) and the "Mandala," pp. 178-82, 187. I am deeply indebted to this author for his detailed construction of a circular oblatory basin-homologized with the goddess's sexual commentary on the superenthronements found in the RA passage, which I repro- organ (yonicakra) and with which mantra-based worship is also to be performed- duce here. is described (2.77-85). 55. In RA 2.8b, the alchemical disciple is described as one devoted to the wor- 64. RA 3.5. ship of the circle of the yoginis and the kula. The presence of the eight mabārasas, 65. RA 3.19, which identifies the "eighty-eight thousand" herbal preparations the principal reagents with mercury in the alchemical samskaras, is troubling. Al- in which the alchemist will later macerate his mercury and other elements with the though these are the "inner circle" of mineral reagents vis-a-vis mercury, they are kula-kbecari, the "sky-going lineage." These are all brought together in the so'bam never divinized and do not seem to fit well into this mandala. bamsa utterance. 56. Rao, Elements, vol. 2, pt. 2, pp. 396-97, and Sanskrit appendix B, pp. 187-91. 66. RA 2.76; 18.6ob. Cf. RRS 6.38; RC 2.47. Their names are: Anantesa, Sūkşma, Sivottama, Ekanetra, Ekarudra, Trimūrti, Śri- 67. RA 3.16. kaņtha, and Śikhaņdi. 68. RA 3.6. 57. Svacchandabhairava rides on the shoulders of the preta Sadasiva in the SuT 69. RA 3.27. (2.81b); Sadāśiva is also the mabāpreta who symbolizes the Goddess's lotus seat in 70. RA 2.97-98. These mantras are used here, in conjunction with mantra-nyāsa, her tantric worship: HT, p. 144. Cf. TA 5.322a. Placed before the tantric image of for predicting the success of a future alchemical operation. Using a hexagonal man- Pancali Bhairava in Kathmandu is the recumbent image of a figure called Vetāla, dala upon whose corners are figured six forms of Bhairava and at whose center is whose name evokes the "vampires" of tantric sorcery who possess corpses (pretas figured Rasabhairava, the practitioner causes girls and boys whom he has brought in this context) and whose services are sought out by powerful tantric practitioners. together to serve as media for posing questions to the divinity: RA 2.92-103, dis- The iconography of this recent image appears to reproduce that of Svacchandab- cussed in Roşu, "Mantra et Yantra," pp. 254-55- hairava. On such vampires, see below, chap. 10, sec. I. 71. Padoux, Vac, p. 47; Gonda, Medieval, p. 171. The title of chapter three of 58. These passages, from book four of the SuT and chap. 22 of the Netra Tantra, the RA is "mantra-nyasa." are discussed in Padoux, Vāc, pp. 404-II. 72. Sanderson, "Mandala," pp. 174-75; Wheelock, "Vedic," in Alper, Mantra, 59. More specifically, Rasabhairava is identifiable with the sakala ("with attri- pp. 103-4; Muller-Ortega, Triadic Heart, pp. 162-81. In the RC (2.51), alchemical butes") form of Svacchandabhairava. See SvT 2.88b-94a, for Svacchandabhairava's mantras are to be imposed upon a patient as a means to curing him of disease: appearance; and 2.115a, which states that his consort is seated on his lap. For a Roşu, "Mantra et Yantra," p. 254.

444 445 Notes to Chapter Six Notes to Chapter Seven

73. On this, see below, chap. 9, nn. 40-47. 87. These include the four-syllabled bamso'bam, identified as the kulakbecari (RA 74. RA 3.2. Her anthropomorphic image, to be worshipped with the five Vidyā 3.19); rasānkušī (3.20b); triširas (3.22a); dāmara (3.23a), etc. bījas ("seed mantras") has already been discussed in 2.69-73. 88. In verse 38, the goddess asks Bhairava a question concerning diksā; verse 75. This is clarified, somewhat, in RA (3.4a): "In the same way as a linga's chas- 83b begins, "The practitioner who has been so initiated ... " Cf. RRS 6.27-31. ing (i.e., the yoni in which it is set) defines the linga's form, so that which 'linga's' - 89. Summarized in letter from Alexis Sanderson, dated 6 July 1992. i.e., māyā (lingimāyā: alternate reading lingitāyā, in ASL MSS no. 4256, fol. 90. Kulārņava Tantra 14.89. 11b.3)-is to that which is "linga'ed." (Or, "that which marks defines that which is 91. On this, see Gonda, "Dīksa," in Change and Continuity, pp. 439, 443. See marked"). The line of the vulva is called by the name "ether" in the fifth house: RA below, chap. 10, sec. 2. 3.4b. Cf. RA 2.114-16, for a list of explanatory synonyms for "ether" as a support 92. John Woodroffe, Principles of Tantra: The Tantratattva of Śrīyukta Śiva Candra for the inner visualization of a tantric deity. On the five houses as the five elements, Vidyārņava Bhattācārya, 2 vols. (Madras: Ganesh and Co., 1970), 2:477, cited in see Kaviraj, Bharatīy 1:555-56. On the five houses as the five components of the Walter, Alchemy and Medicine, p. 70 (n. 23 to p. 16): "In order that this unclean omkara, of which the fifth is the subtle nada, see Eck, Banaras, p. 115. body may be purified, it is dried up by means of the Vayumantra and burnt and 76. RA 3.5-7a. Here, I am following the reading of the Kashmir manuscript reduced to ashes by means of the Agnimantra." See below, chap. 9, sec. 3. used by Ray in his critical edition, as well as emendations suggested by Alexis 93. T.A. Gopinath Rao, Elements, 2:1, pp. 10-11, cited in Gonda, "Dīksā," in Sanderson (letter of 6 July 1992), who substitutes krom-the seed syllable which Change and Continuity, p. 430: "ancient and widespread ideas in connection with animates the mantra and which is known as the "goad"-for krim, which is 'initiation,' 'consecration,' or 'transmutation' have here been embedded in the typi- not a normal seed syllable. On krom as the goad (ankusa), see Nityasodašikārņava- cally Saivite pattern of Hinduism and are put into practice in the framework of rjuvimarsinī, p. 150, line 9. Bhairava's statement in 3.3a, that "four (houses) are Hindu ritual, requiring different kundas (receptacles for fire) and mandalas ('mys- placed in front, while bamsa is the fifth" would also refer to the fact that hamsa is tic drawings')." the final element of the kālapāśa mantra. 77. Cow dung and ashes are standard exoteric and esoteric purifying agents. Chapter Seven 78. RA 3.7-8. 1. RV 1.105.2 describes the exchange of a certain rasa between the partners in 79. RA 3.9-12. The kubjika mantra is mentioned again in RA 3.31a. sexual intercourse. 80. Schoterman, Satsābasra Sambitā, p. 14- 2. See, for example, RV 1.23.19; 9.86.39; Atharva Veda 1.4.4; 1.6.2-3; 3.31.6-11; 81. Ibid., p. 37. Śatapatba Brāhmana 1.6.4.5; 6.2.2.6; 7.3.1.3. See above, chap. I, nn. 40-45. 82. These four are identified with the four cardinal directions on the inner 3. See, for example, Satapatha Brāhmana 7.3.1.26. square of the important Kaula siddhacakra, which is graphically reproduced in 4. As, for example, the prānāgnibotra: Vaikbānasa Smārta Sūtra 2.18. Dyczkowski, Canon, p. 81, and which has been discussed above, chap. 5, nn. 77, 5. Maitri Upanisad 5.2. The term occurs in an emanatory schema, in which 93-94. tamas, the lowest guna, differentiates into rajas, which in turn differentiates into 83. SSS 1.2 with Schoterman's commentary (Satsābasra Sambitā, pp. 32, 36-37). sattva, which "when impelled, flowed forth as rasa." The KJiN (16.50) may be referring to the latter group in its mention of the four 6. Indeed, soma as oblation and agni as sacrificial fire are identified with sun kulasiddhas. and moon in Satapatha Bräbmana 1.6.3.23-25. An alternative reading of samarasa 84. Candakāpālinī is the supreme goddess of the yamalas, the "Union Tantras" is sāmarasya. of the Vidya Pitha; it is to her that the hathayogic Gheranda Sambita is revealed. 7. Goraksa Paddhati 1.71-75, which is quite nearly identical to Yogamārtanda 61- Kālī was worshipped as Caņdakāpālinī in mainstream Saiva funerary rites in medi- 64. The two texts are expansions on the 13th c. Goraksa Sataka. eval Kashmir: Sanderson, "Mandala," p. 201 n. 156. 8. Caraka Sambitā 6.7.71. 85. RA 3.13-16. 9. Dasgupta, History, vol. 2, pp. 357-59 86. The tripurabbairavi mantra is found in 3.25. My thanks to Alexis Sanderson Io. Nātyasāstra 6, cited in G. K. Bhat, Rasa Theory and Allied Problems (Baroda: for clarifying the language of this difficult passage. M. S. University of Baroda, 1984), p. 11.

446 447

Notes to Chapter Seven Notes to Chapter Seven

11. A yogic parallel are the seven fires of yoga (yogägnis), which render the body 24. On rasacikitsā, see Roșu, "Renouveau contemporain," pp. 65-66; and Ma- "cooked": Mārkandeya Purāna, cited in Kaviraj, Bhāratīy, vol. I, p. 178. Cf. Gorakb jumdar, "Medicine," in Bose, ed. Concise History, p. 233. Bani, Sabadī 156-57. A late south Indian alchemical source, the Yogimuni 1000 25. Caraka Sambitā 6.7.1. ([3.10-62], ed. and tr. S. Raghunathan [Madurai: Madurai University, 1982]) de- 26. Śatapatba Brābmana 12.7.1.7: 12.8.I.1, 15. scribes how the conjunction of the five elements with the guru brings about the 27. Śatapatha Brābmana 6.1.3.1-5. ripening (pac) of the elements and metals into gold. On the wide semantic range of 28. Satapatha Brahmana 2.1.1.5; 11.1.6.1-11. In this egg, the silver part of the the verb pac, see Charles Malamoud, "Cuire le monde," in Cuire le monde, pp. 35-71. shell becomes earth and the golden part sky: Chāndogya Upanisad 3.19.1-2. A Ti- 12. Cf. Caraka Sambitā 6.1.78-8o quoted below, chap. I0, n. 102. betan creation myth reproduces these themes: Asian Mythologies, s.v. "Cosmogonic 13. Misra, Ayurvedīya, pp. 517-80. Myths of Tibet." 14. Roşu, "Yoga et alchimie," pp. 376-78. See above, chap. 3, nn. 12, 30-32, 133. 29. Satapatba Brābmana 3.5.2.27; Aitareya Brābmaņa 7.4.6; Maitrāyanī Sambitā 15. RC 11.5. 2.2.2. 16. RA 1.36; RRS 1.76. 30. Needham, Science and Civilisation, vol. 5, pt. 4 (1980), p. 362. The notion of 17. The term rasa in the singular also refers to essential compounds, such as a "ferment" (Greek: matza) that could be used to effect a "biological" replication cinnabar, mica, pyrites, etc., in their native states: Roșu, "Yoga et alchimie," p. 363. of metals is found in the pre-common era Corpus Hermeticum, the Greco-Syriac Kauțilya's ca. Ist c. B.C. (Arthaśāstra 2.12.2-4) reference to rasa would therefore be Pseudo-Zosimus (ca. A.D. 300), the Persian Jabir ibn Hāyyan (9th c.) and the Euro- to smelted ore (and not to mercury, as some have argued). See also Subbarayappa, pean Geber (13th c.): ibid., p. 366. "Chemical Practices," in Bose, Concise History, p. 305. 31. Rāmāyana 1.36.10-22. Cf. Mabābbarata 13.84.68-70. Another such mineral 18. The mabarasas act directly on mercury; the uparasas catalyze the effects of myth is found in the early (pre-8th c. B.c.) Jaiminiya Brabmana (1.223). the mabārasas: Misra, Ayurvedīya, p. 342. Mercurial preparations in which plants 32. Yajur Veda 18.13, cited in Sharma, Ayurved, p. 466. The RC (14.5) states that are used are called rasausadha and rasabbasma rasāyana (RA 12.358). A number of the semen from which vabni-sambhüta ("fire-born") gold was produced was that of "magical waters" are alternatively referred to as rasas or udakas: RA 12.189-292; Siva, semen which Agni vomited after having swallowed it in the myth of the birth RRS 11.125; KCKT 8.18-39; Kardaswamy, History, p. 343; Subbarayappa, "Chemi- of Skanda. The RRS (5.6) agrees, calling such gold (vabnija-svarna). See below, n. cal Practices," in Bose, Concise History, pp. 336-37; Vimalaprabhā 202cd, in Fenner, 56, for identifications of gold with female sexual fluids. Rasāyan Siddbi, p. 153. 33. Dhyanabindu Upanisad 1.5, in The Yoga Upanisads with the Commentary of Śri 19. RC 1.3. Elsewhere (15.2), Somadeva states that the essence of all remedies Upanisad-Brabmayogin, ed. A. Mahadeva Sastri, Adyar Library Series, vol. 6 (Ma- is to be found in mercury (sūta). dras: Adyar Library and Research Center, 1968), p. 189. 20. Prior to the 12th c., this branch of Hindu alchemy was termed lobavāda, 34. As early as the Atbarva Veda (11.3.7-8), the metals are identified with con- "the doctrine of metals": personal communication from Damodar Joshi, Benares, stituents of the body, according to the brahmanic system of adbyātman-adhiyajña- September 1985. The modern Hindi term for alchemy, kīmiyā[gari), is a late bor- adbidaivata system of homologies. Here, flesh is identified with the "dark metals" rowing from the Persian: Ramcandra Varma, ed., Samksipt Hindi Sabd Sagar (Be- (lead and iron?), blood with copper, ashes with tin, and complexion with gold. See nares: Nagaripracarini Sabha, 1981), p. 206. above, chap. I, n. 46. 21. RA 17.165-66: yathā lohe tathā dehe kartavyaḥ sūtakaḥ sadā/ samānam kur- 35. This is the same god who, in Kaula traditions, throws the Kaula scriptures, ute devi pravisan dehalohayoh/ pūrvam lohe parīkseta tato dehe prayojayet. revealed by Siva, into the sea. Matsyendranāth, the "Lord of the Fishes-Nāth," re- 22. rasībhavanti lohāni dehā api susevanāt rasendrastena vikhyātaḥ: RRĀ 2.1.7, covers these, and later transmits them to humanity. See below, chap. 8, sec. 2b. cited in Misra, Ayurvedīya, p. 173, n. 4; Cf. RM, Paris MSS no. 1222, fol. 6a.8; and 36. Wendy Doniger O'Flaherty summarizes a number of such myths from the RA 7.151ab. Misra takes rasī to mean "oxidized" in the case of metals and "joined epics and Puranas, in Siva: The Erotic Ascetic (New York: Oxford University Press, with chyle and the other dhatus" in the case of the human body. 1981), pp. 105, 107-8. 23. RA 16.25b: dvitīya iva Śankaraḥ. 37. Tamils call copper sulfate nilakantha, identifying it with the blue of Siva's

448 449 Notes to Chapter Seven Notes to Chapter Seven

throat after he swallowed the poison that fell into the Ocean of Milk from the mouth of Vasuki, the serpent churning rope employed by the gods and antigods: bCud len gyi man ngag (On the Extraction of the Essence) of Bo dong. This differs

personal communication from N. Sethu Raghunathan, Madurai, January 1985. from the Hindu myth in that Siva is made to be a form taken by Vajrasattva to

38. RC 14-3-4; reproduced in RRS 5.4-5. The latter form of gold is undoubtedly unite with the Goddess; she scatters his seed about to make mercury: Encyclopedia

a reference to the myth of the golden embryo, the biranyagarbba, of Satapatha Brah- Tibetica, vol. 2, pp. 522-23, cited in Fenner, Rasāyan Siddbi, pp. 80-81.

mana II.1.6.I-II. 47. On the adulteration of the mercury in these wells, see RRS 1.80, and AK

39. Rasakāmadbenu 2.1.4, cited in Misra, Āyurvedīya, p. 496. 1.15. Other alchemical works containing this origin myth and references to the five

40. The Lobasarvasvam of Sri Suresvara (vv. 9, 72-74, 103-4, and 116), with Hindi wells include the Rasakautuka (ASL MSS no. 4203, fols. 2b.4-3a.5), Śivakalpadruma

commentary by Pavani Prasad Sharma, ed. Brahmashankar Mishra (Benares: Chow- of Sivanātha (ASL MSS no. 4349, fols. 2a.1-10), and Rasasindbu of Visnudeva (ASL

khamba, 1965); and the Rasendra Bhāskara (3.86-4.119; 5.6-26; 6.12) of Laksmi- MSS no. 4267, fols. ra.9-2a.2).

nārāyaņa (Jaipur: n.p., 1896). On the dating of the former text, see Meulenbeld, 48. On the location of the land of the Daradas see Wink, Al-Hind, p. 232; Ray,

History (draft of May 10, 1994), P. 775. History, vol. I, p. 43, n. I; D. C. Sircar, Studies in the Geography of Ancient and

41. Padmāvat 24.2 (= 245). A complete edition of this work, with a Hindi com- Medieval India, 2d revised ed. (Delhi: Motilal Banarsidass, 1977), p. 34 and n.

mentary, is Rajnath Sharma, Jayasī Granthāvali (Padmāvat- Țīkā Sabit) (Agra: 4; p. 68 and n. 9; and Sharma's notes to RRS, pp. 207-8. See also above, chap. 3,

Vinod Pustak Mandir, 1965, pp. 476-81). A partial edition, but with a superior n. 93.

commentary, is The Padumawati of Malik Mubammad Jaisi, edited with a [Hindi] 49. The word cinnabar is derived from the Greek kinnábari, which has the same

commentary, translation, and critical notes by George Grierson and Sudhakara (unknown) root as the Persian zanjifrab: Oxford English Dictionary, s.v. "cinnabar."

Dvivedi, Bibliotheca Indica, n.s. 877 (Calcutta: Baptist Mission Press, 1896). Par- Another Sanskrit term for cinnabar is hingula, which is identical to the Persian

tial English translations, based on Grierson and Dvivedi's edition, are A. G. Shir- term: Watt, Dictionary, 5:232. The nearest source of naturally-occurring ore cinna-

reff, Padmavatī of Malik Muhammad Jaisi, Bibliotheca Indica, 267 (Calcutta: Royal bar to the Indian subcontinent is Garmsir (Pir Kisri) in Afghanistan (Watt, Diction-

Asiatic Society of Bengal, 1944); and Lakshmi Dhar, Padumavati: A Linguistic Study ary, 5:233).

of the 16th Century Hindi (Avadbi) (London: Luzac & Co., Ltd., 1949). This passage 50. Whence the statements of RA 7.72 and 11.82: "Sulfur is, by its nature and

is found in Grierson and Dvivedi, Padumawāti, p. 530; Sharma, Jayasī Granthāvali, by its form, the form of mercury (rasarüpa)." "None of the rasas or uparasas is supe-

p. 386; and Shirreff, Padmāvati, p. 148. rior to sulfur." The former statement is a reference to the Indian distinction

42. Gorakh Bānī, Sabadī 118, attributed to Ratannāth or Ratan Pir, identifies between essence and existence as "name" (nāma) and "form" (rüpa). In fact, all the

iron with the Muslim holy man (pir), whose yogic teachings he identifies with cop- major iatrochemical preparations of mercury (kajjalī, rasasindhūra, rasaparpatī, and

per. Silver is equated with Muhammad and gold with God (khudai, sudai). makaradbvaja are essentially compounds of mercury and sulphur: Bhagwan Dash,

43. These include the Rasakautuka (ASL MSS no. 4203: fol. 2b.4-11; fol. Alchemy and Metallic Medicines in Ayurveda (New Delhi: Concept Publishing, 1986),

3a.1-5); the Śivakalpadruma of Śivanātha (ASL MSS no. 4349: fol. 2a.1-10); the p. 102.

15th c. Rasasindbu of Visnudeva (ASL MSS no. 4267: fol. 1a.9-11); and the 15th c. 51. RA 7.57-66; RRS 3.2-12; KCKT 44.2-3.

Rasasanketakālika of Camuņdakāyastha (1.2-4). 52. The problematic term in this passage is sambala, which the commentary to

44. Mercury is thus a remedy for täraka-roga, the disease of Taraka: Misra, recension ñ of the MBhT glosses as "that which is released monthly by one's wives."

Āyurvedīya, p. 170. This commentary is indicated by Michael Magee in his translation of the work:

45. The 11th-14th c. Skanda Purāna (1.29.87) seems to refer to this myth in its Sbri Matrika Bheda Tantra, pp. iii and n. I to p. 1.

account of the birth of Skanda. At one point in this myth, the gods all become 53. For the theory concerning submarine volcanos, see Sharma's commentary

pregnant because Agni, who has already swallowed the semen, is the mouth of the to RRS, chap. 11 (p. 210). On the doomsday fire and underwater fire jets or gas

gods. Then, "Siva's semen having torn open their bellies, an ugly mercurial [pārada] vents, see Kalyān Sakti Ank (Gorakhpur: Gita Press, 1934), p. 640. This image is

lake of that [semen], 100 yojanas [in expanse], was formed." transferred to the subtle body in K7nN 2.1-3; and the "Sakta" Goraksa Sambitā

46. A Buddhist version of this myth is found in a 14th c. Tibetan source, the 23.117-18. The RA (9.5, 8) describes, in succession, two types of saline reagents (bīda) called vadavāmukba and jvālāmukha. See below, chap. 8, sec. 2c.

450 451 Notes to Chapter Seven Notes to Chapter Seven

54. Shorter Oxford Economic Atlas of the World, prepared by the Economist Intelli- central India, a metal-working tribal people, for whom the soul is composed of gence Unit, 2d ed. (London: Oxford University Press, 1959), p. 88. Cf. Sharma's iron, see Ruben, Eisenschmiede, p. 88. commentary (p. 184) to chap. 11 of the RRS. Much of the sulfur in India comes 63. Madhihassan, Indian Alchemy, p. 47; Briggs, Gorakbnāth, p. 129. from sulfur springs which are also pitbas of the Goddess: ibid., pp. 211, 214-15. 64. Sharma's commentary to RRS, p. 208; and NSC, p. 295. The tribal Bhīls and 55. On the term lingam, see Watt, Dictionary, 5:232, and Bose, Concise History, Birhors continue the use of blood: Ruben, Eisenschmiede, p. 115. p. 317. 65. The Kalyān Sakti Ank (p. 648) reproduces an unspecified Puranic passage 56. Powdered cinnabar may also be used for the tracing of two-dimensional which states that wherever the body parts of the goddess Sati cut away by Visnu's mystic diagrams (yantras, mandalas). The preparation of gold and mercury used to discus fell, they took the lithic form (pasanatam) of the siddha pītbas. The most form the rasalinga is described in RRS 6.17-18. Here, in contradistinction to the complete discussion of the pithas of the Goddess, and the correspondences to Sati's the epic origin myth of the metal, gold is made to be the mineral homologue of the body parts, is D. C. Sircar, The Sākta Pitbas (Delhi: Motilal Banarsidass, 1973). The Goddess: see Misra, Ayurvediya, p. 83, quoting an uncited source: "The Goddess in Goddess is herself called a stone, or stone or mountain-born (sailaja, Pārvatī) and the form of gold, Sadäsiva in the form of mercury, the linga made through the identified with the lithic in a wide array of Hindu contexts: see Madeleine Biardeau, union of these is called the rasalinga." It should be recalled that two forms of natural L'Hindouisme, Anthropologie d'une civilisation (Paris: Flammarion, 1981), p. 143. gold are said, in the RRS (5.4-5) to arise from the rajoguna and from the placental 66. On the identification of the Kamakhya stone as red arsenic, see Karel R. van material of the egg of Brahmã, respectively: both of these origins evoke the fe- Kooij, Worsbip of the Goddess According to the Kalika Purana, pt. 1, A Translation with male element. an Introduction and Notes of Chapters 54-69 (Leiden: Brill, 1972), p. 26, citing 57. Subbarayappa, "Chemical Practices," in Bose, Concise History, p. 330. Kālikā Purāna 64.85-86 and Heinrich Zimmer, "Die Indische Weltmutter," in Er- 58. This constitutes the tenth of the eighteen branches of Sittar alchemy. The anos Jabrbuch (Ancona, Switzerland: Eranos, 1938), pp. 204-5. Dyczkowski ("Kuņ- practitioner holds three such gulikas in his hand and mouth and attached around dalinī," p. 7, note 13) maintains that the stone is black. the waist to aid in raising the kundalini, and transforming the subtle body: C. S. 67. The Yonitantra, ed. J. A. Schoterman (Delhi: Manohar, 1980), p. 24; Bharati, Narayanaswami Aiyar, "Ancient Indian Chemistry and Alchemy of the Chemico- Tantric Tradition, p. 259-60; Bhattacharya, History, p. 133; Fréderique Apfel- Philosophical Siddhänta System of the Indian Mystics," in Proceedings and Transac- Marglin, Wives of the God-King (New York: Oxford University Press, 1985), p. 240; tions of the Third All-India Oriental Conference, Madras, 22-24 Dec. 1924 (Madras: and Dyczkowski, "Kuņdalinī," p. 7 n. 13. Law Printing House, 1925), pp. 600-8. Purusottamananda Tīrtha, who runs an 68. Dvivedi, Nāth Sampradāy, p. 18; Kalyān Śakti Ank, p. 640; Briggs, Gorakb- Ayurvedic dispensary in Trichur (Tamil Nadu), carries such a gulikā on a belt he nātb, p. 18; Dikshit, Navanātba Caritra, p. 22. Gorakh created mountains of gold, wears around his waist: personal communication from Catherine Clémentin-Ojha, crystal, and ocher (geru) as a means to "awakening" his debauched guru Matsyen- Paris, February 1992. See below, chap. 9, n. 109 on the term gulikā/guțikā. dra, according to the Natha Caritra of Rāja Mān Singh of Jodhpur: NSC, p. 30. 59. MBbT 8.31-33. The term svapuspa, used here for "menstrual blood," is de- 69. Van Kooij, Worship, p. 26, citing Kālikā Purāna 64.73 and 76.89. The ad- fined in MBbT 5.31 as the first menstrual blood shed by a woman after her deflora- jacent mercurial cave is a creation of the 13th c. Simbāsanadvatrimsika (Franklin tion. The MBbT (5.27-33) describes the six types of women whose menstrual blood Edgerton, Vikrama's Adventures or the Thirty-two Tales of the Throne, 2 vols., Harvard may be employed in tantric ritual. Cf. RA 2.25. On ciñcini, see above, chap. 4, n. 41. Oriental Series, 26 (Cambridge, Mass .: Harvard University Press, 1926), 1:182-86; 6o. Needham, Science and Civilisation, vol. 5, pt. 4 (1980), p. 456. Yogiraja Vaidy- 2:166-67). Coloring (rañjana) through amalgamation is the greater part of trans- araja, "Yoga Research on Mercury and the Dhatus," Cakra, A Journal of Tantra and mutation in a number of alchemical contexts: see below, chap. 9, n. 23. Yoga [New Delhi] 4 (March 1972), p. 186. Cf. the "Sākta" Goraksa Sambitā 13.99. 70. On Hinglãj Devī, her geography, pilgrimage, history, and the present-day 61. S. Madhihassan, Indian Alchemy or Rasāyana (New Delhi: Vikas, 1979), pp. relationship of the Nath Siddhas to her cult, see Sastri, Agneya Tirth Hinglaj, 47-48. throughout; Briggs, Gorakbnäth, pp. 89 (on the use of thumrā), 103-10; and Encyclo- 62. Gold is to be placed in the tumuli (samadbis) of the Nath Siddhas: Briggs, pedia of Religion and Ethics (New York: Scribner's, 1914), s.v. "Hinglaj," by William Gorakbnäth, pp. 40, 42. On similar practices among the Munda-speaking Asurs of Crooke.

452 453 Notes to Chapter Seven Notes to Chapter Seven

71. A description of this site is found in a 16th c. Tibetan guide to Ca-ri, written 79. For a discussion of this tantric ritual, see Schoterman, Yonitantra, pp. 26-32. by Padma dkar-po. This is cited at length in R. A. Stein, Grottes-Matrices et Lieux Kunda[m]golakam is an elixir in KCM (ASL MSS no. 3952, fol. 32a.3-5); a kundagola Saints de la déesse en Asie orientale (Paris: Ecole Française d'Extrême Orient, 1988), produced by a Cāndālī or a Rāksasī is also mentioned in the RM (fols. 9b.7-8; pp. 37-43. 17b.3). Cf. MBbT 5.19-21. 72. RA 2.17-19, 25. The kakini is again described in 18.165-67, where she is 80. RA 18.47-49, 115-16, 144, 165-72. An alchemical teaching by the Tibetan called kāminī. She is called kālinī in RRS 6.34. Cf. BhP 4.129-32; and RM fols. Vimalamitra (the sixth of the eight teachings of the Bdud-rtsi-bam-po-brgyad-pa, 35b.3-5; 36a.1-2. called the Amrtakundali in Sanskrit) enjoins that "there be sexual union during the 73. RA 2.23; RRS 6.34. Kakiņī is the name of a female divinity whose worship time of her [the female consort's] menstrual flow ('dzag pa)-her great blood (khrag is described at length in chapters 58-62 of the Rudrayāmala Tantra. Here, she is chen) together with the yogin's semen will be known as the most excellent sort of one of six consorts of Siva [Śaktis], who are identified with the six cakras, all of rasāyana": Walter, Role of Alchemy, pp. 144-45, 168, 173. whose names are derived from Sanskrit phonemes: dākinī, rākinī, lākinī, kākinī, 81. For a historical discussion of the work and its author, as well as a summary šākinī and bākinī. In this context, kākiņī is identified with the anābata cakra, located of the plot, see McGregor, Hindi Literature, pp. 67-70. For full bibliographic de- in the heart (15.54-56) tails, see above, n. 41. 74. RA 2.20-24; RRS 6.32-33. 82. Biravā lonā is an imaginary transmuting herb; however Lona (from the San- 75. RRS 6.35; RA 14.2. A preparation of zinc carbonate, mercury, and menstrual skrit lavana, "salty, beautiful") is also the name of a powerful witch in north India: blood transmutes lead into gold: RA 7.31. In fact, the percentage of sulfur in a see below, chap. 9, n. 163. Haratāra (Sanskrit haratāla), yellow arsenic or orpiment, woman's blood is the highest at the time of her menses: personal communication has a brilliant orange-yellow color which, like red arsenic or realgar (manabsila) from Siddhinandan Misra, March 1985. Sulfur, in addition to increasing the po- lends itself to an identification with female passion and, more specifically, men- tency of mercury, is said in the RRS (3.17, 45) to increase the potency of male strual blood. Mercury "swallows" sulfur in alchemical reactions; the question being semen. asked here is why Ratansen does not allow himself to enjoy the sweetly perfumed 76. KCM fols. 12a.8-9; 2 1a.6-7. A procedure similar to the first of these is also body of Padmavati: see the origin myth of sulfur (gandhaka), which, when churned prescribed in the eponymous KCKT. Here (8.68-70), an alchemical charm for in- out of the ocean of milk was declared "aromatic": above, n. 51. vulnerability in battle is produced by placing equal amounts of gold, silver, and 83. Padmāvat [310-12 = 27.3-5] in Sharma, Jayasī Grantbāvali, pp. 476-81; and copper in menstrual blood, wrapping the whole with a lotus filament, sprinkling it Dhar, Padumavati, p. 56. Cf. a description of such color changes in the laboratory, with cold water, and placing it in the yantra stbana. A nearly identical preparation, in Dash, Alchemy, pp. 141-42. As McGregor (Hindi Literature, p. 70) notes, Padma- called a "tigress pill" (vyaghrī-gutika), is described in the RA (12.348-49) and the vatī is equated with Gorakhnath slightly later in this text (27.14). Thus, the con- RM (fol. 16b.4-5). These latter two sources instruct the alchemist to place this summation of love is also an allegory on a yogin's transformation through initi- preparation inside a woman's vulva, her gubya-sthāna (rather than in a yantra- ation. sthāna). 84. RA 1.28. The same source [11.104-7] describes the immediate and long- 77. BhP 2.5b-6a; 4.7; 5.1. Cf. RHT 3.28b on mercury that is bound when it term effects of eating said mercury. enters the yoni (yāvadvisati na yonau tāvadbandham kuto bhajate). 85. RHT 1.33: divyātanurvidheyā haragaurīsrstisamyogāt. 78. BbP 3.29b-3oa: purandhre gaurirajasā tadbījena bhavantu tat/ jalajādyamla- 86. Vimalaprabbā 2.119, translated in Fenner, Rasāyan Siddbi, p. 138. Menstrual samyuktam bandhanam yāti sūtakah. Later, the same source prescribes a prepara- blood both transmutes copper into gold and, when eaten, affords immortality, ac- tion containing menstrual blood for increasing the powers of mercury to absorb cording to MBbT 5.27-43, especially 37-42. See also Schoterman, Yonitantra, p. 29. (grāsa: literally, "swallow, consume") mica (5.284-87); and for "cutting the wings" 87. MBbT 5.28. A number of tantric sources specifically recommend a nonvir- (paksaccheda) of volatile mercury (6.15-17). Here, one is to dry, on a piece of blue gin, ideally a low-caste woman (especially a Cāndālī or Dombī), for such practices: cloth, a woman's menstrual blood. Through this cloth, one is to strain human Schoterman, Yonitantra, pp. 18-21, 24-25, 31-32; and Bhattacharya, History, breast milk, the which is then macerated together with male semen. p. 136.

454 455 Notes to Chapter Seven Notes to Chapter Seven

88. Sharma, note to RRS, p. 121. of -alaya ("abode, place of"), in which case this term would mean the "seal of the 89. BbP 7.114-15. Such herbs are to be gathered with devotion to Siva, and with place of the penis." Another possible source of the -oli ending is -ovalli, which the bijas (seeds, but also seed mantras) of Sakti, (pronounced) one by one. means, in a tantric context, a "current of doctrine": Dyczkowski, Canon, p. 81 and 90. RC 6.21. On the role of the outcaste Candali or Dombī woman in the wor- n. 233. ship of the female sexual organ, and the subsequent production of the power sub- 99. Needham, Science and Civilisation, vol. 5, part 5 (1983), pp. 270-74, who stance, the yonitattva, through sexual intercourse with her, see Schoterman, Yoni also discusses the Taoist cognate to this: buan ching, "making the Yellow River flow Tantra, pp. 24-25. Candali is also a term for a woman in the first day of her menses backwards." According to Dharmavir Bharati (Siddh Sabity, p. 325), the Näths bor- (ibid., p. 25), and for the female energy located in the abdomen (Kaviraj, Bbaratiy, rowed their theory and practice of the vajroli mudra from the Mahayana Buddhists. 2:262-63. A 15th c. Bengali commentary to the Buddhist Laghu Kalacakra Tantra The practice is referred to in Gorakb Bānī Sabadī 141: bajarī kartā amarī rāșai amarī states that that part of the female organ which resembles the beak of a bird is called karamtām bāī/bhog karamtām je vyamd rāșai te goraș ka gurubhai. Cf. Sabadī 142, "bird mouth." The canal which carries semen down from the bird mouth is called below, chap. 8, n. 88. Jerzy Kosinski (The Hermit of 69th Street [New York: Zebra candali when, instead of semen, it carries menstrual flow. Dombi is that canal which, Books, 1991] p. 69 n. 111) refers to this practice as "acclivity," a term not found in during the menses, points upward. At the time of the discharge of semen, the same the Oxford English Dictionary. Dombī becomes Avadbūtī: Shastri, Catalogue, vol. 2, pp. iv-v. 100. bCud len gyi man ngag of Bo dong, in Encyclopedia Tibetica, vol. 2, pp. 522- 91. Sachau, Alberuni's India, vol. 1, pp. 189-92. See below, chap. 9, n. 178. 23, 556, cited in Fenner, Rasayan Siddbi, p. 82. 92. HYP 3.83-91; 99-103; Yoga Šāstra of Dattātreya 299-312; Goraksa Paddbati, 101. At least two secret Tibetan Buddhist tantric initiations involve these same appendix to 1.82, pp. 50-55. Alternatively, the practitioner may use his partner's elements: sexual intercourse (which is literal and not figurative here), the intermin- yonitattva for purposes of external ritual and worship: Schoterman, Yonitantra, gling of sexual fluids (the red drop of the Prajña-maiden's pudenda (padma) and the p. 28. white drop of the master's vajra), the internal raising of these commingled fluids 93. On these terms, see above, chap. 5, n. 109, and n. 67 of the present chapter. (described as nectar), and the bliss of the realization of Buddha consciousness. The The term sadbbava is found in KJiN 21.10. texts of these initiations are translated, with commentary, by David Snellgrove, 94. TA 29.126, 129, translated in Silburn, Kundalinī, pp. 190-91. These multiple Indo-Tibetan Buddbism: Indian Buddhists and Their Tibetan Successors, 2 vols. (Boston: roles of the female consort and her sexual essence are discussed in Silburn, Kun- Shambhala, 1987), vol. I, pp. 256-64. dalinī, pp. 189-92. Italics are my own. 102. This term is discussed in SSP 5.7; Akulavīra Tantra [A] 43; Goraksa Paddbati 95. Siva Sambita 4.1-5; HYP 3.100, 102. 1.75; Kulānanda Tantra 4, 58; and a host of other hathayogic and tantric sources. 96. Silburn, Kundalinī, p. 192 (commentary to TA 29.128-29): "The process of For discussions, see also Srivastav, "Gorakh Bāni" Viśesānk, pp. 15, 238; idem, "Gor- breath and efficience going from the yogini's mouth to the siddha and vice-versa is aksa" Viśesānk, pp. 45-46, 156-57; Dvivedi, Nāth Sampradāy, pp. 58-59, 72; Gonda, precisely sampuțīkarana ... Prāna and vīrya are so intimately mixed as to be trans- Medieval, p. 224; and Mallik, SSP, p. 21. formed into each other and to become one and the same. According to the tra- 103. The red bindu is said to shine with the same ruddy brilliance as liquid sin- dition, the adept is rejuvenated; white hair and wrinkles disappear." In alchemical düra, mercuric oxide: Yogamartanda 61, in Mallik, SSP, p. 61; and Goraksa Śataka parlance, elixir mercury is valipalita vināśana. 73. This mixture of red and white fluids may have its origins in the combination of 97. Robert Svoboda, Aghora, at the Left Hand of God (Albuquerque, N.M .: Broth- (reddish) soma and white milk in Vedic oblations, and a tantric parallel in the ritual erhood of Life, Inc., 1986), pp. 280-81, quoting his Aghori teacher Vimalānanda. consumption of the yonitattva together with wine: Schoterman, Yonitantra, p. 30. See ibid., pp. 279-90 for an expanded discussion of vajroli mudra. Here as well, 104. Goraksa Paddbati 1.71-75 (= Goraksa Śataka 72-76 = Yogamārtanda 61-64). "heavy" mercury is located at the summit of a hierarchy of fluids: water, milk, clari- Verse 64 reads: śukram candreņa samyuktam rajah sūryeņa samyutam/ dvayos sa- fied butter, honey, and mercury. marasaikatvam yo jānāti sa yogavit: "Semen is conjoined with the moon, uterine 98. Vajra- and sabaja- may also likely refer to the mystic Buddhist traditions blood is conjoined with the sun. He who knows samarasa to be the unity of the known as vajrayāna and sabajayāna. Oli may be a middle Indo-Aryan shortening two, he is a knower of yoga." On the mabābindu, see Gorakb Bānī, Sabadī 237, and

456 457 Notes to Chapter Seven Notes to Chapter Seven

p. 32 of the Gorakhnath Mandir edition of Gorakb Bani, which equates mababindu Lévi, "Manimekhalā, divinité de la mer," in Mémorial Sylvain Lévi (Paris: P. Hart- with mabāsūnya. See also Dhyānabindu Upanisad 87-88, in Varenne, Upanisbads du mann, 1937), pp. 371-83. yoga, p. 11I; and Filliozat, "Taoïsme et yoga," p. 66. 116. Paul Kraus, Jabir ibn Hayyan: Contribution a l'histoire des idees scientifiques 105. Personal communication from N. Sethu Raghnathan, Madurai, January dans l'Islam (Paris: Les Belles Lettres, 1986), pp. 90-93. Kraus identifies Sindiyyat 1985. The titles of a great number of the Tamil alchemical sources are composed with an island off the coast of the Sind, in southern Pakistan. He opines (p. 91, n. of the name of the author followed by the number of verses in the treatise. Cf. HYP 4), however, that Suvarna-bhumi, the "Golden Land," generally identified with the 3.100, with Brahmananda's commentary, on the notion that nāda becomes bindu in "greater India" of southeast Asia, perhaps Burma, may be intended here. On this a woman's body-whence the need for a female partner in erotic tantric rites. On toponym, see Wink, Al-Hind, vol. 1, pp. 334-42. bindu and nāda, see above, chap. 3, n. 102. 117. Needham, Science and Civilisation, vol. 5, pt. 4 (1980), pp. 330-425, espe- 106. Yoga, literally "union" is, in the mystic parlance of the Näth Siddhas, the cially pp. 339-355. union of sun and moon, fire and fluid, ovum and seed, that which is enjoyed (upa- 118. On Darada-deśa, see above, n. 48. On Pārada-deśa, see Sircar, Studies, p. bhogya) and that which enjoys (bhokta), and ultimately, the commingling of the prin- 33 and n. 4; p. 68 and n. 9; and p. 70. ciples of creation and destruction: Dasgupta, Obscure, pp. 235, 238-39. 119. On possible identifications of Bibi Nanī with other mother goddesses, see 107. Siva Sambitā 3.54; Yogatattva Upanisad 74; RA 11.144; 12.251, 265; 18.28, William Crooke, "Hinglaj," Encyclopedia of Religion and Etbics, vol. 6, p. 715. 83, 163. The yogin, whose own body has been rendered golden or adamantine 120. A modern description of the Candrakupa is found in Sastri, Agneya Tirth (Siv Goraks Bāvanī 38, in Goraksa Stavāñjali, ed. Yogi Bhambhulnath [Haridwar: Hinglaj, pp. 29-47. See esp. p. 45. The Life of Apollonius is a work of fiction which Goraksanath Mandir, n.d.], p. 20; Gorakb Bāni Pad 13.1; 50.2) becomes polymor- nonetheless contained numerous geographic and ethnographic details culled by phously perverse at this point, capable of transforming base metals into gold (as Philostratus from ancient travel accounts. Philostratus localizes his "Well of Truth" does mercury), base and noble metals into human bodies, human bodies into met- somewhere in the Indus River region: Philostratus, Life of Apollonius, tr. C. P. Jones als, stones, animals, etc. The mythology of the Näth Siddhas is replete with such and edited, abridged, and introduced by G. W. Bowersock (Harmondsworth: Pen- transformations, often effected by Gorakh. guin, 1970), p. 75. See below, chap. 8, sec. 2b, for a discussion of other "lunar" top- 108. Gorakb Bānī, Sabadī 1.1. Kabīr expresses a similar sentiment in doba 13.23 onyms. in Vaudeville, Kabir Granthavali (Doba), p. 87: "He finds himself face to face with 121. On the rock-hewn images of the sun and moon, see Crooke, "Hinglaj," p. the Lord, and becomes a child (balak) again." For a differing interpretation, see 45; Briggs, Gorakbnath, p. 106 (reporting an account given by Charles Masson in Bharati, Siddh Sabity, pp. 337-38. See above, n. 94, and chap. 10, sec. 2. 1844); and Ctesias, "Indika," in Photios, Bibliotheca (46a.13-19), ed. and tr. René 109. RC 15.13-15; RRS 1.85-88; AK 1.53b-62a; Rasakautuka, ASL MSS no. Henry, 6 vols. (Paris: Les Belles Lettres, 1959), vol. I, p. 136. 4203, fol. 4a.6-9; Sivakalpadruma of Sivanātha, ASL MSS no. 4349, fol. 2b.1-6. 122. Jaan Puhvel, Comparative Religion (Baltimore: Johns Hopkins University 11O. M. F. de Mely, "L'alchimie chez les chinois et l'alchimie grecque," Journal Press, 1986), "Fire in Water," pp. 277-83, especially p. 279, on the Irish Nechtan's Asiatique, 9è série 6 (Sept .- Oct. 1895), pp. 314-40, especially pp. 332-33. well. Cf. Georges Dumézil, Mythe et épopée, vol. 3, part 1, pp. 22-89, especially III. Needham, Science and Civilisation, vol. 5, pt. 4 (1980), p. 337. pp. 27-31. 112. De Mely, "L'alchimie chez les chinois," p. 334- 123. pañcabhūtātmakaḥ sūtastiștatyeva sadāšiva. Cf. BbP 1.32-37. Sadāśiva is, 113. Mababharata 3.110.1-3.113.25. of course, a god who is represented iconographically with five heads: see above, 114. De Mely, "L'alchimie chez les chinois," p. 334. The Renaissance "Hunting chap. 2, nn. 64-66. Elsewhere, the south Indian tradition of Siddha medicine bases of the Greene Lyon," as a metaphor for the quest for the alchemical elixir, may be its practice of pañcikarana ("the five practices," or simply "doing five"), the purifi- of a similar origin: Betty Jo Teeter Dobbs, The Foundations of Newton's Alchemy: "The cation of the elements used in its therapy, upon the assumption that Siva and Sakti Hunting of the Greene Lyon" (London: Cambridge University Press, 1975). are present in every element: Kardaswamy, History, p. 340. 115. This Jabirian account is said to be an elaboration of an Indian legend of 124. Treloar, "Use of Mercury," p. 238. Maņimekhalā, "the jewel engirdled one." Manimekhala's story is referred to in the 125. On this, see David Gordon White, "Why Gurus Are Heavy," Numen 33 Kathāsaritsāgara of Somadeva (Ocean, vol. 9, p. 51). The myth is discussed in Sylvain (1984), pp. 43-44 and notes.

458 459 Notes to Chapter Seven Notes to Chapter Seven

126. The classic theoretical statement is Louis Dumont, Homo hierarchicus, le bam tare taram kanake kanakañ ca līyate sūte. An identical passage is RRS 1.60. RA système des castes et ses implications (Paris: Gallimard, 1978), appendix 4, "Vers une 7.97 makes the same statement in a more elliptical way: "that which precedes [is théorie de la hiéarchie," pp. 396-403. higher on the hierarchy] is imperishable [aksaya] with regard to that which is lower." 127. Each of the five elements is quadrupled within this system: to each of the The Mababharata (5.39.65) describes the hierarchy of the five metals in the follow- five tanmātras (subtle elements, senses) there corresponds a gross element (mabā- ing terms: "Silver is the dross [mala] of gold; copper the dross of silver; tin the bbūta), a "grasping organ" (karmendrīya) and a sensory organ (jñānendrīya). On the dross of copper; lead the dross of tin; and dross [mala: here, iron?] the dross of generation of lower elements from higher in Sāmkhya, see B. N. Seal, "The Physi- lead." Cf. Chandogya Upanisad 4.17.7 for a still earlier formulation. cochemical Theories of the Ancient Hindus," appendix to Ray, History (1956), pp. 136. Two modern empirical explanations for the hierarchy of the metals are 264-65. offered in Misra, Ayurvediya, pp. 494, 498: that which is higher on the hierarchy 128. Caraka Sambitā 4.5.5. The same schema is echoed in the SSS (4.24) of the requires greater heat for its purification than that which is lower; that which is Western Transmission tradition: Schoterman, SSS p. 129. Cf. the slight variations higher on the hierarchy loses less of its mass when heated than that which is lower. suggested by Jean Filliozat, "Taoïsme et yoga," p. 48. 137. RRS 1.30-37; AK 1.18-23. This is an idealized schematization: as noted, 129. Dhatu, "that which supports" (from the Sanskrit root dha, support, the term parada likely derives from the name of the country, located to the west or ground), is the Ayurvedic term for the constituent parts of the body and the al- northwest of India, from which mercury was imported to India: see RC 15.14-15. chemical term for the metals. In both cases, these are the constituent elements of Cf. above, chap. 3, n. 93, and n. 118 of the present chapter. their respective systems, without which the entire (biological or metallic) body 138. RPS 1.17-18; Rasakautuka (ASL MSS no. 4203), fol. 3a.5-6; and Śivakalpa- would collapse and disintegrate. See above, nn. I1, 20. druma (ASL MSS no. 4349), fols. 6a.12-6b.2. The five states or phases of mercury 130. Prior to the 12th c. alchemical synthesis, the term loba, which specifically (avastha) appear not to be arranged hierarchically: RA 10.17-18. means iron, was employed as the generic term for metal: personal communication 139. KCM (ASL MSS no. 3952), fol. 30b.3-4; RA 6.67-68. from Damodar Joshi, Benares, September 1984. 140. The third and fourth members of the hierarchy appear to be reversed in 131. RA 7.97-98. Cf. Chandogya Upanisad 4.17.7: "Just as one would mend gold the discussions of the gatis and vedba: air precedes fire in both cases, in opposition with salt, silver with gold, tin with silver, lead with tin, iron with lead, wood with to the Samkhyan ordering of the five elements, in which fire is the third and air the iron." Enumerations of the metals vary from one source to another, some adding fourth element on the pentadic hierarchy. such alloys as brass, bronze, bell metal, etc. to reach totals of as many as twelve 141. RA 10.13-15; RRS 1.82-83. (in AK 2.1.9) metals. See Misra, Ayurvediya, pp. 493-96, for a discussion of the 142. Mercury is identified with the soul (atman) both here (RRS 1.83) and in nomenclature of the metals in all the major alchemical texts. RU 15.50. The fifth gati of mercury is called jiva-vat in AK 1.40. In the Caraka 132. Syamasundaracarya Vaisya, Rasāyan Sār, 5th ed. (Benares: Syamasundar Sambita's Ayurvedic analysis of semen, only the four lower elements are present Rasayansala, 1971), vol. 1, p. 70. Although this is a modern work (the first edition (on this, see Sharma's notes to RRS 1.81-83, on p. 13 of his edition); it is however appeared in 1914), its statement of the process of resorption (layakrama) respects united, at conception, with äkāśa, ether, which is omnipresent and static in the the classic formulation. ovary: Caraka Sambitā 4.2.4, cited in Dasgupta, History, vol. 2, pp. 302, 307. 133. In a general sense, this south Indian tradition makes greater use of the 143. RRS 1.83. The RA (10.15) stipulates that the fifth gati, the jiva-gati, of system of the five elements, in its formulations of the applications and powers of mercury is invisible, while the first four are visible. Sharma, in his notes to RRS mercurial medicines, than do the northern traditions. Personal communication 1.83 (p. 13), states that, due to its invisibility, the fifth gati of mercury is also called from N. Sethu Raghunathan, Madurai, India, January 1985. Cf. Subbarayappa, daivi-gati, its "divine motion." He suggests the use of such "Siddha mantras" as the "Chemical Practices," in Bose, Concise History, p. 334. agbora, and the adjunction of yogic practices to halt the loss of mercury through 134. HT, p. 176. Cf. Sat-cakra-nirūpana 4-29, trans. with commentary in John this gati, or to render it visible. Woodroffe, The Serpent Power, 9th ed. (Madras: Ganesh and Co., 1971), pp. 144. RA 1.41. Another term for mercury is vyoma-dbārana, "maintaining mica / 330-87. ether [in mental concentration]": Monier-Williams, Sanskrit-English Dictionary, 135. RHT 1.12: kāsthauşadhyo nage nagam vange vangamapi līyate śulbe / śul- s.v. "vyoman."

460 461

Notes to Chapter Seven Notes to Chapter Seven

145. RC 4.106-11; RRS 8.90-94. 159. However, RRS 5.5 states that natural gold (prakrta-svarna) originally arose

146. RRS 8.95. from the rajoguna. See also Biardeau, Hindouisme, pp. 106, 11O. The pivotal, female,

147. RA 5.28, 32-41. and creative role of ether/empty space is set forth in the Caraka Sambitā (4.2.4) 148. These are the "M-words" or antisacraments of tantric practice, possibly discussion of conception: see above, n. 142. subversions of the pancagavya, the five pure products of the cow that were the 160. The association of rajas, the guna of activity, with the Goddess, is supported orthodox brahmin's mainstay: māmsa, meat; matsya, fish; mudra, parched grain; etymologically. Rajas at once means menstrual blood, pollen, redness, and passion, madhu, spiritous liquor; and maithuna, sexual intercourse. all terms with highly feminine connotations in India. 149. The five "nectars," here are bodily secretions used in secret tantric ceremo- 161. Abhinavagupta lends a pivotal role to the fifth element, which he calls nies: blood, semen, urine, feces, and flesh: KJnN 11.5, 11. ākāśa-sakti, the "power of space." On this, he says that, unlike the powers of the 150. These are the "stools" or "seats," pilgrimage sites of the Goddess. As many four lower elements, "the power of space plays no role in sustaining the finite sub- as 134 pithas are enumerated in certain lists. Five are singled out, however, and are ject. For the power of space is inherent in the individual soul as the true subjectivity, identified here with the five elements. Kāmākhyā (here called Kāmarūpa) is one of at once empty of the objects and providing a place in which objects may be known.

these: Schoterman, SSS, appendix 2, pp. 222-25. This is the Emissional Power in its lowest form" (commentary on Parātrīsikā 151. On the five kulas, see Marie-Therèse de Mallmann, "Divinités hindoues Laghuvrtti 5-9a, translated in Muller-Ortega, Triadic Heart, p. 131). See below, note dans le tantrisme bouddhique," Arts Asiatiques 10 (1964), pp. 68, 72-74. On the five 176, on the kañcukas. amnāyas ("transmissions"), emanating out of the five faces of Siva, see Kulārnava 162. The goose seems to be identified with the breaths even before the syllables

Tantra 3.7-10. of its name come to represent the sound of inbreathing and outbreathing them- 152. The more traditional number is nine: however, the Bengali Goraksa Vijaya selves. The original notion would have been that the goose of the breath kept one and other traditions speak of the five original Näths. These are Adinath, the primal alive so long as one of its feet remained tethered to the body: its flight, up and out Nath, and four Naths who arose from his body parts; Gorakhnath, who arose from of the body, is death. This is the implication of Aitareya Aranyaka 2.1.8 (citing Rg his head; Matsyendranāth, who arose from his navel; Kanerināth, who arose from Veda 1.164.38), even if this text does not mention bamsa explicitly. his ears; Jalandharanāth, who arose from his bones; and Cauranginath, who arose 163. Silburn, Kundalinī, p. 23. from his limbs: Mukherji, Gopicandra Nātaka, p. xxxvi. 164. Goraksa Sataka 64-70, especially 67. 153. These three groups of five are treated together in a work of the Näth Sid- 165. RA 2.89; 11.151. Khecarī is also referred to in the context of a thirty-two- dhas entitled Goras Ganes Gusti (in Srivastav, ed., "Gorakbbānī" Višesānk, pp. 388- syllable Kubjikā mantra, in 3.9. A khecarī clan lineage (kula) is named and its mantra 90. All three groups are related, serially, to the five elements. discussed in 3.17-20 and 12.59. Certain herbs are also referred to as kbecarī, be- 154. Dumont, Homo bierarchicus, pp. 104-5, 397. See above, n. 126. cause they produce the power of flight in mercury: 12.52-54. One of the uparasas, 155. Abbraka and gagana are the two more common terms; the RA (6.31) also kasīsa (ferrous sulfate) is called kbecara in RC 11.I and RHT 9.5. It is called khaga mentions vyoma. The RC uses all four terms interchangeably in chapter 16, em- ("sky-going") in RA 7.56. Khecara is one of the eight "magical" siddhis listed (along ploying kha once, in 16.37. Kha-puspa, "cavity-" or "hole-flower," is a common term with rasa-rasāyana) in the 3d-11th c. A.D. Sādhanamālā: Bhattacharya, Sādbana- for menstrual blood in tantrism: Bhattacharya, History, pp. 132, 136. māla 2:1xxxV. 156. The Rasakamadhenu passage is reproduced without verse citation in Shar- 166. RHT 4-5; RC 16.4, 44, 52-55, 75. Clipping the wings of volatile mercury is the remedy for capalyadosa, the "flaw of instability." This is effected through niya- ma's commentary to RRS 2.1. 157. Rasendra Bhāskara 3.42. Abbra connotes wetness in such Vedic passages as mana ("regulation"), the seventh of the eighteen alchemical samskāras: Misra, Atharva Veda 1.12.3 (cited in Dasgupta, History, vol. 2, p. 331, n. 2). Āyurvedīya, p. 195. 158. See above, n. 144. Vyoman is the "firmament" in Atharva Veda 8.18.13. The 167. The play of correspondences is not perfect here, since evaporation can be five akasas are identified with the five vyomas in the SSP and are discussed by Kavi- identified with the element air, rather than ether or space. Both appear to be pres-

raj, Tāntrik Sādbanā, p. 401. ent here. Hamsaga mercury (RA 10.14-15; 11.126-31; 12.68-70; RRS 1.81-83 and

462 463 Notes to Chapter Seven Notes to Chapter Seven

pp. 13, 133) is related to the element air, referring as it does to mercury's tendency trīšikā Laghuvrtti 5-9a, translated ibid., p. 131. Cf. Dasgupta, Obscure, p. 219, and to evaporate, bamsaga, in the sunshine whose beams are called hamsapada in Ath- Srivastav, ed., "Goraksa" Visesānk, p. 272, for Nāth Siddha interpretations. arva Veda 11.4.21-or into the wind as breath, also called bamsa in yoga. 178. BbP 2.5b-7a: tatrādau tu rasah sūtah śivadehādvinirgataḥ/ tato 'nyattu gha- 168. On this, see Cirpatjī kī sabadī 23, in Mallik, SSP, p. 84: "bind the flying nam khyātamumāyonisamudbhavam/ sitam pītam tathā raktam krsnañ caiva tu bamsa" (udaina hamsa lagai bandh). Cf. HYP 4.92 and BbP 5.148, 168-69. varņataḥ // vajrākhyañ caiva māņdūkam dvividham tu prakīrtitam. 169. Elsewhere, the Goraksa Paddhati (p. 60 of the Bombay edition, cited in 179. For abirāja, see RA 13.370. One type of mica is called nāga, "serpent," Bharati, Siddh Sahity, p. 340), calls the ether the void (sunya) in which Siva (whose because of the hissing sounds it makes when heated: RC 10.6. semen is mercury) dwells. 180. The alchemical sources refer repeatedly to lead and tin as pūta, "smelly," a 170. RC 15.4-12. fact that may have inspired this mythic representation of them as mala, dross. In 171. Siddhinandan Misra, who translated this text into Hindi, chooses to see in the gradual transformation of the human body into an alchemical body, the first of the plural use of the term dhatu a reference to the sexual fluids of both Siva and the an eight-stage process is called "piercing" or "transmuting" the skin (rvacāvedha), Goddess (śiv-pārvatī ke antim dhātu): Misra, RC, p. 289. Elsewhere (16.8), Soma- by which the body is transmuted into lead and tin (pannaga): RA 18.150 deva calls mica (abbraka) and mercury (pārada) the final dbatus of the Goddess and 181. Seșa (along with his brothers Ananta, Vasuki and Taksaka) is the Lord of Siva. Here, these dhatus are "final" in the Ayurvedic sense of the term: as the sexual Serpents in Hindu mythology. Somadeva's evocation of the Lord of Serpents who emissions of the two, they are the final distillates of divine processes of digestion. drinks the divine semen that has fallen from the mouth of Agni would thus refer 172. This is the Vedic myth of Indra's theft of Soma, the divine draft of immor- to this same figure. On Seșa as "remains," see Charles Malamoud, "Observations tality. In the Rg Veda (4.26-27), it is Indra who, mounted on an eagle or falcon sur la notion de 'reste' dans le brahmanisme," in Cuire le monde, pp. 28-29. (śyena), steals the soma (later identified with lunar nectar and semen) from the atmo- 182. Siva's name here is to be identified with the fire of yoga (yogāgni) that spheric Gandharvas. In Brahmanic myth (Taittirīya Sambita 6.1.6; Śatapatba Brāb- destroys the fire of time (kalagni) in the microcosm. The universal resorption is a mana 3.6.6.2), the rivalry is between a serpent named Kadru and a falcon named reabsorption of both matter and time: time which generally cooks all creatures is Suparnī. This develops, in later epic myth, into the baroque account of the bird itself cooked by the dissolution. Time yields to eternity, until the universe is re- Garuda's "theft" of the soma from the serpents, or from Indra and the gods: Rāma- emitted by the great yogin. The fire of Siva's yoga is also greater than the fire of yana 3.35; Mababbārata 1.20.1-1.30.25. desire (kamāgni), as illustrated in the myth of his incineration of Kāma, the Hindu 173. For the Babylonian myth of Etana and the Zu bird, as well as a wide array Eros, with fire from his third, yogic, eye. of Indian and Indo-European parallels to this mythic theme, see David M. Knipe, 183. Whence the final verse of the RA (18.228), which states that the perfected "The Heroic Theft: Myths from Rgveda IV and the Ancient Near East," History of Siddha dwells eternally in the highest level of the cosmic egg-sometimes known Religions 6:4 (May 1967), pp. 328-60. Cf. Jarich G. Oosten, The War of the Gods: as Siddha-loka-even when all living beings have been annihilated in the terrible The Social Code in Indo-European Mythology (London: Routledge & Kegan Paul, flood (of the end of the world). 1985), pp. 69-71. 184. In later sources, particularly Saiva Purānas and Tantras, it is Śiva who is 174. Mababbārata 1.20.1-1.30.25. the great yogin, a distinction he has retained down to the present day in India. In 175. RRS 1.80; AK 1.35-37; RC 15.24-25, 43-49; BhP 3.22. Misra (Āyurvedīya, this light, the statement that mercury is the exudation or sweat of the sleeping (i.e., p. 197) describes this purification in terms of a snake shedding its skin. It is note- yogically withdrawn) Siva, is a telling one: RA 1.36. See above, chap. 2, sec. 5. worthy that in Ayurvedic embryology, heating in the uterus generates seven layers 185. Puranic cosmogony is in fact an appropriation, run in reverse, of the stages (kalās) of skin and the seven dhatus around the blood and semen that were originally of yogic withdrawal: Asian Mytbologies, s.v. "Puranic Cosmogony." present at conception: Dasgupta, History, vol. 2, pp. 312, 317. 186. Ibid., and Biardeau, Hindouisme, pp. 121-22. A translation of the Visnu 176. On the kancukas in Saiva Siddhanta thought, see Gonda, Medieval, p. 186; Purāna (6.3.14-41; 6.4.1-10) treatment of cosmic dissolution is in Dimmitt and on the Vidyeśvaras, see Rao, Elements, vol. 2, part 2, p. 397, and above, chap. 6, Van Buitenen, Classical Hindu Mythology, pp. 41-43. n. 56. 87. Mircea Eliade, The Sacred and the Profane: The Nature of Religion (New York: 177. TÃ 13.103-4, translated in Muller-Ortega, Triadic Heart, p. 139. Cf. Parā- Harcourt Brace Jovanovich, 1959), pp. 54-55; Stella Kramrisch, The Hindu Temple,

464 465 Notes to Chapter Eight Notes to Chapter Eight

2 vols. (Calcutta: University of Calcutta, 1946; reprint Benares: Motilal Banarsi- 7. This is the particular perspective of the Nath Siddhas who, anti-Sakta to the dass, 1976), vol. I, p. 62, n. 105. extreme, deny whenever possible any constructive role to the feminine. This is 188. Filliozat, Doctrine classique, p. 19. Nagarjuna is called the son of Vasuki in a in marked contrast to the Trika system, for which the kundalinī is the dynamic Näth source: NSC, p. 226. A Rajput royal lineage calls itself the nāgavamsi, "serpent manifestation of Siva in the universal macrocosm as well as the human microcosm: lineage," claiming descent from one or another of the great serpents of Hindu Silburn, Kundalinī, p. 24. Gorakhnāth compares women to tigresses and succubi: myth: Ruben, Eisenschmiede, p. 110. see below, n. 92. 189. See below, chap. 9, n. 102. 8. Biardeau, Hindouisme, p. 167. Bhogavati is the name of the city or a sacred 190. Kaviraj, Bhāratīy, vol. I, p. 311; Silburn, Kuņdalinī, p. 42, citing Abhinava- tirtha of the serpent race, in the Hindu epics: Monier-Williams, s.v. "[1]bhoga," gupta, Tantrāloka-viveka 22. Cf. Śvetāśvatara Upanisad 1.13, in which meditation on citing the Rāmāyana, Mabābbārata, and Harivamsa. the Om causes the subtle mark (linga) of the divine within all humans to manifest 9. Padoux, Vac, pp. 81-83. Cf. Sanderson, "Mandala and Agamic Identity," pp. itself. 178, 185; Brooks, Secret, p. 111; Silburn, Kundalinī, pp. 19-24, 52-53, 64-66. 10. Silburn, Kundalinī, p. 27, 42, citing Ksemarāja's commentary on Šivasūtra Chapter Eight 2.3. Cf. Kaviraj, Bhāratīy, 1: 309. 1. Toni Schmidt, "Fünfundachtzig Mahäsiddhas," Ethnos 2-3 (1955), p. 120; 11. On the bhukti/mukti dyad see Brooks, Secret, pp. 91, 105. Rasik Vihari Joshi Rolf Stein, "Jardins en miniature," p. 54. A seemingly identical image appears in ("Notes on Guru, Dīksā and Mantra," Ethnos 37 [1972], pp. 106-7) discusses moksa the context of the pilgrimage to the shrine of Hinglaj Devi, in eastern Baluchistan. and bhoga as two forms of tantric initiation. But see Sanderson's ("Saivism," pp. This is a lingam-yoni, the image of the sexual union of Siva and the Goddess, which 667-68) important nuanced discussion of the mutual exclusivity of moksa and bhoga is branded on the upper right forearm of Nath Siddhas travelling to the site. The in early tantrism. Moksa was the goal of "liberation-seeking" celibate ascetics, while image, reproduced by Briggs (Gorakbnath, p. 110), is of a scrolled letter U in the bboga, supernatural experience generally gained through sexual union, was the goal hollow of which are nested two concentric circles. of such "power-seeking" tantric movements as the Mantramārga. 2. HYP 3.1. 12. See MBbT 14.3-10, on divya-, virya-, and pasu-bboga, also in relation to the 3. RHT 1.7; MBbT 3.12-14; SSP 3.5; Goraksa Paddhati 1.60; BhP 9.117-19. Cf. kundalinī. Dvivedi, Nath Sampradāy, pp. 187-88; Misra, Āyurvedīya, p. 28; Kaviraj, Bhāratīy, 13. MBbT 3.1-16; 14.13-14; and Bharati, Tantric Tradition, pp. 252, 260. vol. I, pp. 555-56; Srivastav, "Goraksa" Višesānk, p. 132; idem, "Gorakb Bānī” Viš- 14. HYP 1.27; 3.11; 3.61-69. esank, p. 332; NSC, p. 131. 15. Silburn, Kundalini, pp. 15, 27, 52. 4. Todala Tantra 9.14b-15b: prthvīcakrasya madhye tu svayambhūlingamad- 16. See Silburn (Kundalini, pp. 27, 52) for Abhinavagupta's (TA 3.171) play on bhutam/ sārddhatrivalayākārakuņdalyā vestitam sadā/ lingacchidram svavaktreņa the word vis[a]: when she sleeps, the kundalini holds the poison that destroys human kuņdalyācchādya samsthitā. Cf. 9.21. Alternatively, her mouth is said to cover the vitality; when she awakens, this poison transforms itself into all-pervading (vis) brabmadvara, the "door of brahman," which is the opening to the lower end of the power. suşumņā nādī: Goraksa Paddbati 1.17-19; Yoga Mārtanda 36. Cf. NSC, p. 131, which 17. This is the origin myth of poison, in RRS 29.1-I1. states that she sleeps with her mouth closed around the "western" linga. 18. It is this mythic feat and the quest to, once again, become a "second Siva," 5. Silburn, Kundalinī, pp. 22-23. This discussion, of the place of the kundalini that appears to be the raison d'être of the so-called "Poison Tantras" (visatantras), in the hathayogic system of the Nath Siddhas, differs significantly from Silburn's of which the Yogaratnāvali revealed by Śrīkantha Śiva (ASL MSS no. 4184) is an ex- descriptions (ibid., pp. 19-24, 63-69) of the two kundalinis (the adhah- and ürdhva- ample. kundalinīs) or the three aspects of the kundalinī (sakti-, prāna-, and parā-kundalinīs) 19. He is further identified with Vajrabhairava. This god's consort is Ekajatā: in the Trika Kaula system of Abhinavagupta and others. see Beyer, Tarã pp. 42-43, 54; and above, chap. 3, nn. 89-90. 6. Goraksa Paddhati 1.47-56; Yoga Mārtanda 36-40. The editor of the Goraksa 20. Spanda Kārikā 3.16, trans. in Harvey Alper, "The Cosmos as Siva's Paddbati, citing "other sources," reproduces the following verse (numbered 2) on p. Language-Game," in idem, Mantra (Albany: SUNY Press, 1989), p. 280. 28: "in between the ida and pingala is the child widow, the kundalini." 21. TA 3.171.

466 467

Notes to Chapter Eight Notes to Chapter Eight

22. Gūgã Chauhan or Gūga Pir, whom the Nath Siddhas claim to have been the Goddess in a raised pavilion amidst the Ocean of Milk: Sen, History, p. 44- one of their number, is renowned for his power over serpents: Rose, pp. 395, 409. Another Näth legend explains Matsyendra's piscine origins. A king named Udho- Nepali legend maintains that Gorakhnath held back the "great serpents" of the dhar is cremated. His navel will not burn and is cast into a river, where a fish de- rains in Nepal for some twelve years, a reference that at once harks back to Vedic vours it to give birth to Matsyendra: Rose, Glossary, p. 393; Himavatkbanda (of the myths of Indra taming the rain serpent Vrtra and to Gorakh's yogic control of Skanda Purāņa), p. 73 (appendices); and Nārada Purāna, 2.69.6, cited in NSC, p. 22. the kundalinī. 28. Jayaratha, in his commentary to TĂ 1.6 (vol. 2, p. 24), quotes a source that 23. BhP 9.118b-119a. On the mythic theme, of a disciple becoming his guru's identifies the Siddha Mina with Macchanda, in this episode. The relationship be- teacher and the guru becoming his disciple's pupil, in its relationship to the ideol- tween Goraksanātha (7) and his guru Mīnanātha (6) in HYP 1.5 is symmetrical to ogy of brahmanic sacrifice, according to which that which precedes (e.g. a prior that obtaining between the figures Goraksa-pā (9) and Mīna-pā (8) in the Tibetan sacrifice) is necessary to that which follows (e.g. a present sacrifice, lit from the Grub thob. This squares, moreover, with the statement made by Goraksa in the embers of the prior sacrifice); but also that that which precedes can only persist second verse of his earliest work, the Goraksaśataka, that his guru was Mīnanātha. through that which follows (the prior sacrifice is reborn through the present sacri- It is a later conflation that identifies Matsyendra (the second Siddha, listed after fice), see below, ch. 9, n. 104. Adinatha, in the HYP list) or Lui-pa (the first Siddha of the Grub thob list) with 24. According to the K7iiN (22.10), the Yogini-Kaula doctrines were found "in Mina. The historical Goraksa or Gorakhnath was an early 13th c. figure, whereas every house in Kamarupa." Three Padas (2, 43, and 49) of the Gorakb Banī take the Matsyendra, his purported guru, could not have lived later than the 1oth c .; it is form of teachings made to his guru after his rescue from the Kingdom of Women: Mina who was Gorakh's guru. Srivastav, "Gorakb Bāni" Visesank, p. 327. So too, in the Bengali Min Cetan-Goraksa 29. Matsyendra calls himself matsyaghna in KJiN 16.37, where he says he is Vijay cycle, it is made clear that the yogins of this mystic doctrine and the women from the kaivartta fisher caste (see below, n. 108). He calls himself machli ghana, of the Plantain Forest were one and the same. Indeed, the origin of the "house- "Killer of Fishes," and Machandar Nath, a vernacularization of his Sanskrit name, holder Naths" is also traced back to these yoginis, who inveigled celibate yogins in a song entitled Matsyendranātbjī ka pad, in Mallik, SSP, p. 76. He is also known into moving in with them: Dvivedi, Nath Sampraday, pp. 22, 56. as Macchaghnapāda, Macchendrapāda, Matsyendrapāda, Mīnapāda, Mīnanātha, 25. The opening lines of the 17th c. Bengali work Min Cetan are the Goddess's Macchendra, Macchendapāda, and Macchindranāthapāda: Bagchi, KJñN, p. 6; and question to Siva: "Why is it, my Lord, that thou art immortal and mortal am I? Dyczkowski, Canon, p. 163 (n. 23 to p. 62). On Matsyendra's ambiguous relation- Advise me the truth, O Lord, so that I also may be immortal for ages" (tumhī kene ship to Mina in Nepal, see Locke, Karunamaya, p. 427. tara gosānī āmhi kene mari/ hena tattva kaha dev joge joge tari). In answer to her 30. There are actually two thefts and two recoveries of these scriptures at Can- question, Siva expounds on hatha yoga, in the monologue that Matsyendra over- dradvīpa, and it is only on the second occasion that Siva-Bhairava becomes Matsy- hears: Dasgupta, Obscure, p. 228. endra: KJnN 16.23-36, 52; 22.10-12. The many variants of this legend are re- 26. Matsyendra's name, like that of Nagarjuna, cannot be attached to a single viewed in Bagchi's introduction (pp. 6-32) to the KJiN. historical figure. It is rather to be seen as an initiatory title, given to tantric prac- 31. KJĀN 16.21: yadāvatāritam jñānam kāmarūpe tvayā mayā ("[When] the titioners who had reached a certain advanced level of mastery: Tucci, "Animadversi- [Kaula] gnosis was brought down at Kāmarūpa by you [and] by me"). Cf. 16.26a. ones Indicae," p. 133. However, 22.10b and 12ab state that "this teaching [is found] in the household of 27. Matsyendra is described as dwelling in the belly of the fish in the c. A.D. 1290 every yoginī in Kāmarūpa ... This great teaching was brought down at Candra- Jñāneśvari (18.1729-35) of the Marathi saint Jñāneśvara: Avedyanāth jī Mahārāj, dvīpa. That [teaching] whose abode was the belly of a fish (mabamatsyodarasthitib) "Mahayogī Guru Gorakhnath evam unkī Tapasthalī," in Baba Cunninathji, ed. Tan- is sung at Kāmākhyā." Prakāś (Gorakhpur: Gita Press, 1981), p. 186. The same description is given in 32. Kāmarūpa is generally identified with the important pitha of the Goddess at another Marathi source, the 13th c. YSA. A similar account is found in the Skanda Kāmākhyā (Gauhati district, Assam). It is also a site within the subtle body, located Purāna (6.263.33-61). The Goraksa Vijay and Min Cetan call him Minanath and say at the center of a downturned yoni, which is itself identified with the sexual organ he took the form of a fish to eavesdrop on Siva, who related his secret teachings to of the tantric consort, the yoginī. See below, nn. 89-90.

468 469 Notes to Chapter Eight Notes to Chapter Eight

33. The Grub thob, a Tibetan translation of Abhayadatta's 11th-12th c. Catura- of tanks before emptying into the Varana. It was only in exceptionally heavy flood- sitisiddha Pravrtti ("Acts of the Eighty-four Siddhas"), is found in vol. 86, part I of ing that the volume of water flowing in the Ganges would back up into its Varaņā the Tanjur: see above, chapter 3, nn. 139-40. tributary, and in turn into the Matsyodari channel, reversing its more common 34. Locke, Karunamaya, pp. 282, 297-300, 433 n. 80. northward flow. A small tank, all that remains of the Matsyodari Lake of old, 35. Bagchi, KJiN, p. 22. The color of the image is noted in the earliest datable is today called Macchodari: Eck, Banaras, pp. 116-18. Following James Prinsep's manuscript (11th c.) in which it is mentioned: Locke, Karunamaya, p. 300. 1822 map of Benares (which she reproduces schematically on p. 47), Eck maintains 36. Marie-Therèse de Mallmann, "Divinités hindoues dans le tantrisme boud- that the "reversed" southward current of the Ganges-Matsyodari emptied into the dhique," Arts Asiatiques 10:1 (1964), p. 71, referring to Archive Photograph no. Ganges proper near Benia Talb (the "Confluence Pool"), slightly to the north of 1623416/4, cliché no. 53.24, of the Musée Guimet. The depiction of Matsyendra Daśaśvamedh Ghāt. While such may have been the case in the early 19th century, emerging from the mouth of a fish can be cause for confusion, particularly in west- the Puranic sources indicate that this occurred a mile further to the south, via the ern India where Dāriyanāth (also known as Jhullelāl, Khwāja Khizr, Amarlāl, Zinda Asi River. A string of lesser lakes and ponds, located by Prinsep between the 19th- Pir, etc., according to region) the "lord of the current," is the god of the In- century outlet and the Asi River, could well have served to channel the Matsyodari dus River: Briggs, Gorakhnath, p. 65. In Pushkar, where there is a temple to this down to the Asi in earlier centuries. deity and a temple to the major Nath Siddhas, a Nath custodian of the latter stated 44. Siva Sambitā 5.100, 104, 109. Cf. Eck, Banāras, p. 26. that "Matsyendra was the same as Jhullelal": personal communication from 45. A Hindu commonplace identifies Benares with the two-petaled ajña cakra Dominique-Sila Khan, Pushkar, India, April 1993. which, located at a point behind the bridge of the nose, is the final point of conver- 37. Chattopadhyaya, Taranātha's History, p. 153 n. 28, and p. 178 n. 11; Bagchi, gence of the ida and pingala channels. KJiN, pp. 22-23; Mallik, SSP, p. 15; and V. W. Karambelkar, "Matsyendranāth and 46. Linga Purāņa, Kapileśvara Māhātmya, Omkāranirņaya, and Guhyayātana- His Yogini Cult," Indian Historical Quarterly 31 (1955), p. 362. varnanam, chap. 9 (reproduced in the 12th c. Krtyakalpataru of Bhatta Laksmīdhara 38. TA 5.57b-58a, translated in Muller-Ortega, Triadic Heart, p. 123 (italics my (vol. 8, Tirthavivecanakānda, ed. K. V. Rangaswamy Aiyangar [Baroda: Oriental In- own). Muller-Ortega translates matsyodaradasa as "the condition of the belly of the stitute, 1942]), pp. 34-35. Lolarka Kund lies near the mouth of the Asi River, near fish." The term matsyodarasthitih, found in KJnN 22.12, has the same sense. the southern edge of Benares: Eck, Banāras, pp. 177-78. Lolārka's solar associations 39. Amaraugha Prabodha 10: bibhrānah pavanam hathānniyamitam grāso 'sti make the southern Asi River the obvious macrocosmic homologue of the solar pin- minodare / kaivalyo bhagavānvimucya sahasā yāvannacetatyasau. This text is found galā nādī. This somewhat skews the hathayogic symmetry in another way, however: in Mallik, SSP, p. 49. Gorakh claims to be quoting a text entitled the Samputa. ideally, the Ganges ought to flow backwards, northwards, and upwards from the These words, put in the mouth of Siva and addressed to Pārvatī and Mīnanātha, sun (Asi) towards the moon (Varana); however, since the Ganges' normal flow is appear to constitute the beginning of the teachings offered on the shore of Can- northeasterly at Benares, its backward-flowing stream must flow in a southwesterly dradvīpa. I am grateful to John Roberts for his aid in translating this passage. and "downward" direction, from moon to sun. What is essential here is that mats- 40. Diana Eck, Banaras: City of Light (Princeton: Princeton University Press, yodarī-yoga is identified with the Ganges running against its normal flow. 1983), p. 118. 47. Krtyakalpataru, Tirthavivecanakānda, p. 35. Visuvam also means "at the equi- 41. Skanda Purāna 2.2.12.21-40, cited in Eck, Banāras, p. 94 nox": the flooding that would have precipitated the appearance of this stream could 42. Rudrayāmala Tantra 27.45, which further identifies the three rivers with pür- only have occurred during the summer monsoon: it is therefore possible that mats- aka, recaka, and kumbbaka; and moon, sun, and fire. It further mentions the kun- yodarī yoga fell on the autumnal equinox, at the culmination of the monsoon season. dalini (as kundali), as well as the six cakras, which it associates with six Sivas, six The Darsana Upanisad (4-43b-44a) identifies the two "equinoxes" with the concen- Saktis, and six "shafts" (satšara): 27.42-56. Cf. Caurangīnāth, "Prāņ Sānkalī," verse tration of the life force/yogic breaths in the mūladhara (spring) and the cranial vault 80, in Dvivedi, Nath Siddhom, p. 23. On the geographical disappearance of the (autumn), respectively (in Sastri, ed., Yoga Upanisads). Once again, the latter "equi- Sarasvati River, see Kalyān Sakti Ank, p. 651, which also mentions that a "Sarasvatī nox" would correspond to a generalized upward flow of the life force through the Gangā" exists near Kedārnāth, in the Himalayas (ibid., p. 671): see below, n. 153. medial channel. 43. In a "normal" monsoon, the Matsyodari would flood northward into a series 48. Krtyakalpataru, Tirthavivecanakāņda, p. 59.

470 471 Notes to Chapter Eight Notes to Chapter Eight

49. Quoted from the Kasī Khanda of the Skanda Purāna by Mitramiśra, "Tīrtha Prakāśa," in Vīramitrodaya, p. 240 (cited in Krtyakalpataru, Tīrthavivecanakānda, p. 59. Eck, Banāras, p. 117.

lxxviii). Cf. Narayana Bhatta's (ca. 1580) Tristbalisetu (p. 140, cited in Kubernath 6o. Bagchi identifies Moon Island with the Chinese "double-mountain" known

Sukul, Varanasi Down the Ages [Bhargava Bhushan Press, 1974], p. 200): "The Mat- as K'un-lun, which is itself located within the Taoist subtle body ("The Island of

syodarī tīrtba is indeed twofold because it is [both] there in its own place [the Mat- K'un-Lun and Candradvīpa," in Bharata-Kaumudī: Studies in Indology in Honor of

syodarī Lake] and outside of it, due to its channel that runs around [Benares]. It is Dr. Radba Kumud Mookerji, 2 vols. [Allahabad: Indian Press, 1945, 1947], vol. I, pp.

most preeminent when, during the rainy season, the water of the Ganges, whose 47-55). Cf. idem, K7nN, pp. 29-32. Dasgupta (Obscure, p. 384) locates it on the

course is swollen during the rainy season, flows in a reverse direction via the mouth coastline of the Bakerganj district of Bangladesh. S. M. Ali (The Geography of the Purānas, 2d ed. [New Delhi: People's Publishing House, 1966, 1973], pp. 77-78 of that channel." 50. Also called the brabma-marga, the "path of brahman," which is an echo of [citing Vayu Purana 45.51-58], 86-87, and fig. 8) locates it to the west of Lake

the Linga Purana (in Krtyakalpataru, p. 59) reference to matsyodarī as the fluvial Baikal in central Asia.

[form] of the brabman-syllable. Gorakhnāth refers to the susumnā as the "subterra- 61. On the Candragiri toponym in the Kathmandu Valley, see above, chap. 4, n. 49. nean Ganges": Gorakb Bānī Sabadī 2. 51. On the commonplace identification of kumbbaka with the susumna and the 62. KJnN 16.33-35.

Sarasvati River, see Rudrayāmala Tantra 27.43-45. A cryptic passage from the third 63. Chapter colophons to the K7nN call it the revelation of the "Great Yoginī-

chapter of Durjayacandra's Amitapadā commentary on the Buddhist Catuspīthatan- kaula, brought down to earth by Matsyendrapada at Moon Island." The same is

tra is adduced by Tucci to explain Jayaratha's gloss-in which he gives maccha as also stated in textu in KJnN 22.12.

a synonym for paśa-on Abhinavagupta's reference to Macchanda in TA 1.7. The 64. KJnN 5.6, 12. 65. KJiN 17.23a; 20.11a. For the western linga in the head, see Gorakb Bānī, Buddhist commentary gives the compound makara-mīnaka: Tucci, "Animadversi- Sabadī 187a, and the Raja Rānī Sambad of Gopīcand, v. 105 (in Dvivedi, Nath Sid- ones Indicae," p. 134. I maintain that this compound is a reference to the two riv- erine goddesses, the Gangā and Yamunā, whose marine vehicles are, precisely, the dhom, p. 9). On the "microcosmic west," see also Gorakh Bani, Sabadis 41 and 267.

makara and mina; and that here too, this pair symbolizes the two peripheral chan- 66. KM (London, Wellcome MSS g501) fol. 4a.5-7.

nels of the subtle body, which must be emptied before the susumna can be filled. 67. See above, chap. 3, n. 137.

52. Kumbbaka is compared, in Brahmananda's commentary on HYP 4.27, to the 68. Schoterman, SSS, pp. 36-39. A manuscript of this work, in Newari script,

binding of mercury. is dated to A.D. 1197 by Shastri (Catalogue, 1:lxiv, 11I-12) and to A.D. 1395 by Bag-

53. HYP 2.7-9, 35. The kundalinī doubles as the susumna nadi and is also called chi (KJãN, p. 67).

the "infant widow" in HYP 3.109-10. A Matsyodaratantra, of which the chief subject 69. Dyczkowski, Canon, p. 68-69. This squares with Bäul tradition, which iden-

is the the subtle body of hatha yoga, would appear to take its name from the prac- tifies the fish (above, n. 58) with the moon: "My unknown moon moves through

tices we have been describing: HTSL, p. 102. the water in the form of a fish." (McDaniel, Madness, p. 184).

54. Tristhalīsetu, p. 168, cited in Eck, Banāras, p. 119. 70. Amaraugbaśāsana, p. 1O.

55. Goraksa Sataka, v. 2. This mention of Minanath is found in all recensions of 71. If, in the subtle physiology of the cranial vault, there exists a correspond-

this work. Cf. the ordering of Matsyendra, Mina, and Gorakh in the Grub thob and ing eastern mountain to the western Candraparvata of the KM (see above, note

HYP Siddha lists. See above, n. 28. 66), then the three mountains, taken together, represent the so-called trikuta, the

56. See above, chap. 5, nn. 83-92. "triple-peaked" configuration located in the same region, which is also identified

57. See above, chap. 5, nn. 93-96. with the three phonemes of the syllable Om (on this, see Silburn, Kundalinī, p.

58. McDaniel, Madness, pp. 182-85. The opposition between minaka (fish) and 132). This would further correspond to the sacred geography of the city of Benares,

makara (crocodile) in Durjayacandra's Amitapada (see above, n. 51) may also refer which is built upon three hills, the which are identified with the three tips of Siva's

to this body of sexual practice. trident, called the trisulabbumī in the geography of the subtle body (ibid., p. 58). 72. On the lower western linga, see Goraksa Śataka 19. Cf. SSP 2.2, which states

472 473 Notes to Chapter Eight Notes to Chapter Eight

that the kundalini sleeps with her mouth over the pascima-linga, located in the sec- nirodhe kumbhakaḥ proktaḥ prāņāyāmam prakīrtitam/ dhyātvā kālāgnibījam tu ond cakra, the svadbisthana. It is out of this vortex that seed is emitted in ejacula- yugāntānalasaprabham/ nyaset pādatale mantrī jvālāmālākulam mahat. tion: Srivastav, "Gorakb" Visesānk, p. 131. 81. See above, n. 49. 73. E. Obermiller, ed. and tr., History of Buddhism (Chos-bbyung) by Bu-ston, 2 82. Misra, Āyurvedīya, p. 28; Kaviraj, Bhāratīy, vol. I, pp. 555-56. vols. (Heidelberg: Harrassowitz, 1931-32), vol. 2, "The History of Buddhism in 83. ulatī gang samudrabi sokbai: in Kabīr Granthāvalī, ed. Shyam Sundar Das (Be- India and Tibet," p. 120 (= fol 109a of Bu-ston's text). Obermiller gives the tran- nares: Kashi Nagaripracarini Sabha, 1954), p. 141. Cf. a verse from another Kabīr scription "Badaha" for the name of this island, a name which makes no sense. The anthology, Sant Kabīr, ed. Ramkumar Varma (Allahabad: 1947), p. 20: ulațī gang Nätha-abbyudaya-tantra cited by Bu-ston is called the Mabā-kala-tantra-rāja, in the jamun milāvu/ binu jal sangam man mabim nhāvu. See also Gorakb Bānī Sabadī 55, Kanjur. A microfilm (reel E-1358/7) of a manuscript, in Newari characters, of this 90, 115. work, held in the Nepal National Archives, gives the reading bādāvam (fol. 55b, 84. See above, n. 13. On the "five nectars" of early Buddhist tantra, see Gubyasa- line 5). It is with this reading that I emend Obermiller's translation. māja Tantra (p. 26 of Bhattacharya's 1931 edition), and KJaN 11.15-16. 74. Tucci, Tibetan Painted Scrolls, 1:226-27, citing Tanjur 67, p. 34. The "histori- 85. Kalyāņ Šakti Ank, p. 640. cal" Darika-pā would have lived in Bengal during the 1oth c. reigns of Amrapala, 86. Sharma's commentary to RRS, pp. 39, 184-85; Misra, Ayurvedīya, p. 170; Hastipāla, and Kşāntipāla: Chattopadhyaya, Tarānātha's History, pp. 178, 311. Shorter Oxford Economic Atlas, p. 88. 75. See O'Flaherty, Siva, pp. 286-92, citing Mabābhārata 6.8.26; Bhāgavata Pur- 87. Dyczkowski, Canon, pp. 63-64; Schoterman, ȘSS, p. 73. ana 4.30.45, for the mythology of the submarine fire and the submarine mare. Cf. 88. Gorakb Bānī, Sabadī 142: bhag mușī byand, agani mușī pārā; jo rāsai so guru TÃ 8.98, which locates Vadava to the south of Jambudvīpa, between two mountains, hamara. This is a continuation of Sabadi 141, which is a discussion of vajroli and named Cakra and Mainaka. The southern sea in which the Indian submarine mare/ amarolī mudrās: bajarī karamtām amarī rāșai amarī karamtām bāi / bhog karamtām fire is found parallels the Scandinavian Müspellsheimr, the fire world in the South- je vyand rāșai te goras kā gurubhai. ern Hemisphere whose raging will be the end of the world: Puhvel, Comparative 89. In the more abstract schemata of yonipūja, the tantric worship of the female Mythology, p. 219. This is an alloform of the myth of the birth of Skanda and the sexual organ, the yoni is represented as a downturned triangle (Silburn, Kundalini, origin of mercury: see above, chap. 7, sec. 3. p. 43), at the heart of which is, once again, Kāmarūpa, the abode of the goddess 76. A goddess named Kundali, who appears to be associated with the left side of Kāmākhyā, who is identified with the kundalini and with feminine materiality the body (vamakbya) is mentioned in KJiN 17.23 and 20.11. (prakrti) in the form of menstrual flux (pusparūpini): Schoterman, Yonitantra, p. 24; 77. Siddha Siddbanta Samgraba 3.5, cited in Dvivedi, Nāth Sampraday, pp. Bharati, Tantric Tradition, pp. 259-60; Bhattacharya, History, p. 133; and Dyczkow- 187-88; Goraksa Paddbati 1.60; BhP 9.117-19; Gorakh Bani, p. 332. Cf. MBbT ski, "Kundalinī," p. 7, n. 13. See above, chap. 7, n. 67. 3.12-14. 90. In Assam, specifically at Kāmākhyā, whence he is said to have brought it to 78. HYP 4.17. Cf. HYP 4.108: "the yogin in samadhi is not consumed by time." Nepal: Karambelkar, "Matsyendranäth," p. 365. Cf. KJiN 16.46-49; and Jayara- 79. KJnN 2.2b-3: kālāgnirudrasamjnā tu nakhāgre nityasamsthitam/ yadā prajvalate tha's commentary to TA 1.6. (vol. I, p. 24). As their names indicate, Kāmākhyā and ürddbam sambārantu tadā bhavet/ badavāmukbamabattvanca pātāle sahasamsthitab. Kamarūpa are locations identified with eroticism. One account of Matsyendra's "That which is known as the Destroyer of the Fire of Time is forever located at birth has him born the son of Parvatī at Kāmākhyā: Nārada Purāna, Uttara Khaņda the tip[s] of the toenail[s]. It is when it flares upwards that the universal retraction 69.6, cited in NSC, p. 50. occurs. On the universal level [mahattvam] this is the mouth of the submarine mare, 91. When Matsyendra (Macchanda) is taken to be the founder of Kaulism, it is which is located in the underworld [ patale]." See also Bagchi's commentary, p. 43. his six sons who are noncelibate (adhoretas) who are singled out as qualified to be A late paper manuscript of the Goraksasahasranamastotra found by Bagchi in the revealers of the Kaula cult: Sanderson, "Saivism," p. 681. According to the Candrā- Nepal National Archives calls the land of Gorakhnath's birth "Badava": Bagchi, dityaparamāgama (cited in Bandhopadhyay, Natha Cult, p. 48), Yoganātha (an in- KJiN, p. 64- carnation of Siva) married a maiden named Surati, who was a manifestation of 8o. Thus a parallel passage from the ca. 12th c. Vināsikha Tantra (vv. 70b-71ab): Sakti. They had sixteen sons, beginning with Ādinātha. From them, six house-

474 475 Notes to Chapter Eigbt Notes to Chapter Eight

holding (i.e., noncelibate) sons were born: Ādinatha, Minanatha, Satyanatha, Sacet- "plantain forest." The alchemical BbP (4.108) seemingly evokes this myth when it anātha, Kapila, and Nānaka. Ten others-Giri, Puri, Vāratyādi, Saila, Nāga, Saras- states that a "barren woman born in the Plantain Forest (kadali vanajā vandbyā) vati, Rāmānandi, Šyāmānandi, Sukumāra, and Achyuta-left their homes and repeats the name of Mina in incantation (japayasi)" when a number of ingredients wander from place to place. See above, chap. 5, n. 95. are mixed together. 92. See Gorakb Banī Sabadi 177, on the loss of a yogin's semen as the shame of IO1. Cunninath, Tan-Prakas, p. 198. It is also located on the Malabar coast: the his guru; Gorakb Bāni Pad 48.1-3, on the female vulva as a tigress or vampiress. Cf. site of the old and important Nath monastery of Kadrī is identified with Kadalī: Kabīr, Doba 20.6 (in Vaudeville, Kabir Granthavali, p. 98). On the cleavage, within Narharinath, Nav Nath, p. 7. the Nath Siddhas, between the celibate and hathayogic "Gorakh" and noncelibate, 102. Dvivedi, Nath Sampradāy, p. 45. NSC, pp. 58-59. As a multiple of sixteen, erotico-mystical "Matsyendra" branches, see Bharati, Siddh-Sahity, p. 324. How- the figure of sixteen hundred, surely symbolic, is given in a number of sources: ever, Dvivedi (Nath Sampraday, p. 50) divides the Nath Siddha lineages between Cunninath, Tan-Prakāś, p. 199; Temple, Legends of the Panjab, vol. 2, p. 6 (v. 50). Pir the descendants of the "hathayogic" Gorakh and Matsyendra, and the "erotico- Premnath (Siva-Goraksa, p. 15) places the number at sixteen. Ratansen requests and mystical" Jālandhara and Kānipā. receives sixteen thousand padumini women in Padmavat 62.6. 93. Silburn, Kundalinī, p. 58. 103. Gorakh is conquering death with this feat, and, as such, he is specifically 94. On these works, see Zbavitel, Bengali Literature, pp. 189-90. This legend said to harrow hell to thrash Yama (the god of Death), who had presumed to take cycle is summarized in Sen, History of Bengali Literature, pp. 43-49; Dvivedi, Nāth his guru from him: Goraksa Vijaya pp. 45-48, cited in Dasgupta, Obscure, p. 222. Sampraday, pp. 43-46; and Cunninath, Tan-Prakāś, pp. 197-200. Other versions, 104. Temple, Legends, 2:21 (v. 208); Pancanand Mandal, ed. Gorakha Vijaya (Cal- which differ little from these accounts, are the Maharashtran YSA; the Dharma- cutta: Visvabharati Granthalaya, 1949), Gautam, Gorakbnāth Caritra, pp. 51-55; mangala of Sahadev Cakravarti (summarized in Dasgupta, Obscure, p. 368, n. I), and Dvivedi, Nath Sampradāy, p. 48. See below, nn. 174-78. the 20th c. Natha Caritra, an adaptation of three works from the Jodhpur library of 105. He must also return Matsyendra's soul to his original body: this he does Mãn Singh, the early 19th c. Marwari royal patron of the Nath Siddhas. There with the help of a yaksini, who has reconstituted the butchered body of the Nath, are also vernacular versions of this drama, in Nepali, Maithili, and other medieval and placed it atop Mount Kailash for safekeeping: Gautam, Goraknath Caritra, languages: Mukherji, Gopicandra Nātaka, pp. xxxvi, xliv. See above, chap. 5, n. 80 pp. 96-97. 95. Sen, History of Bengali Literature, p. 44- 106. Gorakba Vijaya, pp. 21-22, cited in Dasgupta, Obscure, p. 223. 96. Ibid., p. 45; NSC, pp. 95-97; and Pir Premnath, Śiva-Goraksa (New Delhi: 107. Gautam, Gorakbnāth Caritra, p. 62; Dvivedi, Nath Sampradāy, p. 46; Das- Vijnana Prakashan, 1982), p. 15. gupta, Obscure, p. 213; Srivastav, "Gorakb" Visesānk, p. 262. In another legend, Gor- 97. Summarized in Dvivedi, Nāth Sampradāy, pp. 48-49. akh kills and revives two other of Matsyendra's sons, the "Jain" Neminath and Pars- 98. Matsyendra's feat of parakāyapraveśa (and Gorakh's role) is also chronicled vanāth: Briggs, Gorakbnāth, pp. 190, 223; Rose, Glossary, 2:394. in the 15th-17th c. Sankara Digvijaya (9.80-84) of Madhava. On this feat, see also 108. Jan Gonda, Visnuism and Sivaism: A Comparison (London: Athlone, 1970), NSC, pp. 27, 57-59; and KJiN 10.26-8. The god Rudra is said to have taken the p. 112. In the case of the group studied in Bengal by Ralph Nicholas ("Ritual Hier- human form of Lakulīśa, the ca. 2d c. A.D. founder of the early Pāsupata sect by archy and Social Relations in Rural Bengal," Contributions to Indian Sociology, n.s. entering and reanimating a brahmin's corpse on a cremation ground: Sanderson, I [Dec. 1967], p. 69), the young men who do the swinging are "temporary re- "Śaivism," p. 664. nouncers" coming from the Mahisya caste, whose original caste name was Kaibar- 99. Alternatively, Matsyendra's queen dices up the body that the yogin has "left rta, i.e., the same as that of the fisherman Matsyendra in the KJnN (16.27-56) myth behind," so that he cannot return to it: Camanlal Gautam, Sri Gorakbnath Caritra of the recovery of the Kaula teachings from the Ocean of Milk: see above, n. 29. (Bareilly: Samskrti Samsthan, 1981), p. 62, 96-97; Dvivedi, Nātb Sampradāy, pp. 109. Dasgupta, Obscure, p. 378 n. 2. Dasgupta notes that the term found in the 48-49. In other sources, his lusty queen is named Paramilā, Mangalā, or Kamalā: Bengali Mangala-kāvyas is kalā-van. Here, a philological identification of plantains Cunninath, Tan-Prakāś, pp. 198-99. (kadala) with lunar digits (kala) is significant one: the full moon would be a bunch 100. Sen, History, pp. 45-46; NSC, pp. 27, 57-59, 61; Dvivedi, Nāth Sampradāy, of sixteen plantains. In a yogic context in which the moon in the cranial vault fills pp. 48-49. Pir Premnath (Siva-Goraksa, p. 15) simply calls this kadalī van, the with semen, the plantain would again bear a sexual significance.

476 477 Notes to Chapter Eight Notes to Chapter Eight

110. Sūnyasampādane 21.12 (vol. 5, p. 399). ho, Guru Maharaj ... your eyes are closed over there in Kajali Woods": Gold, III. Oddly, it is a rejection of Gorakhnath's body-oriented yogic practice that "Gender and Illusion," p. 118. the poet, Prabhudeva, is making here. See above, chap. 4, nn. 115-16, 167. 118. Yogavasistha 3.98-99, in Swami Venkatesananda, Vasistha's Yoga (Albany, 112. See above, chap. 4, nn. 163-64. N.Y., SUNY Press, 1993), pp. 119-20. 113. Mabābbārata 3.146.63-64, cited in Grierson and Dvivedi, Padumawāti, p. 119. On eating poison (visāhār), see Amaraughaśāsana, p. 8. On the thighs of a 253. See also Madeleine Biardeau, Histoires de poteaux: Variations autour de la Déesse woman as a boat to salvation, cf. the words of the Kāpālika Kapota in the Kālika bindoue, Publications de l'Ecole Française d'Extrême Orient, 154 (Paris: Ecole Purāņa (49.1-53.217) myth of Tārāvatī, Candrasekhara, Bhairava, and Vetāla, Française d'Extrême Orient, 1989), p. 286. In the epic, Bhīma encounters his "fa- translated in O'Flaherty, Siva, p. 206. ther," Hanumän, who is guarding the forest, a motif found in one of the legends of 120. With the awakening of the kundalini, the triangle of the downturned the Kadalī-rājya. Dvivedi (Nāth Sampradāy, pp. 53-54), who lists a number of pos- mouth reverses its direction to become an upturned mouth (ürddhavaktra): Silburn, sible locations for a historical Kingdom of Women which may have lain behind Kundalinī, pp. 27, 34-35 (plates), 173-74, 190; cf. Dyczkowski, Canon, p. 64, citing this motif, favors both Garhwal-Kumaon, in the Himalayan foothills of western TA 28.147. Uttar Pradesh, and Kāmarūpa, in Assam, as the most likely venues. 121. On this perennial image in Chinese symbolism of the microcosm and mac- 114. On these three variant readings, see Grierson and Dvivedi, Padumawāti, rocosm, see Rolf Stein, "Jardins en miniature d'Extrême Orient," Bulletin de l'École p. 250. Similarly, Bengali sources on Matsyendra's imprisonment give the variant Française d'Extrême Orient 52 (1942), pp. 100-4; and below, chap. Io, sec. 6. readings of kacali and kachar for kadali: Bagchi, introduction to KJiN, p. 17. On 122. Mababharata 12.255.8, cited in Malamoud, "La brique percee," in Cuire le Alexander's Zulmat, see Shirreff's translation of and notes to Padmavati 166 [12.5] monde, p. 90. and 528 [= 42.5] (pp. 91, 288-89); and Sharma, Jayasī Granthāvalī, p. 804. 123. Van Kooij, Worship of the Goddess, p. 42; Muller-Ortega, Triadic, p. 123; and 115. Kabīr, Sākbī 29.2 (in Vaudeville, Kabīr; p. 290). On Gopīcand, see Grierson Sanderson, "Mandala," pp. 169-76, especially 174-76. and Dvivedi, Padumawāti, pp. 251-52. See also Shirreff's translation of and notes 124. Brbadāranyaka Upanisad 4.2.2-3; 4-4.22; Chāndogya Upanisad 3.14. to Padmāvatī 166 [12.5] (p. 91). 2-3; 8.1.1-4; Maitrī Upanisad 7.2. The heart is also said to be the seat of the mind 116. On the Alha Khand, see Stuart H. Blackburn, Peter J. Claus, Joyce B. (manas) in RV 8.89.5. In the medical tradition, the heart is said to be the seat of Fluekiger, and Susan Wadley, eds., Oral Epics in India (Berkeley: University of Cali- consciousness and the site of the "eighth dhatu" of the body: this is ojas, the fluid fornia Press, 1989), pp. 197-202. Both the legends of Ratansen and of Alha concern that keeps the body alive, of which there are only eight drops: Caraka Sambita 12th c. Rajput princes facing the Turkish conquest of north India, and both hail 1.17.74-75; 1.30.4, 7. from the same Hindi-speaking heartland of north central India. The Gopicand 125. A number of tantric sources locate the ether or void in the heart. These legend flourishes in Bengal as well as across western India: a Rajasthani song cycle include the works of the Trika Kaula reformers (Muller-Ortega, Triadic, especially of Gopicand is translated and interpreted by Ann Grodzins Gold, A Carnival of pp. 142-46; and Padoux, Vac, pp. 28, 128 n. 117, 137 n. 140), in which the heart is Parting: The Tales of King Bharthari and King Gopi Chand as Sung and Told by Madbu the locus of the void/ether, because it is here that sound arises in its subtlest form. Natisar Nath of Ghatiyali, Rajasthan (Berkeley: University of California Press, 1992), Ether is the substrate or sound source of the primal vibration (HYP 4.101). pp. 159-310. Cf. idem., "Gender and Illusion in a Rajasthani Yogic Tradition," in 126. Silburn, Kundalini, p. 40 n. 7, citing Kabīr, Granthavalī 31.1-3. Tale, Text and Time: Interpreting South Asian Expressive Traditions, ed. Frank Korom, 127. Such Upanisadic identifications of the void with the absolute brabman Arjun Appadurai, and Margaret Mills (Philadelphia: University of Pennsylvania (kbam brabman: Brbadāranyaka Upanisad 5.1.1) may have facilitated this localiza- Press, 1990), pp. 102-35. A Punjabi version is given in Temple, Legends, vol. 2, pp. tion: the cranial vault is the locus of brabman, where the central channel (also called 1-77 (legend no. 18). the brabmamärga) culminates in the brabmarandhra. Abhinavagupta (TA 3.137-40) 117. On kajali ban in the legend of Püran Bhagat (who appears to borrow from terms the sivavyoman, the "heaven of Siva" (vyoman, however, also is a term for Kabir when he tells his mother to "loose his elephant in Kajali Van": see below, ether), as the highest of his metaphysical categories and the highest point in the chap. 9, n. 170), see Temple, Legends 2:426 (legend no. 34, "Püran Bhagat," v. 604). subtle body, identifying it with the seventeenth kala or thirty-seventh tattvā: Pa- In a Rajasthani folk cycle, Gopīcand cries out to his guru Jālandharanāth, "Oh ho doux, Vac, p. 91. See also ibid., pp. 95-96, 281, 424, on the connections between

478 479 Notes to Chapter Eight Notes to Chapter Eigbt

this metaphysical concept and mantric practice. Nath Siddha localizations of the dual form ("heaven and earth") that was primordially a singular: Monier-Williams, void in the cranial vault may be found in Gorakb Bani, Sabadi 1.18, 23, 45, 51, 55, Sanskrit-English Dictionary, s.v. "rodas." 176, 231, etc. 138. The full account is found in Temple, Legends, vol. 2, pp. 375-456 and 128. Goraksa Śataka 67. Detailed discussions of the many uses of the term sūnya Charles Swynnerton, Romantic Tales from the Panjab (Westminster: Archibald Con- in the medieval Hindu and Buddhist mysticism of the Bengali Siddhacaryas, Nath stable & Co., 1903), pp. 411-41. It is summarized in Briggs, Gorakhnāth, pp. 184- Siddhas, Vīraaivas, Maharashtran mystics, Kabīr, etc., are Vaudeville, Kabir Gran- 85, 197-98. The original myth of this sort may be the Rgvedic account of Trita, thavali, p. xx; Bharati, Siddh Sabity, pp. 336-69; and Shankar Gopal Tulpole, Mysti- who is thrown down a heavenly well (the dark of the moon?) by his treacherous cism in Medieval India (Wiesbaden: Harrassowitz, 1984), pp. 196-97. brothers. He performs a mock soma sacrifice at the bottom of the well, and the gods 129. gagan mamdal maim ūmdhā kūbā tahām amṛt kā bāsā/ sagurā hoī su bhari come and "flush" him out (the waxing of the moon?): RV 1.105.1-28; Jaiminīya bhari pīvai nigurā jāī piyāsā: Gorakb Bānī, Sabadī 23. On other uses of šūnya, cf. Brabmana 1.184; Mababbarata 9.35.3-51, in O'Flaherty, Tales of Sex and Violence, Sabadī 1, 46, 91, 176. Cf. Siva Sambitā 4.31. pp. 53-57. 130. RV 1.164.36; 8.69.3; Atharva Veda 10.8.9. Cf. Śatapatha Brābmaņa 1.6.4-5; 139. The well, called Puran's Well, is located in a village named Puranwāla, five 6.1.2.4; 7.1.1.10-16; 10.4.2.1; 11.2.5.3; 12.7.3.4. The head is compared to a down- miles outside of Sialkot, in the Punjab (Pakistan): Briggs, Gorakbnāth, pp. 98-99, turned vessel, filled with yasas that is identified with breath in Chāndogya Upanisad 185; and personal communication from Alain Wattelier (who visited the site in the 2.2.3. The classical study on the early history of such identifications is Jan Gonda, 1970s), Paris, June 1985. "Soma, Amrta and the Moon," in Change and Continuity, pp. 38-70. 140. Briggs, Gorakbnath, p. 194. Another Näth Siddha, Kanthadīnāth, also seals 131. Gyān Calisa 24-26, in Srivastav, "Gorakb Bāni" Visesank, p. 352. This work himself into a well to mediate: NSC, p. 22. is not included in Barthwal's edition of the Gorakb Bāni. 141. Tod, Annals and Antiquities, vol. 2, pp. 825-26; NSC 286; Sankarnath 132. Doha 5.45, in Vaudeville, Kabir Granthavali, p. 16. Yogisvar, Śrī Mastnāth carit (Śrī Mastnāth adbbut lilā prakās) (Delhi: Dehati Pus- 133. V. K. Sethi, ed. Kabir: The Weaver of God's Name (Beas, Punjab: Radha tak Bhandar, 1969), p. 136. Through the miracle of this well (related in chap. 20 of Soami Satsang Press, 1984), pp. 233, 461. Cf. Das, Kabīr Granthāvalī 14.40; "Chalu Sankarnath's work), Man Singh acceded to the throne of Marwar in 1804, and it is hamsa va deś." On Kabīr's language, see ibid., pp. 56, 199. a historical fact that he established his Nath Siddha miracle worker as his chief 134. Gorakb Banī, Pad 47.1-8, with Srivastav's commentary, pp. 302-5. Pad 47 minister: see below, chap. I1, sec. 3. is found on pp. 141-42 of Barthwal's edition of the Gorakb Bāni. 142. Briggs, Gorakbnāth, p. 188. 135. Verses 5-8 of Barthwal's edition read as follows: 143. Rose, Glossary, vol. 2, p. 394n. üjad sedā nagar majbārī tali gāgari ūpar panibārī 144. Stein, "Jardins en miniature," pp. 57-58, citing the Yun-ki ts'i-tsein (quoted magarī pari cūlbā dhūndbāi povaņbārā kau rorī kbāi without reference in the P'ei-wen yun-fou). The gourd doubles for one of the three kāmmini jalai agīthī tāpai bici baisamdar tharbar kāmpai "fields" or "wells of cinnabar" located in the Taoist body. Cf. idem, "Architecture ek ju radbiyā radbtī āi babū bivāī sāsū jāī et pensée en Extrême-Orient," Arts Asiatiques 4:3 (1957), pp. 176, 185. nagarī kau pānīm kūi āvai ulațī carcā goras gāvai. 145. On Vedic diksā, see Aitareya Brābmana 1.3. On Ayurvedic kuțīprāveśa, “en- (Gägari, the term employed here for water pot, is defined as a "small metal vessel": tering the hut," a three-chambered hut that is explicitly identified with the female Samksipt Hindi Sabd Sagar, s.v. "gagarā." reproductive organs, see above,chap. 2, nn. 47-48, and Caraka Sambitā 6.1.16: 136. Temple, Legends, vol. I, pp. 166-68 (vv. 513-38 of legend no. 6, "The Leg- "The sages knew of two sorts of rasayana: that of entering the hut, and [the practice] end of Gurū Gūgā"). Verses 536-37 read: "hāl dekho mere tan kã / phor do garhe of wind and sun (vātatāpi)." dīe mahāre." On dogbar[ā], the practice of carrying two superposed water pots on 146. Dalpatram Pranjivan Khakhar, "History of the Kanphatas of Kacch," Indian one's head, see v. 522, and note on p. 167. Antiquary 7 (February 1878), p. 47; Śrināth Tirthāvalī of Mān Singh (Churu, Rajas- 137. Sākbī 20.5 [= Doba 31.11], in Vaudeville, Kabīr, p. 262. Already in the than: n.p., 1951), vv. 5, 42, 283-85. Vedas, heaven and earth are viewed as two halves of a single whole, at once joined 147. Briggs, Gorakbnāth, p. 99; Krtyakalpataru of Lakșmīdhara, vol. 8, p. 84- together and held apart by a pillar (RV 1.160.4; 3.31.12-13, etc.): the term rodasī is a 148. Personal communication from Surya Kumar Yogi, Menal (Rajasthan),

480 48I Notes to Chapter Eight Notes to Chapter Eight

March 1985. For a general discussion, see MBbT 11.23-35. On the importance of 157. The medical and chemical name for mercury is hydrargyrum, from bydros wells at the pithas of Kamakhya and Hinglaj, see Van Kooij, Worsbip, p. 27; and + argentum, whence the symbol Hg in the periodic table; it is therefore possible Briggs, Gorakbnāth, p. 108. to abbreviate "wells of mercury" to "Hg Wells." It will be recalled that an author 149. RC 3.1; Misra, Āyurvedīya, p. 80. named H. G. Wells wrote a book entitled The Food of the Gods, an apt denomination 150. Such mineral theophanies are quite a commonplace at pithas of the God- for the mercury that alchemists ate. dess-or conversely, many sulfurous pools, in areas of geothermic activity, are con- 158. See above, chap. 7, sec. 3. sidered to be manifestations of the Goddess (Sharma's commentary to the Rasarat- 159. A yantra (from the root yam) is that which controls or subdues. In alchemy, nasamucchaya, pp. 10-1I; 184-85; 210-11). See above, chap. 7, nn. 53, 70-71. the term is applied to apparatus of the sort we are describing here, which control 151. Girnar is a site that is shared by Saiva Hindus and Jains alike. In the case such volatile elements as mercury and fire. For other uses of the term in tantrism, of this tirtha, the alchemical mythology is Jaina, and is found in chap. 4 of the see above, chap. 6, n. 93. Vividhatīrthakalpa, trans. John Cort, "Twelve Chapters from The Guidebook to Vari- ous Pilgrimage Places, the Vividhatirthakalpa of Jinaprabhasūri," in Phyllis Granoff, 160. The extraction of mercury from cinnabar, through the use of the vidyā- dhara or ürddhvapātana yantra is described in RRS 9.56. The sublimation of mer- ed., The Clever Adulteress and the Hungry Monk (New York: Mosaic Press, 1990), pp. cury, through amalgamation with copper and in combination with various plant 25-90. Cf. Granoff, "Jain Biographies of Nāgārjuna," pp. 48-49, on Jain mythology substances, also in the ūrddhvapātana yantra, is described in RRS 11.37. Similar of the alchemist Nagarjuna; and Rosu, "Alchemy and Sacred Geography," p. 156. instructions are found in AK 4-38-43, RHT 2.8, RA 10.55, and in every other major 152. The alchemical wonders of Srisailam are praised at great length in the Ra- alchemical work. The RA (7.49) also describes a procedure for the purification of sayāna Khanda [part 4] of the Rasaratnākara of Nityanātha. Chap. 8 of this section is entitled "Srīparvata Sādhana." Cf. Roșu, ibid., pp. 151-56, who questions the cinnabar in the same yantra: the resulting essence (sattva) of cinnabar has the ap- alchemical cachet of Srisailam as described in this and other sources. pearance of mercury (rasasankāśa). 161. RRS 9.24-25. In some sources, the upper pot is upturned and filled with 153. Devī Purāņa, cited in Laksmīdhara, Krtyakalpataru, vol. 8, p. 231; and Skanda Purāna 1.27.33-77; 1.29.87. The same pools, etc. are described in the pil- water, such that mercury condenses on the outside surface of its base: Misra, grim's guide I purchased while on the Kedarnäth pilgrimage in May 1984: Śrī Ked- Ayurvedīya, p. 225. 162. RRS 9.57. ārnāth Mābātmya (Yatrā Gaid, ed. Shersingh Shah (Kedarnath: Shersingh Shah, 163. Rasendracintāmaņi 8.3, cited in Prakash, Prācīn Bhārat, p. 575. n.d.), pp. 41-43. 164. Sharma, identifying the ürddhvapātana yantra with the vidyādbara in his 154. Sastri, Agneyatirtha Hinglaj, p. 54. See above, chap. 7, nn. 70, 120. Given the fact that the Ganges and the Yamuna, identified with the ida and pingala chan- commentary to RRS 1.88 (p. 14), cites the Rasapaddhati. This apparatus is also de- scribed in RC 4-42; 5-51-52. nels of the subtle body, rise out of Himalayan glaciers to the east and west of Kedār 165. Amanaskayoga 2.15: ürddhvamustiradhodrstir ūrddhvabhedastvadhah sirāḥ nāth respectively; and that a "Sarasvatī Gangā" (i.e., susumņā) appears near Ke- dārnāth (Kalyāņ Śakti Ank, p. 671), the subtle physiology of the nādīs is also /dharāyantravidhānena jīvanmukta bhavisyati. The šāmbhavī mudrā is a Kaula technique which would appear to have some connection to the Kubjika cult of the geographically reproduced here. The sixth cakra, the ājña, located between the eye- Western Transmission, in which a "masculinized" form of practice is called sām- brows at that place at which the three major nadis are said to meet, is moreover bbava ("pertaining to Sambhu, Siva"): Sanderson, "Saivism," p. 687. Śāmbbavī-vidyā called Kedar in HYP 3.24. The Nine Naths and "Kedar" are identified with the ten is the subject of the Jñānasankālinītantra (HTSL, p. 102); and sāmbhavī dīksā, a form doors in Gorakb Bāni Pad 9.1. See above, sec. 2b. of initiation particular to the Nath Siddhas (Bharati, Tantric Tradition, pp. 90-91). 155. Śrīnāth Tirthāvalī, vv. 267-72. A similar image, that of hitting a brass cup Śambbavī is also synonymous with the medial channel, the susumna, according to atop seven bamboo poles, is found in the "Marriage of Sakhī Sarwar," legend no. HYP 3.4; and the hathayogic sambbavi mudra is nearly identical with the renowned 22 in Temple, Legends, vol. 2, p. 127. kbecarī mudrā (HYP 4.38-39). On this, see below, sec. 4 of this chapter. 156. Stein, Grottes-Matrices, pp. 37-43. See also ibid., pp. 15-23 for a survey of 166. Goraksa Śataka, v. 62. Other verses (131a, 138c) speak of a pool (dhāra) of Hindu "womb-caves" throughout the Indian subcontinent, including those of lunar nectar in the cranial vault, which the yogin is to drink, lest it fall into the sun Hińglāj and Kāmākhyā. in the lower abdomen.

482 483 Notes to Chapter Eight Notes to Chapter Eight

167. The Yonitantra passage (68a) is translated and discussed by Schoterman on is told in the Tarikb-i hasan of Pir Ghulam Hasan: Jayalal Kaul, Lal Ded (New pp. 12-13 of his edition of the text. Elsewhere, the KCM (fol. 24a9-24b.1) states Delhi: Sahitya Akademi, 1973), p. 14. I am grateful to Patricia Greer for sharing that "by using [a device] composed of dbarā, one attains immortality" (dharāmāyā this reference. pra[yo]gena cāmaratvāpnuyāt). This verse of the Amanaskayoga (ed. Yognath Swami 172. Henri Maspero, Le Taoisme et les religions chinoises, with a preface by Maxime [Poona: Siddh Sahity Samsodhan Prakashan Mandal, 1967]) is singled out by the Kaltenmark (Paris: Gallimard, 1971), p. 492 and n. 6. Cf. Kristofer Schipper, Le great tantric scholar Gopinath Kaviraj (foreword, p. 22) as the most singular por- corps taoïste (Paris: Fayard, 1982), pp. 137-53. The Palace of the Cinnabar Field is tion, if not the very heart of this text (editor's preface, pp. 10-11), and the editor also termed Nihan, the ideogram for which was, in this early period, the same as and Hindi commentator discusses his difficulty in making sense of it (pp. 11-14). that for the nirvana of import Buddhism. Also qualified as the residence of the He refers to the Mababbarata (1.179.14-17) account of Arjuna's winning of Drau- Highest One (Maspero, Le Taoisme, p. 493), its resemblance to the bráhman of Hin- padī, in which the epic hero pierces the target (vivyādba laksyam) (called "piercing duism and the sunya of Siddha traditions is equally striking. the fish" [matsyavedha] in Hindi) placed atop a pole with an arrow shot from the 173. In alchemy as well, there is an apparatus properly known as the damaru bow Siva has previously given him. This he does by looking downward (adbodrstib) yantra. This apparatus is nearly identical to and may be used interchangeably with into a pool of water (dharayantra) as he clenches his fist around his upward-held the urddhvapatana: see RRS 9.57 with Sharma's commentary, p. 143; and Rasabrdaya bow (ürddhvamusti): his head is turned downwards (adbab sirab: but the text gives Tantra, introduction, p. xxiii. sirab) as his target (bbeda) is above. His "piercing of the fish" is at once the hitting 174. The location of a mystic damaru in the head would appear to refer to this of his target, the winning of Draupadi, an inward-looking posture of yoga, or a tantric schema: Bhūtaśuddbi Tantra 3.3, in Tantrasamgraba, 4 vols., ed. Gopinath Kav- reverse technique of sexual intercourse. This interpretation has the merit of pre- iraj and Ramaprasad Tripathi (Benares: Sampurnanand Sanskrit Visvavidyalaya, serving the orthography dhara, whereas my interpretation requires a reading of 1973-81), vol. 4, p. 313. See below, chap. Io, sec. 6. -dbara. 175. Literally, "upward-tending semen," this term is most often translated as 168. RRS 11.38. "celibacy, chastity." The sense, however, was that the ascetic is holding his semen 169. RRS 9.9; 11.39. Cf. Dash, Alchemy, pp. 58-62, on the identification of the up and away from his penis, from which it might all too easily slip away from him. various -patanas as variant processes, all of which make use of the vidyadbara Indian popular tradition is filled with accounts of "semen-headed yogis," a reflec- yantra. tion of this interpretation: G. Morris Carstairs, The Twice Born (London: Hogarth 170. Goraksa Śataka 135. Verses 133-34 read: "In the region of the navel dwells Press, 1958), p. 86. the lonely sun, whose essence is fire; located at the base of the palate is the eternal 176. Four such drums (catalogue numbers R.VII-230, R.XII-835, R.XII-836a, moon, whose essence is nectar. That which rains down from the downturned and R.XII-798a) are presently housed in the National Museum of Denmark in Co- mouth of the moon is swallowed by the upturned mouth of the sun. The practice penhagen. I am grateful to Rolf Gilberg, curator of this collection at the National [of viparītakarana] is to be performed as a means to obtaining the nectar [which Museum, for his ready cooperation in supplying me with this information. would otherwise be lost]." This passage is also found in Gorakhnath's Yogamar- 177. This description of the Tibetan damaru, called the "skull drum" (thod-rna), tanda, vv. 121-122a, 123b. This posture is described in both the HYP (3.77-79) and is found in Mireille Helffer and Marc Gaborieau, "A propos d'un tambour du Ku- Goraksa Paddhati (pp. 48-49), immediately before their respective treatments of the maon et de l'ouest du Népal: Remarques sur l'utilisation des tambours-sabliers vajrolī mudrā. It is dramatized in a legend from Dhinodar, in Kacch (Gujarat), de- dans le monde indien, le Népal et le Tibet," in Studia instrumentorum musicae popu- scribed above, chap. 5, n. 204. lares, Festschrift to E. Emsheimer on the occasion of his 70th birthday (Stockholm: 171. Gorakb Bani, appendix I to Barthwal's edition, p. 242. The full verse reads: Musikhistorika Museet, 1974), p. 78 and pl. 13. ulați yantr dhare sişar āsaņ kare/ ... silahaț madhye kāmvarū jītale/ nirmal dhuni 178. Ibid., p. 75. gagan mämhī. The upside-down yantra here is the (normally) downturned triangle 179. Ibid., p. 78. of the inner yoni, located in the müladbara cakra, here identified with the old Bud- 80. Personal communication from Sangye Sonam, a Tibetan monk at Chatral dhist pītha named Sirihatta, at the center of which is the pītha named Kāmarūpa. Rinpoche's monastery, Pharphing, Nepal, May 1993. I am grateful to Peter Moran, For a discussion of this element of the subtle body, see above, n. 170. Lalla's story who served as my interpreter in this interview.

484 485 Notes to Chapter Eight Notes to Chapter Eight

181. Caryapada 3, with the commentary of Munidatta and the translation (com- An Anthropological Exploration (Varanasi: N. K. Bose Memorial Foundation, 1978), piled in "Yambu," in Nepal) of Kīrticandra, reproduced in Bagchi, "Some Aspects PP. 93-94. of Buddhist Mysticism in the Caryāpadas," Studies, pp. 75, 79, 84-85. I have en- 194. This echoes a statement from the Yogavisaya attributed to Matsyendranātha hanced and "Hinduized" Bagchi's treatment of this passage. (v. 24 in Mallik, ed. SSP, p. 47): "When the nada is heard, the sankhini showers 182. Cf. HYP 3.47 which terms a homologous process "drinking the immortal the body with nectar." The sankbini runs from a point behind the forehead to the liquor" (pibedamaravāruņīm). throat: Silburn, Kundalini, p. 130. Drinking nectar is a commonplace of Nāth mys- 183. See below, sec. 4, for an extended discussion of the "tenth door" of the tic poetry: Gorakb Bānī Sabadī 67, 171; SSP 6.64- subtle body. 195. Amaraughašāsana, p. 7. 184. Yogamartanda 177: "When the upbreath disappears and the mind is ab- 196. Silburn (Kundalini, p. 131) and Dasgupta (Obscure, pp. 239-42 and notes) sorbed, that is then called the unity of samarasa, samādbi." assume there to be but one "tenth door," an impossibility given the descriptions of 185. Goraksa Paddhati 1.68-74. Schoterman (Yonitantra, p. 29) suggests that the this door, which is to be variously opened and closed in different textual descrip- vajrolī mudrā may have been used in lieu of the kbecarī mudra in some cases: instead tions. The mouth of the sankhini is called the tenth door in Gorakhnath's Amar- of maintaining the combined bindus in the head through the latter, the practitioner aughaśāsana (p. 11), and SSP 2.6. The Nine Naths and "Kedar" are identified with would recover the same fluids through urethral suction and then simply drink the ten doors in Gorakh Bani Pad 9.1; the yogin is enjoined to lock this door in them. Pada 23.1, but to "blast it open" by closing the other nine in Atmabodh (in "Gorakh 186. Khecaricakra refers to an elaborate mandala of sixty-four yoginis and to the Bāni" Visesānk, p. 340). In the legend of Pūran Bhagat, the evil stepmother Lūņā is circle in which täntrikas engage in ritualized sex. On this, see Dyczkowski, "Kuņ- identified as the tenth door of the royal palace: Temple, Legends, 2:398 (vv. 265-66); dalini," p. 6; BbP 1.77; RA 3.17, 20; and Vidya Dehejia, Yoginī Cult and Temples: A in the Padmävat, it is the tenth door by which the fortress of Lanka (the body) must Tantric Tradition (New Delhi: National Museum, 1986), pp. 44-45. The contents be stormed: see below, sec. 5 and n. 243; and chap. 9, sec. 7. of the KhV have been discussed in detail: see above, chap. 5, sec. 5m. 197. The clearest discussion of this doubling of the tenth door is found in Kavi- 187. Maitrī Upanisad (6.19): "By pressing the tip of his tongue against the palate raj, Aspects, p. 237. On the sankbini's two mouths, see Mandal, ed., Gorakba Vijaya, ... one sees brabman through contemplation." Another early reference is found in p. 141, cited in Dasgupta, Obscure, p. 240. The Nath Siddhas generally identify the KJnN 6.18-28. Cf. Goraksa Śataka 67; Goraksa Paddhati 1.68; Yogamārtanda 58; Gor- tenth door with the brabmarandhra, by which the fontanelle is intended: Gorakb akb Bānī Sabadī 133; HYP 3.41-42. Bănī Sabad 135 (in which the yogin is said to throw a "door panel" [kapāt] across 188. Kakar, Shamans, p. 203. See Bharati, Tantric Tradition, pp. 242-43, for a this opening). The lower mouth is located in the tälu-cakra, the sixth cakra in the general discussion of the multiple meanings of this term. nine-cakra system which expands upon the more conventional system of 6 + 1: 189. Atharva Veda 10.8.43; Śvetāśvatara Upanisad 3.18. The yogin seals himself SSP 2.6. Cf. Gorakb Bāni Pad 23.1, with commentary; and Atmabodh 2 (in Srivastav, off doubly from the outside by enclosing himself in a windowless room to meditate: "Gorakb Bāni" Visesānk, p. 340). Yogatattva Upanisad 32, in Varenne, Upanishads du yoga, p. 74. This recalls as well 198. Amaraugbaśāsana, discussed in Silburn, Kundalinī, p. 129: she is so called the initiation hut of Vedic diksa and the Ayurvedic practice of kutīpravesa. See because when she sleeps, she is coiled three and one half times around the linga above, chap. 2, nn. 47-48. located in the abdominal region. The SSP (1.67), also apparently identifying san- 190. Kaviraj, Aspects of Indian Thought, pp. 236-37; Dasgupta, History, 2:253. kbinī with kundalini, states that the sankbini flows in the urethra (lingadvāra) and via 191. Amaraughaśāsana, pp. 10-11, discussed in Silburn, Kundalinī, pp. 131-32; a straight path up into the brabmarandbra and thence into all of the ten doors. Dasgupta, Obscure, pp. 239-42, citing Gorakha Vijaya, ed. Mandal, pp. 141, 143-44- 199. Gorakb Bānī Sabadī 219a. Identical to this is verse 5 of "Mahādev jī kī Cf. Yogavisaya of Matsyendranath, vv. 30-31, in Mallik, ed. SSP, p. 47 sabadī" (Dvivedi, Nāth Siddbom, p. 66). The same concept is found in the SSS (chap. 192. In order to do so, he has already cut away the frenum, which anchors the 42, in Schoterman, SSS, p. 87, cited in Dyczkowski, Canon, p. 65), which states tongue to the lower palate, through a technique described in HYP 3.32. that the goal of the practitioner is "the union of the two [the highest and the lowest] 193. On the Jūna Ākhādā, see Rajesh Bedi, Sadhus: The Holy Men at India (New mouths," identified with the highest and lowest of the seven cakras, as well as with Delhi: Brijbasi Printers, 1991), p. 79; S. Sinha and B. Saraswati, Ascetics of Kasbi: the upper and lower yonis of the Sākta subtle body.

486 487 Notes to Chapter Eight Notes to Chapter Eigbt

200. Here, a "crooked" goddess may be easily assimilated to a curved duct, and major alchemical work. Their construction and use are summarized in Dash, Al- one of the synonyms for both the sankbini and the kundalinī is banka nala, "curved chemy, pp. 197-99. duct." This term also has an alchemical application: this is a curved tube which, 211. Silburn, Kundalinī, pp. 169, 192, in which sampuțīkarana is an exchange of tapered at one end, is used as a blowpipe for ventilating a flame. Yet, the 14th c. male and female energies that occurs through, but is not limited to, the act of sexual Gorakb Bodb (v. 62 in Barthwal's numbering, Gorakb Bāni, p. 186) distinguishes be- intercourse. The same term applies to the telescoping of elements back into their tween susumnā, śankbinī, and banka nala. On the banka nāla of the subtle body, see source, in the ascension of the kundalini or in the distillation of the 51 phonemes also Yogavisaya v. 22 (in SSP, p. 46) and Gorakb Bani Pad 53.3. The alchemical of the Sanskrit language into the bisyllabic abam, the universal I (ibid., p. 9). The banka nāla is described in RA 4.58 and RRS 10.45. Cf. Misra, Ayurvedīya, pp. term samputa means "skull" or "cranium" in Hindi: Samksipt Hindi Sabd Sāgar, s.v. 147-48. "sampuța." 201. For the western linga in the head, see Gorakh Bani Sabadi 187a and the Raja 212. Kaviraj, Bhāratīy 1:276, 282, 287-89; KM (London, Wellcome MSS no. Ranī Sambad of Gopīcand, v. 105 (in Nath Siddbom, p. 9). On the "microcosmic g501) fol. 35a.7-8, describing the kaulaketu ritual of the Siddha Mārga. west," see also Gorakh Bani, Sabadi 41 and 267. The KJnN (17.23, 20.11) also asso- 213. Tucci, "Animadversiones Indicae," pp. 154-55. This is a portion of a text ciates a coiled (kundaläkrti, kundali) goddess with the left side. On the lower west- describing a tantric Buddhist initiation. Cf. KM (London, Wellcome MSS no. ern linga, see Goraksa Śataka 19. Cf. SSP 2.2, which states that the kundalini sleeps g501) fol. 35a.7-8; KJñN 18.22-23; and KPT (in Bhattacarya, ed. Indrajāla vidyā- with her mouth over the pascima-linga, located in the second cakra, the svādbisthāna. samgraba, p. 265) for similar language. It is out of this vortex that seed is emitted in ejaculation: Srivastav, "Gorakh" Vis- 214. Kaviraj, Bhāratīy, 2:293 n. 1: "The mouth of the sankbinī nādī is, in tantric esānk, p. 131. treatises, precisely the mouth of the guru; it is otherwise designated as Siva's upper 202. MBbT 7.18; Śrīnātbakatbāsāra of Dwārkanāth, ed Narharināth Yogi (Be- mouth. It is from this site that the teaching of the [initiatory] tāraka mantra is nares: n.p. 1951), p. 23. transmitted into the disciple's right ear." 203. Dyczkowski, "Kuņdalinī," p. 3 n. 7. 215. The Sarada Tilaka (5.139) has the guru transfer "that which has oozed in- 204. Yogabīja 135, 148, 159. Cf. Kaviraj, Bhāratīy, vol. 2, p. 285. side his mouth" (vidruvakträntara) into the mouth of his disciple. Often, the guru 205. TA 29.124-29. On the female sexual organ as the yoginivaktra, the "mouth is said to directly penetrate the heart of his disciple with the divine energy he has of the yogini," see Tantrāloka Varttika, vol. 2, p. 104, cited in Padoux, Vāc, p. 61. stored up inside himself. On this notion, and the similarities and differences be- 206. Vamamārga (in Hindi), by Pandit Vanisidhar Sukul Vaidyaraj (Allahabad: tween tantric initiation between males and ritualized sex between a male and a Kalyan Mandir, 1951), v. 110, English translation in Bharati, Tantric Tradition, p. female, see Silburn, Kundalinī, pp. 175-76. 264. The Hindi reads (ibid.,p. 277): "phir śakti ko gaurī kī apne ko siv kī bhāvanā 216. Goraksa Paddbati, appendix to 1.62 [p. 32] and 2.16; HYP 3.12, 27. kar .. . mātrmukh mem pitrmukh arpit kare." 217. Rudrayāmala Tantra 17.73, 85-87. 207. Silburn, Kundalini, p. 190, citing TA 29.124-26, with the commentary of 218. Khecari mudra is also the term employed by a Nath suborder, the Gūdaras, Jayaratha; and Dyczkowski, Canon, p. 64, citing TA 28.147. For further discussion, for the cylinders of wood which they pass through their earlobes. The Gūdaras see Silburn, Kundalinī, pp. 27, 173-74, and plates on pp. 34-35- were founded by a member of the Dasnami sect and are not generally considered 208. Manthānabbairava Tantra, Kumārikākhanda 13.110-43, summarized in by Näth sampradäyins to be one of their own: Briggs, Gorakhnäth, p. 11. Given the Dyczkowski, "Kundalinī," pp. 17-18 n. 27, who remarks that "although one can fact that the piercing of the ears opens a channel without which yogic practice distinguish two Yonis, there is in fact only one." cannot be effective, the term is an apt one: on this, see below, chap. Io, sec. 4. 209. RA 11.17-18, 29-41; RRS 8.75-79. The image may be based on moveable 219. RA 11.151, 154, 162-63, 219. Siva lingas, over which "mouthed" sheaths could be superimposed: Bagchi, "Fur- 220. On gutikā bandha, see Gorakb Bānī Sabadī 49; on the kbecarī and other al- ther Notes on Tantrik Texts Studied in Ancient Kambuja," Studies, p. 21. chemical gutikas, see RA 12.336-37, RHT 19.65-76; RRS 11.88; and Brahmānan- 210. On the term mudra in alchemical terminology, see Sharpe, Eight-Hundred da's commentary to HYP 4.27. On the siddhi gutikā, see Padmāvat 222 [= 23.1] (p. Year Old Book, p. 106. Putas are discussed in RA 6.101-22; RRS 10.47-64, and every 482 of Grierson and Dvivedi, Padumawāti; and p. 135 of Shirreff, Padmāvati). On

488 48g Notes to Chapter Eight Notes to Chapter Eight

kbecari mercury, see RA 2.89; 3.9, 17-20; 11.98-107, 149-53, 162-63; 12.380; on devoted to the Goddess's tour of her footstools, across an Indian subcontinent that khecarī śakti, see KJnN 5.4; 6.18-28; 7.14; 20.10; and Dyczkowski, Canon, p. 168 is situated, in its entirety, within her body: Schoterman, ȘSS, p. 148. n. 52. A folktale from Benares casts Gorakhnath in the miraculous recovery of 233. For the alchemical lore relative to Hinglāj and Kāmākhyā, see above, chap. thousands of such gutikās: personal communication from Siddhinandan Misra, Be- 7 nn. 65-70. Kedār has been discussed above, n. 153. Alchemical descriptions of nares, March 1985 Srisailam are found in RRÃ 4.8.1-185. The RA description of Kailash is in 1.2-3. 221. In the RRS (6.13-14), there is a "division of labor" between the practical 234. On the symbolism of the Nine Näths, see Dasgupta, Obscure, p. 206, citing and the spiritual components of the alchemist's craft. On the one hand, the alche- the Goraksa Siddbanta Samgraba, pp. 44-45; the Tantrarāja Tantra (2.2-5, in Wood- mist carries out his chemical operations in a laboratory (rasasala); on the other, roffe, Tantrarāja Tantra, p. 19); and Schoterman, ȘSS, pp. 31-39. See above, chap. he effects his daily worship of the alchemical gods in a temple (rasamandapa) built 4, nn. 53-60. nearby. 235. The Gorakb Bodb, a vernacular work cast in the form of a conversation 222. The RRS (6.21) says the rasalinga is to be located on the eastern side of the between Gorakhnäth and his guru Matsyendranäth, is a veritable "skeleton key" to alchemical altar (itself located at the center of the rasamandapa). The Chandi Bukit the "intentional language" of the Nath Siddhas concerning the bodily microcosm. Batu Pahat (Malaysia) alchemical temple site studied by Francis Treloar ("Use of Cf. the Prän Sankali of Caurangīnāth, vv. 206-341 (in Dvivedi, Nāth Siddhom, pp. Mercury," pp. 232-40) has its rasalinga at the center of the structure. The rasaman- 19-28); and in Sanskrit, the third chapter of Gorakhnath's SSP. dapa described in the RA (2.52) also has the rasalinga at its center. 236. Temple, Legends, 2:441; Swynnerton, Romantic Tales, pp. 426-33. 223. On this mandala in the RA and RC, see above, chap. 6, sec. 2. The RC 237. Dvivedi, Nath Sampraday, pp. 185-86; Grierson and Dvivedi, Padumawāti, mandala is graphically reproduced in Roșu, "Mantra et Yantra," p. 252 (fig. 5). A p. 251-52. Jayasī himself speaks of Simhala and Lanka as two different places (Shir- simpler mandala is described in Misra, Ayurvedīya, p. 83. reff's notes to Padmavati 2.16), even if he clearly knows Simhala or Sangaldīp to be 224. See above, chap. 7, sec. 3 and nn. 56-59. an island situated to the south of India. There is, however, much more of fairyland 225. RA 1.43; 2.52; RRS 1.23-28; 6.14-23; MBbT 8.1-10; RC 1.37-38. See also than hard geography to Jayasi's Simhala, just as there is to the Kingdom of Women Treloar, "Use of Mercury," especially pp. 237-38; and Siddhinandan Misra's intro- into which Matsyendra falls in the myths we related above. duction to the Rasendracudamani, p. 5, in which he discusses the alchemical grounds 238. It is Püran Bhagat (Cauranginath) himself who claims Matsyendra for his for the foundation and installation of the rasalinga. guru: Prān Sankali vv. 207, 221, 307, in NSB, pp. 19, 20, 26. On the localization of 226. Misra, Ayurvediya, p. 85. In his description of the Chandi Bukit Batu Pahat Matsyendra's "Plaintain Forest" in Sri Lanka, see Cunninath, Tan-Prakāś, p. 198; alchemical temple ruin, Treloar states that the mercurial linga unearthed there was Temple, Legends, 2:19 (v. 194); and above, n. 101. Gorakb Bānī Pad 1.1-4 presents set in a silver semicircle: Treloar, "Use of Mercury," pp. 237-38. Matsyendra's imprisonment in the Kingdom of Women as the result of his Kaula 227. RA 2.77-82. The Vimalaprabba commentary to the Buddhist Kala- propensities. The text of the poem consists of Gorakh's arguments against sexual cakra Tantra (2.213-15) presents a similar iconographic representation of this wor- intercourse as a component of tantric practice. ship support, for use in the alchemical Jambhala rite: Fenner, Rasāyan Siddhi, pp. 239. Padmāvat 219-22 [= 22.8-10; 23.1], in Grierson and Dvivedi, Padumawāti, 160-63. pp. 478-82; Sharma, Jayasī Granthāvali, pp. 347-55; and Shirreff, Padmāvatī, pp. 228. RA 3.63-73; RRS 6.22-23. For the iconography of these images, see above, 134-35. A similar description is in 2.16-18. This translation relies to some extent chap. 6, nn. 61-63. on Shirreff's translation; it relies more heavily, however, on Dvivedi's and Sharma's 229. See below, chap. 9, n. 23. superior Hindi tika and commentary. Words in parentheses correspond to Shirreff's 230. RHT 1.12. See above, chap. 7 sec. 5. annotations; words in square brackets are my own additions. 231. Bhattacharya, History, pp. 376-78. 240. The six cakras and the sabasrāra: see above, chap. 2, n. 95. Similar imagery 232. The classic source on the pithas, which includes a critical edition of the is found in such popular traditions as the legend of the marriage of a Muslim fakir 16th c. Bengali Pithanirnaya, is Sircar, Sākta Pīthas, passim; for a lively rejoinder, named Sakhi Sarwar. In a preliminary test, Sakhī Sarwar is made to hit with an however, see Pal, Hindu Religion, pp. 24-27. The Nath Siddha reckoning is found arrow a brass cup that has been perched atop seven superimposed bamboo poles: in Gorakh Banī Sabadī 163. Much of the latter part of the $SS (especially 4.1-11) is "Marriage of Sakhī Sarwar," legend no. 22 in Temple, Legends, vol. 2, p. 127.

490 491 Notes to Chapter Nine Notes to Chapter Nine

241. Hath, the modern Indo-Aryan form of batha, as in batha yoga. Fall, through the sexual union of Adam and Eve: Mircea Eliade, The Two and the 242. Cinbe (pl.), "marks," "signs," are the emblems a yogin wears (earrings, horn, One, trans. J. M. Cohen (Chicago: University of Chicago Press, 1979), p. 104. wallet, etc .: see chap. 9, nn. 82-84) which are so many marks of his initiation into 2. RA 17.165a; 1.18ab. Cf. RU 15.50 (cited in Misra, Ayurvedīya, p. 31 n. 2), a given-in this case the Nath Siddha-order. which describes the knower of mercury (rasavid) and the knower of the soul (ät- 243. The nine gates and the tenth door are the bodily orifices and the inner mavid) as the two "subtle seers" (sūksmadarsinau). One may choose to see a precur- brahmarandbra or śankhinī nādi (also known as the banka nāla); the five constables sor to this twofold method in the Caraka Sambita's (6.1.16) reference to the two are the five breaths: see above, chap. 2, n. 29. complementary elixir therapies (rasāyanas), i.e., kuțīpraveśa and vātātapika, "[the 244. Bank[i], an alloform of banka, as in the bent tube, the banka nala, another practice of] wind and sun": Roșu, "Consideration," p. 7. On kutīpraveśa, see above, name for the śankbini nadi, associated with the dasama dvāra, the tenth door, located chap. 2, nn. 47-48. in the head. The reference to the body as "bent" or "crooked" at the beginning of 3. ausadhyo 'dhyātmakaśceti rājayogo dvidhā kvacit / hatho 'pi dvividhaḥ kvāpi this description would also refer to the crooked body of the sankbini, which doubles vāyubindunișevanāt. Cf. RA 4.23, which calls for the conjoined use of botanicals as the kundalini. See above, n. 200. and mantras. 245. Bhed, "penetrate" has the extended meaning of "secret," as such are gener- 4. Sabara's commentary to Jaiminīya Sūtra 3.1.3 and Tantravārttika, p. 1078, ally impenetrable until one has a clue with which to unlock them. This is the stan- cited in Pandey, Hindu Samskāras, p. 16, who also gives many specialized uses of dard hathayogic term used for the "piercing" of the cakras by the kundalini. In al- the term. Cf. Caraka Sambitā 1.26.30. See also Kapani, Notion de samskāra, passim. chemical parlance, this term comes to be closely identified with vedb, "pierce, 5. See RA 1.36 on sūta; RRS 1.79-80 on divine jealousy as ground for the al- penetrate," which is the term employed to signify transmutation, in which mercury chemical samskāras; and Pandeya's introduction to the BhP (Goraksa Sambitā, vol. is said to penetrate and thereby transform base metals into gold and mortals into 2, p. gba), for a commentary. See Manu Smrti 2.27 and Yajñavalkya Smrti 1.13 on immortals. the rites of passage. Cf. Kapani's discussion: Notion de samskāra, vol. I, pp. 89-94- 246. Cf. the "pools" in the ascent of Kedārnäth (above, sec. 3, nn. 109-11); and See above, chap. 7, sec. 2, on origin myths of metals, in which gold is accompanied the channeling of the Ganges associated with matsyodari. The path rising by various metallic "afterbirths." out of the pool at the base of the fortress (the mūladhara cakra, where the kundalini 6. The figure of eighteen alchemical samskaras is reached by taking the classic sleeps) is clearly the medial susumņā nādī: see above, n. 153. number of sixteen samskāras (cf. Siddhinandan Misra's commentary to the discus- 247. Cf. the discussion of the reverse (viparita) yogic posture and the alchemical sion of the samskäras in the RC preceding 4.85, on p. 54.) and adding to these the vidyadbara and other yantras: above, nn. 120-21. resulting processes of transmutation and bodily transubstantiation. The Vaikbānasa 248. The idea here is that one must dive down into the pool at the base of the Smārta Sūtra is a source that lists eighteen life-cycle samskāras: Kapani, Notion fortress (the müladhara) in order to gain access to the path (the susumnā nādi) that de samskāra, vol. 1, p. 86; for a figure of sixteen, see Mrgendrāgama, Kriyā 6.9-11 rises out of its depths. As I show in chap. 9, sec. 46, such can only be effected by (cited in Gonda, Medieval, p. 171). See above, chap. 2, sec. 4. restraining or binding (bandb) mind, breath, and seed. 7. The eighteen samskāras are listed in RHT 2.1-2. Succinct descriptions are 249. See above, n. 220. found in RC 4.85-106. This material is copied, for the most part, by the RRS (8.62- 250. At once a reference to the opening of the lotuses that are the cakras when 88), which also lists them in 11.15-16. A synoptic table of the samskāras, taken they are pierced and the initiatory role of the padmini, the Lotus Maiden, in tantric from fifteen major alchemical sources, is given in Misra, Ayurvediya, pp. 212-13. initiation: see below, chap. 10, sec. 2. An adequate English-language description of the mechanics of the eighteen sam- skāras is given by Subbarayappa, "Chemical Practice," in Bose, Concise History, pp. Chapter Nine 320-22. Subbarayappa's informant was Dr. Damodar Joshi, of the Rasasāstra De- 1. Padoux, Vac, pp. 82, 124-26; Silburn, Kundalinī, pp. 19-24, 52-53, 64-66. partment of the Ayurvedic College of Benares Hindu University. Dr. Joshi was also The human body is the turning point of another important medieval metaphysical my principal informant on such matters, during my stay in Benares, from August system. This is the Neoplatonist emanation and participation of John Scot's "Divi- to January 1984 and in March 1985. sion of Nature," in which the return towards cosmic wholeness begins, after the 8. Gheranda Sambitā 1.9-11.

492 493 Notes to Chapter Nine Notes to Chapter Nine

9. Misra, Ayurvedīya, pp. 94-96, 198-99; and Dash, Alchemy, pp. 91-92, 185-86. 26. Śatapatha Brābmana 3.8.2.27: amrtam āyur hiranyam, literally "gold is non- 10. On this distinction, see Misra, Ayurvedīya, p. 211. death; gold is longevity." Cf. Satapatha Brabmana 7.4.1.15-17 and 13.4.1.7. 11. The process of murcchana is described in RRS 11.35. The properties of 27. Śatapatha Brabmana 6.1.2.13-26 and 7.3.1.16, discussed in Malamoud, swooned mercury are described in RA 11.199. "Briques et mots: Obervations sur le corps des dieux dans l'Inde vedique," in Cuire 12. On this distinction, see Dash, Alchemy, pp. 92-93; and Misra, Ayurvediya, le monde, pp. 262-63; and Kapani, Notion de samskāra, vol. I, pp. 58-71. Cf. White, "Dakkbina and Agnicayana," pp. 192-94, 212-13. p. 248. 13. The RRS is the most systematic source on bandhana, which it defines (8.66; 28. Śatapatha Brābmana 7.2.2.7-19; 7.2.4.1-26; 7.3.1.9-11; and Āpastamba 11.60); it also gives a list (11.61-64) and practical descriptions (11.65-112) of the Śrauta Sūtra 16.19.11-13, discussed in Charles Malamoud, "Cosmologie prescrip- twenty-five means for fixing mercury. Cf. Misra, Ayurvediya, pp. 309-21. tive: Observations sur le monde et le non-monde dans l'Inde ancienne," in Le temps 14. Specific procedures for the killing of mercury are given in RRS 11.113-21; de la réflexion 1O (1989), pp. 320-21; idem, "Village et forêt," in Cuire le monde, 30.1-21. The powers and properties of killed mercury (bhasmasūta, mrtasūtaka) are p. 101 n. 35; and Kapani, Notion de samskāra, vol. 1, pp. 63-65. given in RA 1.22 and 11.200; and RM fol. ga.2-3. The RA (12.82) and RC (1.26) 29. Malamoud, "Cuire le monde," in Cuire le monde, pp. 60-62. stipulate that mercury, as the semen of Siva, is not truly killed when reduced to 30. Aitareya Brābmana 1.3. Dīksā is defined as tattva-suddbi, "purification of the ashes, but merely "greatly swooned." On the killing of other metals, see Subbar- constituent elements," in Kaviraj, Bhāratīy, 1:282. ayappa, "Chemical Practice," in Bose, Concise History, pp. 325-26, citing RRS 5.13 31. Maitrī Upanisad 2.5; Manu Smrti 8.96; 12.12,14; Yajñavalkya Smrti 3.178; and RHT 9.16. Bhagavad Gītā 13.1-2; Caraka Sambitā 4.1.65. Cf. Dasgupta, History, 1:214. Ksetra 15. This process has been discussed at length in chap. 8, sec. 3. refers to the lower evolutes of Prakrti, feminine materiality, while ksetrajña is to be 16. RRS 11.47-48. The salt in question, called ambuja ("water-born") or sain- identified with male Purusa, spirit. dhava ("from the Sindhu [Indus River] region"), is said to have arisen from a combi- 32. White, "Dakkbina and Agnicayana," pp. 205-6. nation of "male and female sexual emissions and the menstrual blood of a forbidden 33. Personal communication from Siddhinandan Misra, Benares, March 1985. woman": Sharma's commentary to RRS 11.48 (p. 167), citing the Rasāvatara. See also Misra, Ayurvedīya, p. 326, and above, chap. 2, sec. 2. 17. RRS 11.49. Cf. Dash, Alchemy, pp. 63-64. 34. Zimmerman, "Rtu-Sātmya," pp. 97-99. Cf. Sharma, Ayurved, pp. 276-77, 18. RRS 11.51-52. Cf. Misra, Ayurvedīya, p. 235. 495-503. 19. RRS 11.59. Cf. Dash, Alchemy, p. 65. The first eight samskāras are treated 35. RA 18.2-19; RRS 11.66. The alchemist's preparation in fact begins with his summarily in RHT 2.1-20. initiation, by a guru, into the alchemical arts. Once again, parallels-between hu- 20. RRS 8.72. man birth and the samskāras as rites of passage, and the alchemical samskāras-are 21. RRS 8.80-85. Dash, Alchemy, pp. 65-73. The RHT devotes all of its third explicit: Pandeya's introduction to BbP (Goraksa Sambitā, vol. 2), p. ña. chapter to gräsa, its fourth chapter to cärana, its fifth chapter to garbhadruti, and its 36. Ayurveda Prakāśa 1.489, cited in Misra, Āyurvedīya, p. 327. In preparing fifteenth chapter to babyadruti. it for its ingestion, one is to nourish mercury itself with bijas of gold and silver: 22. The alchemist knows he has reached the stage of jarana when the mercury RA 8.16-22. Cf. KCM (London, Wellcome MSS no. g473) fol. 45.4-9, in which no longer increases in weight after absorbing a given mineral substance: ibid., p. menstrual blood is used as an adjunct. On the ksetrasamskāras (garbbādāna, pumsa- 73. Chapter six of the RHT is devoted to jārana. vana, and sīmantonnayana), see Manu Smrti 9.33, 36-38. In legal terminology, the 23. RRS 8.87; RHT 8.1-19. Indeed, the coloring of metals with mercury colored surrogate male who fathers a child on a widow (ksetra) who has lost her husband in this way may have been the greater part of so-called transmutation in this tradi- (ksetrin) is called the "inseminator" (bījin): Gautama Dbarma Sūtra 28.30, cited in tion: Subbarayappa, "Chemical Practice," in Bose, Concise History, p. 320. Kane, History of Dharmaśāstra vol. 2, pt. 1, p. 599, quoted in Gail Hinich Suther- 24. RRS 8.88; RHT 16.1-36. land, "Bija (Seed) and Ksetra (Field): Male Surrogacy or Niyoga in the Mahābhār- 25. RA 11.216-17; RHT 17.1-8. Neither the RC nor the RRS contains descrip- ata," Contributions to Indian Sociology, n.s. 24 (1990), p. 78. tions of krāmana. 37. Joshi, "Notes," pp. 106-7; HT, p. 85; Kaviraj, Bhāratīy, 1:279.

494 495 Notes to Chapter Nine Notes to Chapter Nine

38. MBbT 10.3-4. The unity of Siva and Sakti is portrayed as that of seed (bija) 48. Tantravārttika, p. 1078, cited in Pandey, Hindu Samskāras, p. 16. Cf. Kapani, and sprout (ankura) in a number of sources: Narharinath, Nav Nath Caurāsi Siddh Notion de samskāra, vol. I, pp. 89-94. (Benares: Rashtriya Press, 1968), p. 25 (v. 3). Cf. Kāmakalāvilāsa, quoted in Wood- 49. Sodhana is derived from the same root as śuddbi, in bbūtaśuddhi. Apart from roffe, Serpent Power, p. 132. the first eight samskaras, relatively pure mercury may be obtained through extrac- 39. There is a certain symmetry here between the father-son relationship of tion from cinnabar (in the vidyadhara yantra, through the process of distillation, Agni-Prajäpati and the guru-disciple relationship of Gorakh-Matsyendra: see be- pātana: see above, chap. 8, sec. 3), and a number of simpler methods. Svedana is low, n. 74 of the present chapter. described in RA 10.41; RHT 2.3; RC 15.36; and RRS 8.62; 11.29. 40. HYP 2.23. Cf. Renou and Filliozat, L'Inde classique, vol. 2, p. 52. 50. HYP 2.12, 19. This process is widely reported in sources (quoted at length 41. HYP 2.22-38. This identification, of satkrīya as the means to bodily bbūta- in Brahmananda's commentary on 2.12) that range from the Yajnavalkya Smrti to śuddbi, was confirmed for me in a personal communication from Mahant Avedya- the Linga Purāna. Cf. Caraka Sambitā 6.15.29-30 (cited in Dasgupta, History 2:326) nath, Gorakhpur, January 1985. on the body's natural voiding of waste products through the bodily orifices and 42. Śatapatha Brābmana 7.2.2.18-20; 7.2.4.10. pores. The KJiN (14.32) calls the first of a list of five sets of Kaula schools "Starting 43. This is clearly stated in Padmāvat 22.8 (= 219), in Grierson and Dvivedi, at the Pores" (romakūpādi). Padumawāti, p. 479, and Shirreff, Padmāvatī, p. 134: "Now that you are a Siddha, 51. HYP 2.13. you have obtained a state of purity (sudbi). The mirror (darpan) of your body has 52. Yoga Sütra 1.2 reads yogas cittāvrtti nirodhab, of which the most generally been cleared of dust." Cf. Śvetāśvatara Upanisad 2.14. accepted and elegant translation is "yoga is the suppression of the states of con- 44. HT, p. 81. Two texts entitled Bhutasuddbi Tantra may be found, in edited sciousness." form, in the Tantrasamgraba, vol. 3, pp. 565-625; and vol. 4, pp. 308-16. 53. HYP 1.56. 45. Jaggi, Yogic and Tantric Medicine, p. 117. The term may originally have re- 54. The most succinct rendering is found in RA 1.19: for variant readings in ferred to the casting out of "beings" or "spirits" inhabiting the area upon which other works, see above, chap. 5, nn. 145, 161, 218, 222, 237. the ritual was to be performed. Bhuta at once means "element" and "spirit": Van 55. The verse is found in HYP 4.27, in which the Sanskrit reads: murcchito Kooij, Worship, p. 21. Cf. BhP 7.368-74. harate vyādhīn mrto jīvayati svayam/ baddhaḥ khecaratām dhatte raso vāyuśca pār- 46. Descriptions of outer and inner bhūtasuddbi are found in Kalika Purāna vati. Cf. RA 1.19: mūrcchito harate vyādhim mrto jīvayati svayam/ baddhaḥ khecar- 52.18-19; 53.13-14; 57.93-108 (in B. N. Shastri, Kālikapurāna [text, introduction, atām kuryāt raso vāyuśca bhairavi. and translation into English], 2 vols. [Delhi: Nag Publishers, 1991], vol. 2, pp. 751, 56. Rājā rānī sambād, v. 18 in Dvivedi, Nāth Siddom, p. 9. Statements of this kind 756, 816-19). (The same passages, with different chapter and verse numbering, are are legion in the hathayogic sources. Cf. HYP 2.2 and 4.28 for a more sober formu- translated in Van Kooij, Worship, pp. 45, 166). Cf. HT, pp. 136, 140, 143; San- lation. derson, "Mandala," pp. 174-75; and Woodroffe Principles of Tantra, 2:477 (cited in 57. HYP 4.26, 28; RA 2.117-18. Cf. RA 1.20-22, the verses immediately follow- Walter, Role of Alchemy, p. 70). ing the classic statement of 1.19. 47. Kālikā Purāna 55.21-35 (in Van Kooij, Worsbip, pp. 45-46) gives a detailed 58. Gorakb Bānī Sabadī 142. Cf. Sabadī 238 and RA 2.117-18. description of the divinity who is to be projected upon this empty space. Cf. 59. RA 11.197. The seven forms-mūrcchita, mrtasūta, jalūkabandba, mūtti- Mabānirvāna Tantra (ed. Woodroffe), pp. 76-77; Bharati, Tantric Tradition, p. 246; bandba, pattabandha, bhasmasūta, and khota[bandha]-are described in 11.198-208. Gupta, HT, pp. 136, 140, 143; Sanderson, "Mandala," pp. 174-75; and Wheelock, RA 10.29 establishes a metaphorical, if not literal, identification between various "Mantra in Vedic and Tantric Ritual," in Alper, ed., Mantra, p. 121. Compare RA forms of treated mercury and Hindu divinities: "Through the power of the Work 18.208-210. On a Theravāda Buddhist precursor of this technique, see Paul Mus, (karmayogabalāt) mercury takes the form of Brahmā when it is purified (ārota); "La notion de temps réversible dans la mythologie bouddhique," École Pratique des (Visnu) Janärdana when it is swooned (mürcchita); and Rudra when it is bound (bad- Hautes Etudes. Annuaire 1938-1939 (Melun: Imprimerie Administrative, 1938), dha). Other sources (RRA 1.1.12a-13a, RM fol. 1a.8-10; Rasasindhu, ASL MSS no. pp. 30-31. 4267, fol. 1.5) follow this enumeration, adding other hypostases of Śiva (Maheś-

496 497 Notes to Chapter Nine Notes to Chapter Nine

vara, Sadāsiva, Isāna) to their lists to correspond to further transformations of 168-69; 6.176; 8.81; 9.81; and RC 16.44, 52-55, 75. See also above, chap. 7, n. 166. mercury. 70. RV 10.113.6, cited in Jean Varenne, Cosmogonies védiques (Milan: Archè, 6o. HYP 2.7-9,35. The final member of the first of the six preliminary practices 1982), p. 114. of yogic purification (the satkarmani), called agnisara ("the coursing of the inner 71. The myth of Gorakh's binding of the nine serpent clouds is related in the fire"), involves a similar body of practices which are said to activate the five inner ca. A.D. 1820 Śrī Nāth Tīrthāvalī of Mān Singh, vv. 330-36 and Lévi, Le Népal, vol. fires (of digestion, etc.). In agnisara, one is to repeatedly fill and empty the abdomen I, pp. 351, 372, citing the "lost" Nepali Buddba Purana. This binding may be a of air, after the fashion of a blacksmith's bellows: Renou and Filliozat, L'Inde reference to the nine cakras of a system described in Gorakhnäth's SSP (2.1-9). classique, vol. 2, pp. 52-53. More likely, it refers to a commonplace of Siddha literature, that through a given 61. HYP 2.69. HYP 3.12, calls the outer nādīs "lifeless"; 3.73 says they are "para- practice, one becomes more powerful than nine serpents: KCM fol. 20a.8; BbP lyzed"; and 3.74 calls them "bound," all synonyms for "swooned." See below, n. 89. (ASL MSS 4401, fol. 3b.49). The RA (12.314) raises this figure to ten. 62. Brahmananda's commentary to HYP 4.27. Murcchana has a number of other 72. HT, p. 53, citing Paramārthasāra 1.15; and Kaviraj, Tantrik Vanmay, p. 58. meanings, including "swelling," "coagulation," "assuming shape," which together 73. Yogabija 52: chedabandhairmukto. Cf. Abhinavagupta's praise of Matsyendra lead to an image of a bruise which, taking shape as it swells, becomes deadened to as one who "tore apart the glowing net," in TA 1.7. Lee Siegel's delightful Net of touch: Monier-Williams, Sanskrit-English Dictionary, s.v. "murch." Magic is a wide-ranging discussion of this entire symbol system. 63. HYP 3.55-6. Uddiyana is also the name of one of the four original pithas of 74. K7nN 17.33-37. Cf. Gorakhnath's statement that "bound in the noose [of both Buddhist and Hindu tantrism. In the Nath Siddha system of the nine cakras Death] one becomes a living creature. Freed from the noose, one becomes Sada- (which alternates with the 6 + 1 cakra system), uddiyāna (here called odyāna) is lo- siva": Sahagal, Gorakh Darsan, p. 107 (without citation). The Paraśurāma Kalpa Sū- cated at the level of the second cakra, the svadhisthāna: SSP 2.2. tra (1.5) makes a nearly identical statement. Tantrics also have the power to kill. 64. HYP 3.70-73. Jalamdbara is also the name of one of the four original pithas, This one of the "six acts" (satkarmāni) of tantric black magic is called mārana: HT, located in the Nath Siddha system of the nine cakras at the level of the brabmaran- p. 35. dhra-nirvana cakra, the eighth cakra identified with the fontanelle: SSP 2.8. The 75. HYP 4.17. On the three knots, HYP 4.70, 73, 76. Näth Siddhas locate the two other original pithas, Kāmarūpa and Pūrnagiri, at the 76. HYP 4.94. lowest and highest cakras, respectively: SSP 2.1, 9. The form of diaphragmatic re- 77. Yogamārtanda 180, in SSP, p. 71; Yogabīja 52; and HYP 3.24. Cf. HYP 4.108, tention called mūrccha, discussed above (n. 60) in fact combines breath retention which states that the realized yogin is not bound by karma, not eaten by time, and with the jalamdhara bandha: HYP 2.69. is beholden to no one; and Gorakh Bani Sabadī 85, which makes a similar statement 65. HYP 4.26. in the context of what it calls the anahat bandh, the "lock of the unstruck [sound]." 66. On the seven types of alchemical bandbas, see RPS 2.2, and above, n. 59. See also Akulavīratantra [B] 30, 49-50, 65, in Bagchi, KJiN, pp. 96-99. The ava- The verse concerning the raksasa named Method is RA 12.3, and the commentary dhuta is one who has thrown down (ava-dhu) the bonds of existence: NSC, p. 126. is from the glossary to Tripathi's edition, p. 373. Cf. SSP 6.1. 67. See above, chap. 8, n. 220. 78. Death (named Yama or Kala) carries a noose, as does the divine enforcer of 68. Gorakb Bānī Sabadī 88. Cf. HYP 4.91, which calls the nada an elephant goad order and binding contracts, Varuna. The latter, as a first-function god, is a special- for stabilizing the wild elephant of the mind. This passage is the first of a series of ist in invisible binding weapons, not unlike the Scandinavian Odin who ensnares nine verses (HYP 4.91-99) which metaphorically compare the effects of the nāda enemies on the field of battle with his invisible fetters. The bonds of death are an on the mind to all manner of immobilization in nature and culture. In Saiva ritual, Indo-European theme, on which see Bruce Lincoln, "Mithra(s) as Sun and Savior," the ankusa mudra (a fist with the index-finger bent in the form of a hook) is em- in Death, War, and Sacrifice: Studies in Ideology and Practice (Chicago: University ployed to generate nectar from within the subtle body: Gonda, Visnuism, p. 180. of Chicago Press, 1991), especially pp. 78-81. On kālavañcana, see HYP 3.3; KJāN 69. The notion dates back to Atharva Veda 11.4.21. On clipping the wings of 17.17; KCM (ASL MSS no. 3952) fol. 31a.3, 31b.2. the cosmic goose, the bamsa, see Carpati jī kī sabadi 47, in Dvivedi, Nāth Siddbom, 79. Gopicandrer Gan, summarized in Dasgupta, Obscure, p. 227. Gorakhnāth p. 17; and HYP 4-92. On the clipping of mercury's wings, see RHT 4.5; BbP 5.148, does the same: Goraksa Vijaya, pp. 45-48, cited ibid., p. 222.

498 499 Notes to Chapter Nine Notes to Chapter Nine

8o. Gopicandrer Pancali, p. 345, translated in Dasgupta, Obscure, p. 226. 90. HYP 3.4. In classical Indian thought as well, one enjoys one's "third" and 81. Gopīcandrer Sannyās, p. 418, cited in Dasgupta, Obscure, p. 215. most glorious birth, this time in heaven, when one's corpse has been incinerated in 82. Ernest Trump, The Adi Granth (London: 1877), p. xl, cited in Briggs, the cremation fire: Satapatha Brāhmana 11.2.1.1. Gorakbnāth, p. 203. Cf. Padmävat 129 [=12.1] for an extended description of a 91. On vajroli, see above, chap. 7, nn. 90-102. Sabajoli and amarolī are described fully equipped Näth Siddha; and Briggs, Gorakbnath, pp. 6-23 and Dvivedi, Nath in HYP 3.92-98. These practices appear to be connected to the broader tantric Sampradāy, pp. 15-21, for discussion. ritual practice of amrtikarana, which involves the identification of ashes, water, and 83. The wavering mind is itself compared to a gazelle in the HYP (4.94, 99), a nectar: on this practice, see Gonda, Visnuism, pp. 72, 180. metaphor dramatized in the legend of Gorakh's disciple Bhartrhari: Briggs, 92. See RA 12.79 on the state of the question. It is also called bhasma-nirmāna Gorakbnātb, p. 132. pārada: MBbT 5.1. See above, n. 14. 84. Gorakb Bänī Sabadī 48a, in which the kantha is identified with the five ele- 93. RA 7.142. Personal communication from Siddhinandan Misra, Benares, ments. Cf. the Jain legend of Kanthadi, a disciple of Gorakh, who transfers the heat March 1985. of a fever (or his yogic energy) into his kantha: Prabandba Cintāmani, pp. 22-23, 94. An early example is Vajasaneyi Sambitā 40.15: "Now my breath is the im- cited in Dvivedi, Nath Sampradāy, p. 51. mortal wind, my body is bhasma," cited in David Knipe, "Night of the Growing 85. On dying to the world and performing one's cremations rites, see Pandurang Dead: A Cult of Virabhadra in Coastal Andhra," in Alf Hiltebeitel, ed. Criminal Vaman Kane, History of Dharmaśāstra, 5 vols. 2d ed. (Poona: Bhandarkar Oriental Gods and Demon Devotees: Essays on the Guardians of Popular Hinduism (Albany: Institute), vol. 2, part 2, pp. 954-55, 985. On the prānāgnibotra, see Vaikbānasa SUNY Press, 1989). Smarta Sütra 2.18. This tradition goes back to the Taittirīya Āranyaka: Madeleine 95. Maitrī Upanisad 6.15. Biardeau and Charles Malamoud, Le Sacrifice dans l'Inde Ancienne (Paris: Presses 96. Malamoud, "Village et foret dans l'idéologie de l'Inde brahmanique," in Universitaires de France, 1976), pp. 67-68. A Prānāgnibotra Upanișad, a compi- Cuire le monde, p. 102; and Kapani, Notion de samskāra, vol. I, p. 60 n. 34, citing lation of earlier Upanisadic descriptions of this practice, is found in Deussen, Sixty Śatapatha Brabmana 6.2.2.6; 6.3.1.1 .; and 6.5.3.1. Upanisads, 2:645-51. Because the renouncer has performed his own cremation rites 97. Gopicand's statement (in Temple, Legends of the Panjab, vol. 2, p. 48 [legend upon taking up sannyāsa, his body is not burned, but rather inhumed or set adrift, no. 18, vv. 499-500) reads kachā bartan bove jidbar phere phir jāe/bam to jogi būe gurū upon its apparent "death." On this see Paingala Upanisad 4-5-8, cited in Kapani, ne dīe pakāe. Cf. Śvetāśvatara Upanisad (2.12), which states that a body fired by yoga Notion de samskāra, vol. I, p. 153. (yogāgnim mayam śariram) cannot be reached by either sickness, aging, or death. 86. RA 1.19. The optimum fixation (bandha) of mercury, effected through its Whence also the invocation to Agni and the gods, "May we be well cooked!": RV "assimilation" in female sulfur and mica, leaves it in an ashen state (bhasmīkrta) 9.83.1. Cf. Śiva Sambitā, 2.32-34; and Yogabīja 34-35, 51, 76, which distinguishes in which it has been rendered "lifeless" (nirjiva: in opposition to the quite useless between a body that is "cooked" or "fired" (pakva) by yoga and one that is not sajīva-bandba, which leaves mercury subject to evaporation and has little medical (apakvā); and evokes a "seven-dhātu body fired in the fire of yoga." use: RRS 11.75), i.e., completely immobilized and impervious to heat and thereby 98. Here we are paraphrasing the formulation of Benjamin Walker, Hindu most effective against disease: RRS 11.76. World, s.v. "ashes." 87. Gorakb Bānī Sabadī 26: marau be jogī marau, maraņ hai mīthā/ tis maranīm 99. On the Pāsupatas, see Lorenzen, "Śaivism: Pāśupatas," in Encyclopedia of Reli- marau jis maranīm gora mari dīthā. Cf. Yogabīja 57-58a, and Kabir, Sakbī 19.13 gion, vol. 13, p. 18. On vibhüti, Monier-Williams, Sanskrit-English Dictionary, (= Doba 41.8), in Vaudeville, Kabīr; p. 260. s.v. "vibhuti." On the consumption of ashes as a component of dīksā, see Manu 88. Padmavat 24.2, in Grierson and Dvivedi, Padumawāti, p. 530. Smrti 6.25, 38. 89. Amaraugba Prabodha 40b-41, which is nearly identical to HYP 3.27b-28. I 100. Walker, Hindu World, s.v. "ashes." That is, Siva ejaculated at the sight of have translated the problematic jānīyāt in this passage as a gerundive. The fourth Pärvatī, in which case the combination of elements which Siva smeared over his state of consciousness is called either mūrccha or turīya ("trance"). A fifth may be body evokes those of the amaroli mudrā. added. This is marana, "death": commentary to English translation of HYP 4.107, IO1. Mababbarata 3.81.98-118; 9.37.34-50; Skanda Purana 5.2.2.2-37; Padma p. 107. Purāna 1.27.1-15; 5.18.132; Vāmana Purāna 17.1-22. A brief Mabābhārata

500 501 Notes to Chapter Nine Notes to Chapter Nine

(13.17.92) reference to the same myth has Siva calling himself bhasmabbūta, "made detail concerning milk and menses, is adapted into Hindi in Gautam, Srī Gorakh- of ashes": O'Flaherty, Siva, p. 245. nāth Caritra, p. 6. A Punjabi variant is briefly recounted in Rose, Glossary, p. 393. 102. Brabmānda Purāņa 1.2.27.106b, 108ab, 109b, 113ab, 121a, 123a, 124b. Another version of the same account, from the Tabqiqat-i-Cisti, is cited in Briggs, 103. Kāvya Usanas's poetic powers are hymned in RV 9.87.3; a Brahmanic myth Gorakbnāth, pp. 182-83. NSC (p. 49) gives a north Indian version of this legend. of his rivalry with Indra and the gods is found in Jaiminiya Brabmana 1.125-27. 113. These specially prepared ashes are in fact twice calcinated, by virtue of See O'Flaherty, Tales of Sex and Violence, pp. 87-90. which they take on their distinctive snowy-white color. On this process, see Bedi, 104. Georges Dumézil, Mythe et épopée, vol. 2, part 2, "Entre les dieux et les Sadbus, pp. 78-79. démons: un sorcier," pp. 161-66; 197-205, 208. The myth is found in Mababbār- 114. Briggs, Gorakbnāth, p. 21. The present-day Gorakhnāth temple complex ata 1.71.2-1.72.25. of Gorakhpur houses both an eternal flame (akhand jyoti) and an eternal dhūnī (a- 105. Dumézil, Mythe et épopée, 2:200-4. kband dhūna). 106. On the alternation between kavi and Kāvya Usanas's name, see Dumézil, 115. On Gorakh's initiation, see Dvivedi, Nath Sampraday, p. 19. On Carpati's Mytbe et épopée, vol. 2, part 2, pp. 148-56, 204. On the incantatory element in Vedic conception from Gorakh's langoti water and Guga's conception from ashes given healing, see Zysk, Religious Healing, pp. 8 and 241, citing Sayana's commentary to by Gorakh to his mother, see NSC, pp. 193, 256. Gorakh initiates Pūran Bhagat Atharva Veda 4.2.6. It should be recalled here that "magical alchemy" was an affair with ashes: Temple, Legends, 2:445-46. of Asuras: see above, chap. 3, nn. 35-40. On the alchemical kavi and the medical 116. Temple, Legends, 2:436-37; William Crooke, "A Version of the Guga Leg- kavirāj, see above, chap. 1, nn. 49-50; below, chap. 1o, sec. 5. end," Indian Antiquary 24 (1895), pp. 52-53; Gautam, Śrī Gorakbnāth Caritra, 107. Knipe, "Night," in Hiltebeitel, Criminal Gods, pp. 149-50. In the Punjabi p. 106; NSC, p. 276. Hills cults of the Seven Sisters, the term pindi is applied to "a lump of stone, some- 117. Rose, Glossary, vol. 2, pp. 390-2; Srivastav, ed., "Gorakb" Viśesānk, p. 90. what resembling a Siva-linga" that are manifestations of the Goddess herself: Erndl, On the calcinating breath of the cosmic serpents, see Temple, Legends, vol. I, p. Victory to the Mother; p. 66. An extensive discussion of the term pinda is found in 177 (legend no. 6 ["Guru Gūggã"], v. 641). Kapani, Notion de samskāra, vol. I, pp. 129-37. 118. Above, chap. 8, sec. 4. 108. See above, chap. 7, n. 98 on the -oli suffix; and chap. 8, n. 194, on the 119. HYP 1.27, 48; 2.66; 3.111, 115; 4.10, 11, 19. Somānanda, in his Šakti- relationship between the Näga suborders of the Dasnämi sects and the Nath sam- vijnana, uses the term utthapana (vv. 12-13) for the act of raising the kundalini; and pradāya. On ashen golās carried by the former, see Sadananda Giri, Society and Sann- bodbana (vv. 14-15) for the subsequent act of causing the kundalini to pierce the yāsin: A History of the Dasanāmī Sannyāsins (Benares: n.p., 1976), p. 26. granthis of Brahma, Visnu, Rudra, and enter into Isvara, who is found between the 109. The alchemical texts generally keep the reading gutikā; however, see RA eyebrows, and finally into Sadāiva, via the brabmadvāra: Silburn, Kundalinī, p. 113. 12.330 for gulikā. See Monier-Williams, Sanskrit-English Dictionary, s.v. "guda." 120. Padoux, Vāc, pp. 86-121 (especially p. 87, citing Sārada Tilaka 1.7-8) de- 110. Pir Premnath, Siva-Goraksa, pp. 29, 61. scribes this interaction at length. See Gonda, Medieval, p. 186 (citing Mrgendrā- 11I. Śiva-Goraksa, pp. 29-34, 36, 41-46. Throughout his account, Premnath gama, Kriyāpada 1.2); and Silburn, Kundalinī, p. 3 for alternative terminology. identifies Vicarnath with Bhartrhari. The Śri Nath Tirtbavali of Man Singh (v. 364) 121. HT, pp. 174, 177. On the Kaulāvalīnirņaya of Jñānānandagiri Parama- relates that Bhartrhari revived a dead child (whom this source names Siddha Kāyā- hamsa, see HTSL, p. 144. For a discussion of the related term nirodbini in the man- nath) in a cave within "Gorakh Tilla" which, although it is usually identified with tric practice of chapter 21 of the pre-roth c. Netra Tantra, see Padoux, Vac, pp. the prestigious Näth Siddha monastery of of Sialkot (Jhelum dist., Punjab, Paki- 103-4. Suppression (nirodba) is used as a synonym for prāņāyāma in HYP 2.2 and stan), is located by this author at Pharphing, in the Kathmandu Valley (an A.D Gorakb Bānī Sabadī 82, 92. 1393 inscription at this site states that Goraksanätha's sandals were established 122. Lilian Silburn, "Le vide, le rien, l'abîme," in Le Vide: Expérience spirituelle there by Acintanätha; Tibetan traditions identify this figure with Matsyendra: Bag- (Paris: Hermes, 1969), pp. 30-33. This technique, also known to Mahāyāna Bud- chi, introduction to Kaulajñānanirņaya, p. 23)! dhism, was transmitted via China to the Japanese Zen masters, where it took the 112. Lévi, Le Népal, vol. 1, pp. 351, 372 citing the Buddba Purāna; and Dvivedi, form of the koan. Nath Sampraday, p. 47, citing the YSA. This latter account, which includes the 123. RA 10.11: jāranād bandbanam bhavet.

502 503 Notes to Chapter Nine Notes to Chapter Nine

124. RA 11.50-54. 139. Maitrī Upanisad 6.28; Gorakb Bānī Sabadī 33, 215. 125. RA 11.12: grāsam grbņa mama prabbo. 140. Khakhar, "History," p. 49. 126. On the successive absorption by mercury of six times its mass of mica, see 141. Mircea Eliade, Shamanism, Archaic Techniques of Ecstasy, tr. Willard R. Trask RA 11.70-73. More recent sources (Rasacintamani 5.73) suggest that sulfur should (Princeton: Princeton University Press, 1972). be used before mica in this process: Misra, Ayurvediya, p. 243. In modern practice, 142. Although most traditions make Gopicand a king in Bengal, the oldest re- sulfur, rather than mica, is used: Ray, History, vol. 2, pp. xliii-xliv; and Dash, Al- censions of his song cycle appear to be from Nepal: Mukherji, Gopicandra Nātaka, chemy, p. 100. It should be noted here that mercury has six times the atomic weight p. xliv. of sulfur and that synthetic cinnabar is composed of 84% mercury and 16% sulfur: 143. Legend cycles of Jalandhara (or Hādipa) and Gopīcand are found in Bengal personal communication from Damodar Joshi, Benares, March 1985. (Govindacandrer Gän, summarized in Dvivedi, Nath Sampraday, pp. 184-86), the 127. RA 11.98-106. Punjab (summarized in Rose, Glossary, p. 395, and Temple, Legends, 2:1-71 [legend 128. Personal communication from Damodar Joshi, Benares, August 1984. Cf. 18]); and Rajasthan (Gold, Carnival, pp. 159-310). A Marwari recension is the Gop- Dash, Alcbemy, p. 100. īcand kā Akbyān, ed. Sridhara Sivalal (Bombay: Jnan Sagar, 1890). The Bengali 129. Gorakb Bānī Sabadī 252a. Cf. Sabadī 13b, in which the yogin is enjoined to Gopīcand cycle intersects that of Gorakh's rescue of Matsyendra from the Plantain absorb (jaranam) [his mind] in [the words issuing from] his guru's mouth. Forest, in the trilogy comprising of Goraksa Vijaya, Gopīcandrer Gân, and Anila Pur- 130. This practice, dramatized in the legend of Mastnäth (Sankarnath, Śri Mast- āna: Mukherji, Gopicandra Nāțaka, p. xxxvi. Three bānīs, entitled Gopīcand jī kī Sa- näth Carit, pp. 109-10), has for its outward correlate the playing of the singnad, badi, Rājā Rānī Sambād, and Rāg Rāmagrī, offer fragmentary accounts of the Gopī- the antelope's horn worn by Nath Siddhas, which is made to emit a (barely aud- cand legend: Dvivedi, Nath Siddham, pp. 6-11 (vv. 47-125). ible) sound by blowing: Véronique Bouillier, "La caste sectaire des Kānphațā Jogī 144. Kabīr, in one of his Sākbīs (4.8), refers to the Ketakī (vernacular kevarā; dans le royaume du Nepal: l'exemple de Gorkha," in Bulletin de l'Ecole Française Latin Pandarnus odoratissimus), a fragrant shrub whose creamy colored flowers at- d'Extrême-Orient 75 (1986), p. 147. tract bees: Vaudeville, Kabir, p. 179 and n. 7. 131. Gorakhnäth mentions (Gorakb Bānī, Sabadi 85) the anabat bandb, a power- 145. The motif is an archaic one, going back at least as far as the Kausītaki ful lock presumably effected at the level of the heart, by which the yogin is ren- Brābmana (6.1-2, translated in O'Flaherty, Hindu Myths, p. 31). dered invulnerable. 146. Grierson, "The Song of Manik Candra," Journal of the Royal Asiatic Society 132. HYP 4.85-86. of Bengal 47 (1878), no. 3, pt. I, p. 218, stipulates that she survived this ordeal by 133. HYP 4.100. taking the form of a mustard seed (siddbartha) 134. baddham vimuktacāncalyam nādagandhakajāraņāt/ manah pāradam āpnoti 147. Gopīcandrer Pāñcalī, pp. 366-69; Govinda-candra-gīta, pp. 71-73; Gopī- nirālambākhyakhe 'țanam. candrer Gân, Bujbān Khanda, pp. 87-130, summarized in Dasgupta, Obscure, p. 226. 135. RRS 11.78. 148. RA 18.228. 136. HYP 1.17, 27. 149. Temple, Legends, 2:23, cited in Briggs, Gorakbnāth, p. 194. NSC (p. 210) 137. A matra (4 seconds) is defined as the time it takes for a sleeping man to maintains that these effigies were made of the seven metals (sapta-dbatu); other breathe in and out once: six of these are a pala (Yogacintāmani, cited by Brahmā- Bengali accounts say all three are made of gold (Mukherji, Gopicandra Nātaka, p. nanda, in his commentary to HYP 2.12). Alternatively, it is the time it takes to xli). The Bengali Gopicandrer Git says merely that Kanha-pa made three Gopīcand circle the knee three times with the palm of the hand, and then snap one's fingers effigies: Dvivedi, Nath Sampradāy, p. 186. (loc. cit., citing Yajñavalkya Smrti). Geometric progressions of hathayogic palas are 150. Another version, from the Punjab, states that Gorakh created a man from found in Amanaska Yoga 1.50-98; the relation between temporal duration and the a blanket, who called himself "Gopicand" when Jalandharipa asked him who he successive stages of Pātañjala yoga are also described in Brahmānanda's commen- was. The angry yogin reduced him to ashes seven times, after which the real Gopī- tary to HYP 2.12. See below, chap. Io, nn. 57-58. cand was brought forward. Jalandharīpa then declared that since he had not been 138. RA 18.56-60. consumed by fire, he should become immortal: Rose, Glossary, p. 395. Cf. the

504 505 Notes to Chapter Nine Notes to Chapter Nine

Marwari version, which states that Jalandhara reduced seven Gopīcand effigies to sanga was Caurangi's surrogate father. The Punjabi legend recounted in Temple ashes, after which he declared his disciple immortal: Sivalal, Gopīcand kā Akbyān, states that Puran Bhagat was conceived by Acchran "as soon as the Sun saw her" pp. 33-35. (v. 93). 151. Gorakb Bānī Pad 34.5, glossed in Srivastav, ed., "Gorakb Bānī" Visesānk, pp. 162. This he does in spite of Acchran's warning, which evokes a yogic notion of 282-83. Alternatively, he uses magic to restore these limbs, which his stepmother the body already discussed: "There are nine gates to the city, go not to the tenth, had cut off (see below, for an identical motif, in the legend of Pūran Bhagat): The tenth is the palace of thy stepmother, Nūnan": Temple, Legends, 2:398 (vv. Briggs, Gorakhnäth, p. 191. On a parallel, from the Tibetan Grub thob, see above, 265-66). See above, chap. 8, sec. 4. chap. 4, n. 148. 163. Lūņa, who is said in Temple's Punjabi recension of this legend (Legends, 152. Gorakb Bānī, Panc Mātrā 21a, in Srivastav, ed., "Gorakb Bānī" Visesānk, 2:392 [v. 187]) to be of the Chamari (currier) subcaste, is widely known as Lona p. 392; and Barthwal, Gorakh Bānī, p. 221. (or Nonā) Chamäri, the most terrible sorceress of north India. She is said to have 153. Briggs, Gorakbnath, p. 197. gained her powers by eating the corpse of Dhanvantari, the divine founder of In- 154. Gautam, rī Gorakhnāth Caritra, pp. 14-16. dian medicine, who had died of snakebite at the fangs of the great nāga Taksaka: 155. Ibid., pp. 96-97. Crooke, Religion and Folklore, p. 437. Veiled references are given in Temple, Legends, 156. Ibid., pp. 59-62. 2:386 and 413 (vv. 118-19, 441). Curiously, Salwan is himself said to be the son 157. Briggs, Gorakbnath, p. 72. of another great nāga, Vāsuki (Bāsak) in another version of this legend: Briggs, 158. Ibid., p. 187. Gorakbnāth, p. 184. The Jain Prabandba Cintāmaņi gives Nāgārjuna a similar pedi- 159. The principal recensions of the Püran Bhagat/Cauranginath legend are gree: see above, chap. 4, n. 194. Noņā Chamāri is said to have belonged to an those found in Temple, Legends, vol. 2, pp. 375-455 (no. 34); Swynnerton, Romantic Islamic subdivision of the Nath sampradaya, that of the Ismail Jogis: Rose, Glossary, Tales, pp. 411-41, both of which are Punjabi; and the twenty opening verses of the p. 396. Prän Sankalī of Caurangīnāth (in Dvivedi, Nāth Siddhom, pp. 19-20), written in a 164. Temple, Legends, 2:408 (vv. 383-86). medieval eastern Rajasthani dialect (ibid., p. 16 of the preface). It is also recounted 165. Ibid., 2:411 (vv. 418-19). by Jñāneśvara in his YSA (p. 372, cited in Dvivedi, Nath Sampraday, p. 177), and in 166. Ibid., pp. 414-15 (vv. 457-468). Gautam, Srī Gorakbnāth Caritra, pp. 81-89. 167. Ibid., p. 417 (vv. 498-502). 160. Asian Mythologies, s.v. "Turkish and Mongolian Shamanism," by Jean-Paul 168. Ibid., pp. 419-20 (vv. 517-33). It is Lūņā's maid, Hīrā ("Diamond"), who Roux; idem., "Le nom du chaman dans les textes turco-mongols, Anthropos 53 tests the blood. It will be recalled that Hira was the name of the sorceress who (1958), pp. 113-42; Diane M. Coccari, "The Bir Babas of Banaras and the Deified enslaved Gopīcand for twelve years: see above, n. 149. Dead," in Hiltebeitel, ed., Criminal Gods, p. 256; and William Crooke, Religion and 169. Ibid., p. 422-23 (vv. 555-67). The well is so described in Narharinath, ed., Folklore of North India (London: Oxford University Press, 1926), p. 319. Sialkot, Śrīnātbkathāsāra, p. 18. the ancient Sakala, the venue of the Puran Bhagat legend, was the capital of the 170. Temple, Legends, p. 426 (v. 604). Cf. Kabīr, Sākbī 29.2 (in Vaudeville, Kabīr; central Asian Hunas, the Ephthalite Huns, in the latter half of the first millennium: p. 290), quoted above, chap. 8, n. 115. the Punjabi Jats, of which Püran Bhagat would have been a scion (Briggs, Gorakh- 171. Temple, Legends, pp. 428-34 (vv. 630-701). näth, p. 239), are descended from this people. On this, see White, Myths of the Dog- 172. Ibid., pp. 440-46 (vv. 774-846). Man, p. 120. The term bhagat is employed for the rasdbari actors whose troupe 173. Ibid., pp. 448-55 (vv. 875-956). Gorakhnath joins to release Matsyendra from the clutches of the women of the 174. See above, chap. 7, nn. 82-83, for alchemical allegory in the Padmāvat. Plantain Forest: Rose, Glossary, p. 394 [n.]. 175. Both sankba and [birwa] lona are also herbs used in alchemical operations: 161. Prān Sānkalī, in Dvivedi, Nāth Siddbom, p. 19 [= v. 206]. On the Śankh/ KCM fol. 12a.9; and Padmavat 310 [= 27.3]. Salwan alternation, see Temple, Legends, 2:376, 378 [legend 34, vv. 5, 28], who says 176. On this forest, see above, chap. 8, sec. 2d. in a note that Sankh is Salwan's father. Gautam, Śrī Gorakbnath Caritra (p. 82) calls 177. This would parallel such Hindi constructions as dayālū, from dayā, kindness the king Sasanga and maintains that Caurangīnath was born from Siva's seed: Sa- (thus "kind"); or krpālū, from krpa, "mercy" (thus "merciful").

506 507 Notes to Chapter Nine Notes to Chapter Ten 178. Sachau, Alberuni's India, vol. I, p. 192. From here, the story takes a turn and level of thought which is sometimes taken to be the homologue of samarasa. The becomes an account of the sack of Valabhi, related above: see chap. 4, nn. 200-202. gold that comes in (or "permeates"), by its very nature [sabaja], refers to the irre- 179. Sachau, Alberuni's India, vol. 1, pp. 191-92. Such traditions have remained sistable flow of refined semen realized at the end of one's hathayogic practice and common in the regions of the Sind and Kacch, on the western fringe of India, to the effects of the alchemical touchstone, which transmutes all it touches into down to the 19th c .: Madhihassan, Indian Alchemy, pp. 58-59. gold. 180. Gorakb Bānī, Pad 6.1-5 in Srivastav, "Gorakb Bānī" Visesank, pp. 260-61. 181. This is the body that is rejuvenated through its calcination in the fires of Chapter Ten yoga: cf. Kabīr, Doba 46.64 (in Vaudeville, Kabir Granthavali, p. 70). Cf. Lallā- I. E. M. Forster, The Hill of Devi (Harmondsworth: Penguin, 1965), p. 88. Vakyāni 100, cited in Silburn, Kundalinī, p. 45: "Give thou breath to the bellows 2. In particular, it is the ojas, the fluid of life which exists, in extremely limited even as doth the blacksmith. Then will thine iron turn to gold ... " The Vedic quantity in the heart, that is to be restored. See above, chap. 2, n. 44; and Ananta- blacksmith's bellows are fanned with birds' wings: Maurice Winternitz, A History charya, Rasayana, pp. 20-21. of Indian Literature, trans. V. Srinivasa Sarma, 5 vols. (Delhi: Motilal Banarsidass, 3. Gorakb Bānī Pad 6; 13.1; 50.2; Šiv Goraks Bāvanī 38. 1981), 1:59. 4. Gorakb Bānī Sabadī 60, 211; Pad 5.1; 15.1; BhP 5.260; 5.297. 182. raļtļtī kā kām māse kī corī literally means "the work [or desire, or semen] of 5. See above, chap. 5, n. 113. one rati's weight [is] the theft of one masa's weight." According to the weights and 6. Alchemical and other uses of the root vedh likely remount to the metallurgical measures used in precious metalworking, 8 ratīs (the seed of Abrus precatorius, 21/2 technique of amalgamation of mercury with gold for the extraction of gold from grains, 121 milligrams) equal one mäã (lentil; one-twelfth of a tola). Below, in verse the ores in which it naturally occurs. The earliest such use of the term (rasaviddba) 5, the unit of one gadiyana is used: one gadiyāna equals 6 masas and therefore 48 is found in the Arthasāstra of Kautilya (2.12.2). ratīs. Kabīr makes poetic use of the terms ratti and tila (sesame seed, another unit 7. On this relationship, see, for example, Brhadāranyaka Upanisad 1.4.11; Ma- of weight) in doba 35.7-8 (in Vaudeville, Kabir Grantbavali, p. 56). habbarata 12.72.9-12 and 13.8.21-22; Manu Smrti 8.37-39. There is a symmetry 183. Because I have found it overly "spiritualized" (to the neglect of most of between the brahman-ksatriya relationship and that obtaining between the ksa- this poem's concrete referents), I have generally avoided the tīka provided by Sri- triyas and the third varna, the vaisyas: Satapatba Brābmana 12.7.3.8. vastav and the Nath Siddha editors of the "Gorakh Bani" Visesank. Here, however, 8. Śatapatba Brāhmana 5-4.4.7-19. I find their commentary useful: the grain is the individual soul, into which the 9. The myth is related in Devibhāgavata Purāna 7.17.45-7.18.58 and Mārkandeya universal soul (the lentil's weight) has emanated. Gorakhnath, who has stolen his Purana 7.1-69; 8.1-270. For a discussion see White, Myths of the Dog-Man, pp. way into the universal soul, identifies himself with it and dwells within it, is lost in 80-86. it. There is, however, another probable reading of this verse: The play of propor- 10. The Mārkandeya Purana version dates from the 3d-4th c. A.D. tions, of a grain acting on a mass eight times its weight, is a common alchemical 11. The Canda-Kausika, ed. Sibani Das Gupta (Calcutta: The Asiatic Society, phenomenon, in which an infinitesimal quantity of perfected mercury can trans- 1962). mute up to a billion times its weight of base metals into gold. 12. Candakauśika 4.25ff. 184. Cf. HYP 4.56. 13. Candakausika 4.30-31. This is a slightly altered version of the old Buddhist 185. Here, the triadic void is the trikuti, the three-crested peak, located in the set of eight siddhis. cranial vault, which is the culminating point of the three channels [nādīs]. Alter- 14. Candakauśika 4.32, 34. nately, it is the downturned triangle from which the nectar of immortality the yogin 15. M. A. Stein, Kalhana's Räjataramgini, Chronicle of the Kings of Kashmir, 2 vols. has generated drips downward, thereby transforming his body. At this point, all of (London: Constable, 1900; reprint Delhi: Motilal Banarsidass, 1979), I:102. mundane existence, including the pure/impure opposition, has been transcended. 16. This is the third and final regnal period of Cakravarman. He had also ruled 186. Unmani or unmana is a transcendent state of consciousness, located at the in A.D. 923-33 and during part of 935: Rajatarangini 5.288-92, 297-302. highest level of the subtle body. It is also a term which connotes equanimity, a pure 17. Kānha-pā calls himself a Kāpālika who dances with a Dombī (before killing

508 509 Notes to Chapter Ten Notes to Chapter Ten

her) in Caryapada 10 (in Kværne, Anthology, p. 113). On the Dombi as portion of 27. Kulārnava Tantra, 14.62-68, 78-79. the female sexual organ, see the A.D. 1446 Bengali commentary to the Laghu Kāla- 28. Sarada Tilaka 5.128-41. An English translation of this passage is found in cakratantra, reproduced in Shastri, Catalogue, vol 2, p. v. HT, pp. 86-87. 18. Ucchista Candāli ("Outcaste-Leftover") is a name for the tantric Goddess/ 29. Goraksa Sambitā, vol. 1, ed. Janardana Pandeya. The initiation is called consort in a number of traditions: Apfel-Marglin, Wives of the God-King, p. 240. vedhavatī in 14.251-52 and loosely described in 14.254-74. This Sākta text also 19. Tod, Annals and Antiquities, 2:1196. See above, chap. 4, n. 240. calls itself the Kādibbeda Tantra: HTSL, p. 55 and n. 92. 20. Tod, Annals and Antiquities, 2:107. This is portrayed graphically in 19th c. 30. TA 29.239-40, citing the Gabvara Tantra. miniature paintings, found in miniatures held in the Jodhpur Fort Museum as well 31. Sārada Tilaka 5.138b: tam punarguruvaktre tu yojayed. See also above, chap. as murals on the outer walls of the shrine of the Mahamandir, of the Marwar king 7, n. 94; and chap. 8, nn. 213-15. Man Singh together with his minister "Jalandharanath." In these miniatures, the 32. TĂ 29.271. king is standing or pictured in a subservient position to his Näth Siddha, who is 33. Šārada Tilaka 5.140. seated, in conical cap, under the royal parasol. 34. Silburn, Kundalinī, p. 95. 21. Gorakbāvamšāvalī, ed. Yogi Narharinath (Benares: Rashtriya Press, 1964), p. 35. Tucci, "Animadversiones Indicae," pp. 138, 154: vaktreņa vaktram dattvā 95 (in Nepali): duvai bāt thāpnu bhayo ra uhi hātmă dabi chādi lau sã bhanda. Cf. Nay- tadhrdi dhyānamukhamāpūrya vajrabhrtā 'stottarasatamantritam krtvā muhe mu- araj Pant, Śrī 5 Prthivīnārāyan Sabko Upades (Lalitpur: n.p., n.d.), pp. 545, 641-42. ham dei mela [mukhe mukham dehi, me]. See also Snellgrove, Indo-Tibetan Bud- This episode is graphically portrayed in a painting on the walls of the Näth Siddha dhism, vol. 1 pp. 254-62. monastery of Caughera (Srigau district, Dang, Nepal): personal communication 36. Sarada Tilaka 15.137b. The editor suggests two readings-vidhuvakkrāntare from Véronique Bouillier, Paris, January 1993. A Hindi translation of this episode and visnuvakkräntare-neither of which makes sense and which I amend to vidru- is found in Srivastav, ed., "Gorakh" Visesānk, p. 338. The crucial phrase, in Hindi, vakträntare. In the kaulaketu rite of the siddha märga, the guru's mouth is said to be is gorakbnāthjī ne wabī dabī uskī añjali mem mukb se ulat diyā aur kabā ki isko kbāo. In "encased" or "interlocked" (samputa) in or with that of his disciple: KM (London, the Hindi, the verb ulatna means both "vomit" and "reverse," as in the hathayogic Wellcome MSS no. g501) fol. 35a.7-8. ulatā sadhana. This latter reading would imply that Gorakhnath drew the yogurt 37. Yogakundali Upanisad 71-72, in Varenne, Upanishads du yoga, p. 132. These down to the base of his subtle body before reversing it and bringing it up out of verses are borrowed directly from the KhV: the Yogakundali Upanisad is composed his mouth. of the seventy-five verses of chap. I of the KhV, together with 164 "original" verses: 22. For the historical reality and political intrigue behind this legend, see Véro- Bouy, Nath Yogin, p. 41 n. 157. nique Bouillier, "The King and His Yogī: Prthivinarayan Sāh, Bhagavantanāth and 38. Jonathan Parry, "Sacrificial Death and the Necrophagous Ascetic," in Mau- the Unification of Nepal in the Eighteenth Century," in John P. Neelsen, ed., Gen- rice Bloch and Jonathan Parry, eds. Death and the Regeneration of Life (Cambridge: der, Caste and Power in South Asia: Social Status and Mobility in a Transitional Society Cambridge University Press, 1982, p. 96, citing H. W. Barrow, "On Aghoris and (Delhi: Munshiram Manoharlal, 1991), pp. 1-21. Aghoripanthis," Journal of the Anthropological Society of Bombay 3 (1893), p. 241. Cf. 23. Sachau, Alberuni's India, vol 1, p. 180. the Hevajra Tantra (2.12) description of a secret Indo-Tibetan Buddhist initiation 24. Stern, "Le temple d'Eklingjī," nn. 8, 19. This detail of Bāppā Rāwal's initia- ritual, in which the initiate is enjoined to eat the semen that his guru has shed in tion is not found in the 15th c. Sri Ekalinga Mabatmya. It is, however, related in the pudendum of a "Wisdom Maiden": Snellgrove, Indo-Tibetan Buddhism, vol. 1, Tod, Annals and Antiquities, vol. I, pp. 264-65. See above, chap. 4, nn. 226-27. pp. 258-59. 25. Mujtaba and Shah, "Taming of the Serpent," pp. 83-84. 39. KJiN 18.22b-23c (the conclusion of its chapter on initiation): vaktrādvak- 26. Abhinavagupta's discussions of the various vedba diksas are found in TĀ tram višeseņa siddhibhāgyaḥ samānyathā/ sāmānye kathitam kumbhe śankhādvak- 29.236-75. A partial English translation of the description of vedba-diksa is found tram viśesataḥ. in Silburn, Kundalini, pp. 91-103; a full Italian translation is in Raineiro Gnoli, 40. I am grateful to Alexis Sanderson for this reference. Luce delle sacre scritture (Tantrāloka) (Torino: Unione Tipografico-Editrice Torinese, 41. rasendreņa yathā viddhamayaḥ suvarņatām vrajet/ dīkșāviddhastathā hyātmā 1972), pp. 708-12. sivatvam labhate priye: Kulārnava Tantra 14.89.

5I0 51 Notes to Chapter Ten Notes to Chapter Ten

42. Sādbanamālā (vol. I, p. 82 of Bhattacharya's edition); TĀ 5.151; Parāparapra- Brahmã. On this, see O'Flaherty, Dreams, pp. 204-5; and Malamoud, "Cosmolog- kāšika 9 and Bāsava Purāna A. 38.87 (both cited in Sadasiv Balwant Kulkarni, ed. ies prescriptives," pp. 307-17. and Marathi tr., Rasaratnasamucchaya, 2 vols. [Kolhapur: Sivaji University, 1970, 57. For definitions of this temporal unit, see above, chap. 3, n. 112, and chap. 1972], vol. I [1970], pp. 23-24); Amanaska Yoga 2.48. 9, n. 137. 43. RA 17.164-65. See above, chap. 7, n. 21. 58. Bhāgavata Purāna 3.11.9-12, 16-22, 32. 44. rasaśca pavanaśceti karmayogo dvidbā matab. 59. For the tantric perspective on breath as the basis for all time reckoning, see 45. The "three fruits" are the fruits of three varieties of myrobalan; the "three Abhinavagupta, Tantrasāra, chap. 6 (pp. 46-57 of The Tantrasāra of Abbinavagupta, hot substances" are dry ginger, long pepper, and black pepper; "rock water" is a ed. with notes by Mukunda Ram Sastri [reprint ed. Delhi: Bani Prakashan, 1982].) naturally occurring mineral acid. 60. For summaries of Gorakhnäth's metaphysics, see Dvivedi, Nāth Sampradãy, 46. RA 12.331-37. pp. 114-26; and Aksaya Kumar Banerjea, Philosopby of Gorakbnāth with Goraksa- 47. As a unit of weight, one pala equals eight tolas (RA 10.34b). A tola is slightly Vacana-Sangraba, 2d ed. (Delhi: Motilal Banarsidass, 1983). more than ten grams. Therefore a pala is 82.624 grams, or about three ounces. 61. This notion is particularly transparent in Jñānakārika (2.2b-4b), a text at- This value of the pala is calculated from equivalents given in RA 10.32-34. Other tributed to Matsyendra (in Bagchi, KJnN, p. 116). Cf. Jean-Paul Sartre, Being and alchemical sources yield different equivalents. Nothingness, trans. with an introduction by Hazel E. Barnes (New York: Washing- 48. Other alchemical texts make the same claim. See for example BbP 5.255, ton Square Press, 1972), p. 56: "Nothingness lies coiled in the heart of being-like 7.224. a worm." 49. Amanaska Yoga 1.50-98, on tāraka yoga. Cf. Srivastav, "Goraksa" Višesānk, 62. Gorakb Bānī Gyān Tilak 27-28, in Srivastav, "Gorakb Bānī" Višesānk, p. 362. p. 152. 63. Mabävedba is treated on p. 47 of this edition, whose versification is totally 50. As a measure of time, one pala equals one sixtieth of a ghatika of twenty- confused. It quite identical to the description given in the 15th c. HYP 3.26-31. four minutes, i.e., twenty-four seconds. For a similar quantification, this time of the progressive angas of Pātañjala yoga, 51. This ordering respects the hierarchy of the five elements, with their corre- see Brahmananda's commentary to HYP 2.12. sponding senses. 64. Cf. Heimann's (Facets, pp. 95-100) discussion of the term sünya, which has 52. The powers acquired between days fourteen and twenty-eight correspond many of the same valences in the Sanskrit language. to the eight classical Hindu siddbis. 65. See the early Aitareya Upanisad 1.1-3.12, especially 3.11-12. This notion 53. This is Intellect, the second tattva or metaphysical category, according to carries over into the tantric worldview, in which the endpoint of emanated creation Samkhyan thought. Purusa, the universal principle alone, is higher than this. is the adbah-kundalini, who enters the human microcosm through the fontanelle, 54. I have arrived at this measure of mass by extrapolating data from the Bhāga- to slumber at the base of the system of cakras, until she is awakened by yogic prac- vata Purāna (3.11.5-6) and RA 10.32-33a. According to the former source, two tice. On this, see above, chap. 8, sec. I. paramānus make one anu, three anus constitute a trasarenu (or trisarenu), and three 66. Stephen W. Hawking, A Brief History of Time: From the Big Bang to Black Holes trasarenus equal one truți ("mote"-and one hundred trutis equal one vedha!); ac- (New York: Bantam Books, 1988), pp. 81-113, especially p. 89, in which Hawking cording to the latter, six trutis equal one liksa ("nit"), six liksas one yuka ("louse"), discusses the improbability of black holes issuing into "wormholes" in another part six yūkas one rajas ("pollen grain"), and six rajases one sarsapa (one mustard grain), of space-time. which has the mass of approximately o.004 grams. 67. RA 18.217-20. 55. Bhāgavata Purāna 3.11.4. 68. On sabdavedha, see RA 12.70 and RRS 8.95. On transmuting with bodily 56. Blaise Pascal, Les Pensées sur la religion et sur quelques autres sujets (Paris: Édi- secretions, see Yogatattva Upanisad 74, and RA 18.28. tions du Luxembourg, 1951), pp. 134-39. Douglas Adams, The Restaurant at the 69. Gorakb Bāni Sabadī 171a, 148: nīñjhar jharaņaim ammīmras pīvaņām sat dal End of the Universe (New York: Ballantine Books, 1995). A Hindu equivalent of this bedhyā jāi / ... vyand hīm jog vyand hĩm bhog vyand hĩm harai causathi rog / yã "restaurant" is the cosmological construct called Lokāloka ("World/non-World"), bind ka koī jānņaim bhev / so āpaim karatā āpaim dev. located at the outermost edge of the disk of the earth, within the cosmic Egg of 70. For a discussion of these texts, see above, chap. 5, sec. z.

512 513 Notes to Chapter Ten Notes to Chapter Ten

71. Gorakb Upanisad, in Mallik, ed., SSP, p. 74, which suggests that the outer their vital fluids, through them. In Matsyendra's Akulavira Tantra (verse 78a of ver- piercing of the ears is tantamount to the inner khecarī mudra. Cf. Briggs, Gorakb- sion "A" on p. 91 of Bagchi's edition of the K7nN), it is said that "that which is nath p. 6, citing the testimony of Naths in Hardwar. difficult for divine Siddhas is easily accessible for the Yoginīs." 72. See above, chap. 8, nn. 213-15. 84. KJiN 16.47-50; 24.4-12. On the "Naths of the Four Ages," of which Matsy- 73. See above, n. 17. endra is the fourth in this tradition, see above, chap. 5, n. 93. 74. Here, the mythic paradigm for the biological relationship between priestly 85. Matsyendra's recovery of the Kaula teachings is described in chap. 16 of teacher and disciple may go back to even pre-Vedic times. See above, chap. 9, n. the KJiN: see above, chap. 8, sec. 2. Matsyendra appears to identify himself with 104. "Siddhanātha," who revealed the akulavīra doctrines to the world: Akulavīra Tantra 75. Here I am following the alternative reading suggested in Tripathi's edition (A), vv. 1-2a; Akulavīra Tantra (B), vv. 39b, 142a. Cf. Jñānakārikā 2.1 (in Bagchi, of the RA (p. 221 n. 1). in place of haribaramagabbirab, which makes no sense, the KJiN, pp. 94, 97, 106, 116). See above, chap. 3, n. 137. editor suggests the reading baribarasamavīrab. 86. KJiN 14.40, 55b-56, 63b-65b. 76. Kularnava Tantra 5.93. Cf. KJAN 11.21-23. 87. Pāsupata Sūtras 1.33-38, quoted in Gonda, Medieval, p. 218; Candakaušika 77. This theme is especially dear to the KCM, which devotes over a fourth of (ed. Bhattacharya), pp. 109-I1. Kșemīśvara's Kāpālika alchemist is described as its content to techniques for attracting Siddha maidens, nymphs, goddesses, and females of every stripe into one's embrace. Similar themes are the frequent subject possessing eight siddhis that are quite identical to the eight Buddhist "magical pow- ers": see above, chap. 3, n. 14. of the BbP: see above, chap. 5, n. 214- 88. Harsacarita 3.112-28. 78. On the Asura siddbi of rasa-rasāyana, see above, chap. 3, nn. 34-40. On Kāvya Uśanas, see above, chap. 9, nn. 103-6. On the Bengali kavirājas, see Roşu, 89. Ocean, vol. 2, p. 236. Mahāvratin is a generic term for a Saiva ascetic-a

"Liétard et Cordier," pp. lxxxiv-lxxxv. Pāśupata or Kāpālika-referring as it does to the "great vow" (i.e., the slaying of a 79. KJiN 7.2 1a, 14.18a, 26b; RA 12.337; KCM fol. 24a.8; Goraksa Śataka 147b; brahmin followed by twelve years of expiation) undertaken by them in their initi- ation. KbV (NNA MSS no. 5-6568), fol. 9b, line 3; Siva Sambita 3.73. 90. RM (Paris MSS no. 1222, fol. 28b.8-10; 29a.1; Gondal MSS no. 861, fol. 8o. On the alternation between kavi and Kāvya Usanas's name, see Dumézil, 36b.4-9). This passage is clearly an expansion on RA 18.208-28, discussed above, Mythe et épopée, vol. 2, part 2, pp. 148-56, 204. On the incantatory element in Vedic healing, see Zysk, Religious Healing, pp. 8, 241, citing Sayana's commentary to Atb- chap. 5, nn. 163-66. This apotheosis corresponds quite closely to the goal of the Pāsupata, as described in the Pasupata Sūtras: "(one) moves unobstructed every- arva Veda 4.2.6. Cf. Padoux, Vāc, p. 6. 81. The archaic cults of these beings go back to at least the first centuries of the where; being equipped with these qualities one becomes the great chief of the ganas of Bhagavan Mahādeva." See above, chap. 4, n. 85. common era, in Hindu and Buddhist traditions alike. The Siddhas and Vidyādharas 91. RA 7.58. are listed as beings "born from divine wombs" in the 6th c. A.D. Amarakośa I.1.11. 92. MBhT 7.36. See also Jean Przyluski, "Les Vidyaraja," Bulletin de l'Ecole Francaise d'Extrême Orient 93. Siva Sambitā 4.46; 5.202, 204. 23 (1923), pp. 301-18; J. A. van Buitenen, "The Indian Hero as Vidyādhara," in Milton Singer, ed., Traditional India: Structure and Change (Philadelphia, American 94. For a complete account of this technique, as it is found in the RRS and other alchemical sources, see above, chap. 7, n. 109. Folklore Society, 1959), pp. 99-105; and David Seyfort Ruegg, "Sur les rapports," 95. AK 1.1.61. p. 83. As Przyluski notes (p. 317), these demigods figure in the entourage of Siva (Gorakh) in Hinduism, and Avalokiteśvara (Matsyendra) in Buddhism. 96. RA 12.252-58, especially 12.254 and 257. 82. K7nN 8.3oa. Italics my own. In his Yogabīja (63), Gorakhnāth speaks of Sid- 97. RA 11.104b-106. Cf. 12.337. 98. RA 18.222b-227. Cf. Amanaska Yoga 1.98. dhas through whose grace (krpa) one becomes a yogin. 99. RA 18.228. See also RA 11.107: "There where the gods are absorbed [at the 83. KJnN 9.1. Yet, one is enjoined to worship the same three figures in 18.4b. end of a cosmic eon], there too the Siddha is absorbed." In fact, three half-verses The term yogini is employed both for human "witches" or "sorceresses" with whom follow RA 18.228. Cf. Yogabija 65, which states whereas the three great gods pass tantric practitioners had commerce and for the goddesses to whom they offered away with the universal dissolution, the perfected yogin is indestructible.

514 Notes to Chapter Ten Notes to Chapter Ten

100. See above, chap. I, note 13. 114. The "Raivatācala Māhātmya" constitutes chapters ten through thirteen of 101. This is the text of an inscription in the Visvanath Temple at Benares Hindu the Jain Satruñjaya Mābātmya (translated in James Burgess, Report on the Antiquities University. Cf. RM fol. 8a.8; 26a.2; BbP 2.1; and RA 15.16: "Eaten, (mercury) ef- of Kathiawad and Kacch, Being the Result of the Second Season's Operations of the Archaeo- fects the destruction of aging and poverty." logical Survey of Western India, 1874-1875 [London: India Museum, 1876; reprint 102. Caraka Sambitā 6.1.78, 8o. Delhi: Indological Book House, 1971], p. 157, note). See above, chap. 4, n. 213 for 103. Stein, "Jardins en miniature," pp. 53-55, 58. other Jain references to this site. 104. Ibid., p. 57, citing the Yun-ki ts'i-tsein (cited in P'ei-wen yun-fou). 115. The bulk of the Matsya Purana is older than this; the praise of the Narmada 105. Ibid., p. 58; and idem, "Architecture et pensée en Extrême-Orient," Arts River region in which Raivataka is mentioned is a late addition made by a Saiva Asiatiques 4:3 (1957), pp. 176, 185. resident of Maharashtra: Surinder Mohan Bharadvaj, Hindu Places of Pilgrimage 106. Bbagavata Purana 5.16.7. This passage describes that portion of Meru in India (A Study in Cultural Geograpby) (Berkeley: University of California Press, which rises up from the earth's surface: a mirror image of this mountain extends 1973), pp. 66-67; R. C. Hazra, Studies in the Puranic Records on Hindu Rites and Cus- below the surface of the earth, into the subterranean worlds of the demonic beings toms, 2d ed. (Delhi: Motilal Banarsidass, 1975), p. 46; S. G. Kantawala, Cultural who inhabit them. The "lower half" of Meru is of lesser dimensions than the upper History from the Matsya Purāna (Baroda: M.S. University, 1964), appendix III. half, according to the Puranic texts. On this, see Ali, Geograpby, p. 48. 116. Skanda Purāna 7.2.1-15 (on Girnar). On the dating of this Purana, and of 107. Ibid., p. 49; and I. W. Mabbett, "The Symbolism of Mount Meru," History book seven, the Prabhāsa Khanda in particular, see Hazra, Purānic Records, p. 165. of Religions 23 (1983), pp. 68, 71. The Bhagavata Purāna (5.24.4) locates the Siddhas, 117. Watters, On Yuan-Chwang's Travels, vol. 2, pp. 248-49; Beal, Su-yu-ki, vol. Vidyädharas, and Cāranas at the highest atmospheric (but not heavenly) level, im- 2, p. 269. Hsuan-tsang's contemporary, the Indian author Banabhatta, describing a mediately below the spheres of the sun and Rähu, the "descending node" of the conclave of such "rishis," names the Pāsupatas: Harsacarita 8.226. moon. See above, chap. I, n. II. 118. On the identification of Raivata and Gomanta, see Vettam Mani, Puranic 108. Here, I concentrate on TA 8.119-38 (vol. 4, pp. 1441-51), with the com- Encyclopedia (Delhi: Motilal Banarsidass, 1975), s.v. "Gomanta I." This is the first mentary of Jayaratha, who indicates selected passages borrowed from the Svac- English edition, a translation of the original 1964 Malayalam edition. chanda Tantra (10.424-51). On the millennarian importance of this latter text and 119. Mababharata 2.13.53 of the Bengali (B1m.2-4) and Bombay Government col- the cult of Svacchanda Bhairava in Kashmir, see Sanderson, "Saivism: Saivism in lection (D) manuscripts only. The critical reading is Bhavanta. Another peak men- Kashmir," in Encyclopedia of Religion, vol. 13, p. 16. tioned in the Mahabharata has also been identified with Girnar: this is Ujjayanta, 109. Vajranka in TA 8.128, but vajranga ("Lightning-Limbed") in SuT 10.446, which the epic (3.86.18-20) describes as one of the holy places of Saurashtra (i.e., which adds that the "lowest-level Vidyadharas are travelers on the winds of the eastern Gujarat). The Mabābhārata (2.42.8) names, without describing it, a Raiva- mind (manahpavanagaminab)." tāka Hill, which it also locates in Gujarat. The site has also been called Girinagara IIO. TA 8.133. The names of these Siddhas are [go]rocanā, añjana, and bhasma. and Girinārāyana, of which Girnar is a vernacularization. Gorocana is in fact an organic dye having the same intense yellow color as orpiment 120. On this king, one of the earliest mythic devotees of Rudra-Siva, and his (auripigmentum). Gorocana is made from the urine of the cow. use of mountain caves, see Dumézil, Mythe et épopée, vol. 2, pp. 96-105. I1I. SuT 10.452. "Love's body" is a siddbi enjoyed by the consummate alchemist 121. The passage concerning Gomanta is found only in the Bombay and Cal- (RA 12.366: madana iva sukāntib) and yogin (see above, n. 53). cutta recensions of the Harivamsa (2.40, entitled "The Climbing of Gomanta" or 112. See above, chap. 3, n. 137, and chap. 8, n. 68. "The Journey to Gomanta"); in the critical edition, it forms a portion of appendix 113. The thirty-sixth and final chapter of the Tantraraja Tantra (pp. 115-21 of 17 and all of appendix 18, found in vol. 2, pp. 92-98 (lines 380-507). Cf. Mani, Woodroffe's edition), entitled "The Siddha," seems to second this account of the Purānic Encyclopedia, s.v. "Gomanta I." Siddhas, albeit in a rather tame way (p. 121): "The Siddha has prior (to death) been 122. Harivamśa, appendix 17, lines 381-82, 386. Girnar is a cluster of peaks, of freed from (attachment to) the body ... and whenever, wherever, and howsoever which two twin crags, today identified by Hindus as Gorakh and Dattātreya, are he may leave the body, he goes to the Good Path, for he was liberated whilst liv- by far the highest. Lines 390-91 state that Krsna and Balarama would later defeat ing (jīvanmukta)." Jarāsandha at that site; the battle is drawn in line 487.

516 517 Notes to Epilogue Notes to Epilogue 123. Harivamsa, appendix 18, lines 448-49. See above, n. 106, citing the later 9. T. R. Anantharaman, "Transformations-Metallurgical and Mental," Prof. (post-6th c. A.D.) Bhagavata Purana (5.24-4), which gives a similar description of N. P. Gandhi Memorial Lecture, Varanasi, 18 December 1973, pp. 1-25. Mount Meru. 10. Personal communication, 3 November 1984, Kathmandu. On this, see also 124. Burgess, Report, p. 159. Yogananda, Autobiography, p. 163. 125. Girnar figures prominently in both the 13th c. YSA and the early 19th c. 11. Personal communications from Surya Kumar Yogi, Bhilbara (Rajasthan), Natha Caritra of Man Singh, which devotes no less than fourteen chapters (1.2; March 1985; and Narharinath Yogi, Kathmandu, Nepal, October 1984. See Kakar 2.1-13) to the site and in particular to Gopicand's travels there. (Shamans, p. 182) for names and descriptions of a number of similar illustrious 126. Grierson, "The Song of Mánik Candra," p. 209; Temple, Legends, vol. 2, figures. p. 375. 12. Trevor Fishlock, India File (London: John Murray, 1983; reprint Calcutta: 127. Dasgupta, Obscure, p. 207; and Marie Lecomte-Tilouine, "Des dieux aux Rupa and Co., 1984), p. 36, quoting the advertising of a "sex clinic." sommets (Népal)," in Veronique Bouillier and Gérard Toffin, eds., Classer les Dieux? 13. Tayumanavar, Vannam, unpublished translation by Swami Sevananda. Des Panthéons en Asie du Sud [Purusārtha 15](Paris: EHESS, 1993), pp. 153-72, esp. 14. MBbT 2.5-6; Suśruta Sambitā 1.14.14-15; but see 6.15.32 for a figure of six pp. 159-62. to seven days. 128. Punjab States Gazetteer, vol. 22A, Chamba State with Maps, 1904 (Lahore: 15. According to both the medical and folk traditions, it takes forty drops of Civil and Military Gazette Press, 1910), pp. 183-84. blood to produce one drop of semen: Carstairs, The Twice-Born, pp. 83-84. Cf. 129. See above, chap. 4, n. 156. SSP 1.73. 130. The Hindu and Jain Vidyadharas had their Buddhist homologues in the 16. Conception joins the four lower mababbutas, already present in semen, to Vidyārājas: Reugg, "Sur les rapports," p. 83. In the Mañjuśri Mlakalpa, the bodhi- ether, the empty space of the female womb. Also present are mind and karma, sattva Vajrapāņi, the foremost of the Vidyārājas, is iconographically placed to the attached to the soul: Dasgupta, History, 2:302, 307. left of Sākyamuni: Snellgrove, Buddhist Himalaya, p. 287. I have noted (chap. 3, nn. 17. Caraka Sambitā 4.2.11-12; Śarngadbara Sambitā 1.6.12-13; MBbT 2.13-14; 75-91) Nāgārjuna's connection with Tārā: as the leader of the Mahāvidyā god- SSP 1.71. When semen is predominant, the fetus is male and spherical; when desses, she is called "Vidyārajñi": Bhattacharya, History, p. 225. blood, it is female and elliptical; when equal, is hermaphrodite and hemispherical: Dasgupta, History, 2:314. In alchemy, similar genders are assigned to diamonds Epilogue (vajras) of different shapes: RA 6.69-71. I. E. M. Forster, The Hill of Devi (Harmondsworth: Penguin, 1963), p. 65. 18. Gananath Obeyesekere, cited in Kakar, Shamans, p. 234. On the myth of 2. See above, chap. 3, n. 143. King Moon, see above, chap. 2, nn. 38-40. 3. See above, chap. 4, n. 156. 19. In one of his mystic poems on hathayogic practice, Gorakhnāth identifies 4. On this process in medieval Europe, see Jacques LeGoff, Pour un autre the vulva as a vampiress and tigress: Gorakb Bāni Pad 48.1-3. An identical sentiment Moyen Age, Temps, travail et culture en Occident: 18 essais (Paris: Gallimard, 1977), is voiced by Mayana, mother of the Nath Siddha Gopicand, in Gopicandrer Gān, p. 231. cited in Dasgupta, Obscure, p. 246 n. 1. The alchemical RA (18.103-6) provides a 5. Personal communication from Siddhinandan Misra, Benares, March 1985. mantra by which the practitioner may protect his semen, mercury, blood, flesh, and 6. Personal communication from Siddhinandan Misra, Benares, March 1985. A bones from hordes of goddesses and succubi who would trouble his dreams and similar experiment, in which the production of gold from mercury was authenti- suck him dry in the night. This may be a reference to early kula tantrism, which cated by a recognized authority, C. P. N. Singh, is documented by S. N. Khandel- imagined the "extraction" of vital fluids from within the body by yoginīs and sakinīs, wal in the 6-12 November 1983 edition of Saptabik Hindustan, pp. 41-44. who would in turn offer the same to Mahabhairava enthroned in the heart (Sand- 7. Sanjukta Gupta, "The Mandala as an Image of Man," in Richard Gombrich, erson, "Purity," p. 213 n. 89, citing Netra Tantra 20.1-40 and other sources). See ed., Indian Ritual and Its Exegesis (Delhi: Oxford University Press, 1988), p. 35 above, chap. 5, n. 84. 8. Paramahansa Yogananda, Autobiograpby of a Yogi, 2d Indian ed. (Bombay: 20. Svoboda, Aghora, pp. 280-81. On ojas, see Dasgupta, History vol. 2, p. 343 Jaico, 1975), p. 70. n. 2, citing Cakrapāņi's commentary to Caraka Sambitā 1.30.6.

518 519 Notes to Epilogue Notes to Epilogue

21. On this and related attitudes as rationales for the practice of widow burning, 32. Tod, Annals and Antiquities, 1:300; and 2:298 ff., cited in Kolff, p. 82. Italics see the masterful work of Catherine Weinberger-Thomas, Cendres d'immortalité: la my own. crémation des veuves en Inde (Paris: Editions du Seuil, 1996; translation forthcoming, 33. George A. Grierson, "Some Bihari Folk-Songs," 7RAS 16 (1884), p. 236, University of Chicago Press). cited in Kolff, p. 77. 22. This claim was displayed prominently on the billboard over the storefront 34. Van der Veer, Gods on Earth, pp. 126-30, 176. of a gupta rog clinic in the Lanka district of Benares in 1985. Recall here that the 35. Yubaraj Ghimire, "The Rise of the Sadhus," India Today, pp. 61-63. A photo- eighth and ultimate limb of traditional Āyurveda is vājīkarana, sexual therapy: see graph on page 62 shows Avedyanäth seated, sixth from the left, with the other eight above, chap. 2, n. 46. members of the VHP brain trust. See also Véronique Bouillier, "La violence des 23. Here I am simply summarizing two detailed studies by Véronique Bouillier: non-violents, ou les ascètes au combat," in Denis Vidal, Gilles Tarabout, and Eric "The King and His Yogi," pp. 1-21; and "Growth and Decay of a Kanphata Yogi Meyer, eds., Violences et non-violences en Inde (Purusartha, vol. 16) (Paris: Editions Monastery in South-west Nepal," The Indian Economic and Social History Review 28:2 de l'EHESS, 1994), pp. 213-43 and especially p. 233 for a historical overview of (1991), pp. 151-70, especially pp. 152-60. this phenomenon. 24. This legend forms chapter 20 of the late 19th c. Śri Mastnath Carita (Acts 36. Cf. Van der Veer, Gods on Earth, p. 133: "Ascetics use violence on their own of the Illustrious Mastnäth) of Sankarnäth Yogi, which places the action in the for- bodies to acquire power over the microcosm of the body and over the connected tified city of Chittor, in southeastern Rajasthan. macrocosm of nature, and they use violence to acquire power in society." 25. Padmaja Sharma, Mabaraja Man Singh of Jodbpur and His Times (1803-1843 37. See above, chap. 4, n. 204. The dire and irreversible effects of a Yogi's curse A.D.) (Agra: Shiva Lal Agarwala & Company, 1972), pp. 155-56. are feared throughout India. E. M. Forster takes a different view, as stated in the 26. Tod, Annals and Antiquities, vol. 2, pp. 825-27. The most complete and in- epigraph to this chapter. sightful accounts of the relations between Man Singh and Ayas Dev Nath and his 38. See above, chap. 3, n. 6. successors are Sharma, Mabaraja Man Singb, pp. 153-82; and Daniel Gold, "The 39. Oman, Cults, "A Group of Yogis, One Man Enjoying His Churrus Pipe," Instability of the King: Magical Insanity and the Yogi's Power in the Politics of facing p. 4; "A Party of Wandering Yogis," facing p. 29. Cf. pp.7,29. Jodhpur, 1803-1843," in David N. Lorenzen, ed., Bhakti Religion in North India: 40. See above, chap. 4, n. 220. Community Identity and Political Action (Albany, N.Y .: SUNY Press, 1995), pp. 41. rank se rāv rāv se rankā, chinmem karaim nabīm kuch šankā: Śankarnāth, Śri 120-32. Mastnath Carita, p. 108. 27. Peter van der Veer, Gods on Earth: The Management of Religious Experience and 42. Illustrious Nath Siddhas are inhumed under burial tumuli called samādhis Identity in a North Indian Pilgrimage Centre, London School of Economics Mono- after death: see above, chap. 4, n. 71; and chap. 9, n. 113. graphs on Social Anthropology, 59 (London: Athlone, 1988). 43. The most common form of greeting between Nāth Siddhas, ades means 28. The ascetic sects were, by the 178os, the dominant money-lending and "[what is your] command [?]"; a mystic interpretation holds the term means "Ādi property-owning group in Allahabad, Benares, and Mirzapur: ibid., p. 134, cit- [-nāth] is Lord [īśa]." ing C. A. Bayly, Rulers, Townsmen and Bazaars: North Indian Society in the Age of 44. On this serpent-keeping and -charming suborder of the Nāth sampradāya, British Expansion, 1770-1870 (Cambridge: Cambridge University Press, 1983), p. see Briggs, Gorakhnath, pp. 59-61 (who calls them Sepalas); and Rose, Glossary, vol. 143. 2, p. 409. 29. David Lorenzen, "Warrior Ascetics in Indian History," Journal of the Ameri- 45. Like many of the itinerant Näth suborders, the Sampelas generally live by can Oriental Society 98 (1978), pp. 68-70. begging (and by snake charming). 30. Dirk H. A. Kolff, Naukar, Rajput and Sepoy: The Ethnobistory of the Military 46. See above, chap. 9, n. 28, for this Brahmanic identification. Labour Market in Hindustan, 1450-1850 (Cambridge: Cambridge University Press, 47. See above, chapter 7, part 3. 1990), pp. 74-85: "The warrior-ascetic in song, ballad, and legend." 48. "Om, Victory to the Lord of Animals!" Pasupatinath, located in eastern 31. Ibid., pp. 76, 81. Kathmandu, is the national shrine of Nepal.

520 Notes to Epilogue

49. See above, chap. 9, n. 173, for this identification. SELECTED BIBLIOGRAPHY 50. The "wallet" in which Nath Siddhas carry their numerous paraphernalia. The jboli is the Siddha's magical bag of tricks. 51. See above, chap. 8, n. 220. 52. I am grateful to Bob Ladd, of the Department of Materials Sciences at the University of Virginia, for carrying out electromicroscopic spectroscopy on my sample.

Manuscripts Anandakanda of Mahābhairava, 907 fols. Jamnagar. GAU MSS no. 830/2. This is a copy of the Tanjore Sarasvati Mahal Sanskrit Library manuscript upon which S. V. Radhakrishna Sastri's 1952 edition is based. Āścaryayogamālā of Nāgārjuna, 45 fols. Bikaner. ASL MSS no. 3941. A.D. 1674. Carpat Rasāyan of Carpatnāth, 2 fols., incomplete (padas 19-52). Jodhpur. MSL MSS no. 1178/669 (Hindi). Dattagoraksasamvāda. Jodhpur. MSL MSS no. 1557. This work claims to comprise chapters 27-29 of the Tantra Maharnava. Dhatukalpa of the Rudrayamala Tantra, 112 fols. Jamnagar. GAU MSS no. 843/18. Gandbakakalpa of the Rudrayāmala Tantra. Jamnagar. GAU MSS no. 835/7. Itself a work in twenty-three chapters, this work claims, in its colophon, to be a portion of chapter 28 of the Rudrayamala Tantra. Ghodā Colī. Jaipur. Śrī Rāmacaraņa Prāya Vidyāpīth MSS no. 1.14.ii.19. Ghodācoli-vākya, 1 fol. Jodhpur. MSL MSS nos. 1481, 1482[a], 1483[b]. Goraksa Sambitā, 34 fols. (incomplete). Jamnagar. GAU MSS no. 841/13. This is a compendium of passages from various tantric works, including the Kaksaputa. It bears no resemblance to either the "Sakta" Goraksa Sambitā or the Bhütipra- karaņa. Goraksa Sambitā, Bbūtiprakarana, 174 fols. Jodhpur. MSL MSS no. 1496/1431. 18th c. -, 59 fols. Jodhpur. MSL MSS no. 1497/1432. 19th c. Goraksa Sambitā, Svacchandasaktyavatāra, Bhūtiprakarana, 119 fol. Jodhpur. MSL MSS no. 1499/1434 (V.S. 1881). See also below, Svaccbandasaktyavatāra. -, Mabamanthanabhairava Tantrantargatab. Jodhpur. MSL MSS no. 1502/ 1437, 310 fol. This is the manuscript of the "Sakta" Goraksa Sambita that Pandey edited as Goraksa Sambitā, vol. I (1976). . Kathmandu. NNA MSS no. 5-3978 (Ayurveda 89). Reel no. A-213/21, 118 fols. This is a copy of the Bhūtiprakarana.

521

522 523 Selected Bibliograpby Selected Bibliography

Goraksa Sambitā bhāsa. Jodhpur. MSL MSS no. 649 (Hindi). This is a Hindi com- Matsyendra Sambitā, 179 fol. Jodhpur. MSL MSS no. 1604/1782. 19th c. mentary on the "Sākta" Goraksa Sambitā. Natha Caritra of Mansimha, 85 fol. Jodhpur. MSL MSS no. 1573/1645. 19th c. Kākacandeśvarīmata, 32 fols. Kathmandu. NNA MSS no. 5-3969. Reel no. A211/19. Pāradakalpa of the Rudrayāmala Tantra, 244 fols. Jamnagar. GAU MSS no. 21/849. Itself a work in twenty-five chapters, this work claims, in its colophon, to be a -, 57 fols. Kathmandu. NNA MSS no. 3-118 (Ayurveda 71). Reel no. A-211/8. portion of chapter 28 of the Rudrayamala Tantra. See also below, Rudrayāmala- tantrapāradakalpa. -, 37 fols. Kathmandu. Reel no. E-1796-8b. London. Wellcome Institute for the History of Medicine MSS no. g473. Rasabrdaya of Śrigovinda-bhagavat-pād. Kathmandu, NNA MSS no. T118/277 (Āyurveda). This is a copy of NNA MSS no. 3-118. Rasakautuka, 5 fols. Bikaner. ASL MSS no. 4203. , 43 fols. Bikaner. ASL MSS no. 3952. Rasaprabodha of Nāgadeva, 13 fols. (incomplete). Bikaner. ASL MSS no. 4216. Kaksaputam of Srī Nāgārjuna, 140 fols. Jodhpur. MSL MSS no. 2959/1318. 19th c. of Siddha Nāgārjuna, 40 fols. Jodhpur. MSL MSS no. 2960/1319. 18th c. Rasarājaširomaņi of Paraśurāma. Bikaner. ASL MSS no. 4258.

Kautukacintāmani of Pratāpadeva, 53 fols. Bikaner. ASL MSS no. 3967. Rasaratnadīpika of Rāmarāja, 31 fols. Bikaner. ASL MSS no. 4222.

Khecarī Patala of Ādinātha, 19 fols. Jodhpur. MSL MSS no. 1469/1375. A.D.1726. Rasaratnākara of Nityanātha. This work is divided into five sections (khandas), which appear as separate manuscripts and edited works. They are the (1) Rasa; This is a manuscript of the Khecarī Vidyā. Khecarī Vidyā of Ādinātha, 16 fols. Jodhpur. MSL MSS no. 1468/1374. 19th c. (2) Rasendra; (3) Vāda; (4) Rasāyana; and (5) Mantra-khaņdas. The Rasa, Vāda,

, 23 fols. Jodhpur. MSL MSS no. 1470/1376. 19th c. and Rasāyana-khandas have been edited; the Rasendra and Mantra-khandas

, 15 fols. Jodhpur. MSL MSS no. 1471/1377. A.D. 1683. only exist in manuscript form. Rasa Khanda, 68 fols. Kathmandu. NNA. Reel no. D2/7. -, 9 fols. Jodhpur. MSL MSS no. 1472/1378. 19th c. -, 15 fols. Kathmandu. NNA MSS no. 6-1636 (Jyautisa 7). Reel no. A-999/ Rasa and Rasendra Khandas, 99 fols. Kathmandu. NNA. Reel no. H108/1. Rasa and Rasendra Khandas, 134 fols. Kathmandu. NNA. Reel no. E2086/3. 7. A.D. 1678. -, 15 fols. Kathmandu. NNA. Reel no. M-23/10. . Siddha [Mantra] Khanda, 22 fols. Kathmandu. NNA MSS no. 5-3092 (Ãy-

-, 17 fols. Kathmandu. Kaiser Library MSS no. 316 (Yoga). Reel no. C32/12. urveda 266) Reel no. B163/19.

-, 11 fols. Kathmandu. NNA MSS no. 4-1817 (Tantra 347). Reel no. A- -. Siddba [Mantra] Khanda, 49 fols. Kathmandu. NNA. Reel no. H253/20.

1289-9. Rasārnava, 106 fols. Bikaner. ASL MSS no. 4256. A.D. 1675. Erroneously cata-

Kriyākālagunottara Tantra, 14 fols. Kathmandu. NNA MSS no. 3-392 (Śaivatantra logued under the title of Rasarājasankara.

67). Reel no. B25/32. A.D. 1184. -, 109 fols. Bikaner. ASL MSS no. 4273. A.D. 1627.

-, 88 fols. Kathmandu. NNA MSS no. 5-4947. Reel no. A149/2. , 100 fols. (incomplete). Bikaner. ASL MSS no. 4274.

-, 88 fols. Kathmandu. Kaiser Library MSS no. 297. Reel no. C30/16. Rasasāra of Govindācārya, 81 fols. (incomplete). Bikaner. ASL MSS no. 4260. A.D. 1649. Kubjikāmata. London. Wellcome Institute for the History of Medicine. MSS no. Rasasindbu of Vişnudeva, 129 fols. Bikaner. ASL MSS no. 4267. A.D. 1564. g501. This is a copy of NNA MSS no. 1-285 ka. Madanavinodanigbantu of Rāmarāja, 71 fols. Kathmandu. NNA MSS no. 4-2224 Rasāvatāra, 10 fols. Bikaner. ASL MSS no. 4269.

(Āyurveda no. 199) / reel no. A1289/4 Rasendrakalpadruma of Rāmakrsnabhatta, 51 fols. Bikaner. ASL MSS no. 4280.

Mahākālatantrarājā. Kathmandu. Reel no. E-1358/7.93 fols. in Newari script. This Rasendramangala of Nāgārjuna, 45 fol. Bikaner. ASL MSS no. 4281.

is the same work as the Nathabbyudayatantra, referred to by Bu-ston in his 39 fols. Paris, Bibliotheque Nationale. Fonds Palmyr Cordier. MSS no.

History of Buddbism. See Chos-bbyung of Bu-ston. 1222 (Sanscrit). This is a copy of ASL MSS no. 4281.

Mabākālayogašāstra, 13 fols. NNA MSS no. 5-6568. Reel no. Azo7/6. This is a , 37 fol. Jamnagar. GAU MSS no. 862/34.

manuscript of the Khecarī Vidyā. Rudrayāmalatantrapāradakalpa, 144 fols. Bikaner. ASL MSS no. 4290. A.D. 1662. See also Pāradakalpa of the Rudrayāmala Tantra.

524 525 Selected Bibliograpby Selected Bibliography

Śivakalpadruma of Sivanātha, 45 fols. Bikaner. ASL MSS no. 4349. bCud len gyi man ngag of Bo dong. In Encyclopedia Tibetica: The Collected Works of Bo Svacchandaśaktyavatāra of Śrīkaņtha, 130 fols. Bikaner. ASL MSS no. 4401. This is don Pan-chen phyogs-las-rnam-rgyal, 13 vols. Ed. S. T. Kazi. New Delhi: Tibet the Bhūtiprakarana. See also Goraksa Sambitā. House, 1970, vol. 2, pp. 507-601. Yogaratnāvali of Śrīkaņtha Šiva, 175 fols. Bikaner. ASL MSS no. 4184. A.D. 1640. Bhāgavata Purāņa. Śrimadbbāgavata Purāņa, 2 vols. Ed. and trans. C. L. Goswami and M. A. Sastri. Gorakhpur: Gita Press, 1971. Primary Sources: Bhūtaśuddbi Tantra. In Tantrasamgraba, vol. 3, pp. 565-625; and vol. 4, pp. 308-16. Advayavajrasamgraba. Ed. Haraprasad Shastri. Gaekwad's Oriental Series, no. 40. Bhūtiprakarana. Goraksa Sambitā (Part Two). Ed. Janardana Pandeya. Sarasvati- Baroda: Oriental Institute, 1927. bhavana-Granthamala, vol. 111. Benares: Sampurnananda Sanskrit Visvavidya- Aitareya Åranyaka. With the commentary of Sāyana. Ed. Rajendralal Misra. Cal- laya, 1977. cutta: Ganesa Press, 1876. Brabmānda Purāna. Brabmānda Purāna of Sage Krsna Dvaipāyana Vyāsa. Ed. J. L. -. Trans. Arthur Berriedale Keith. Oxford: Clarendon Press, 1969. Shastri. Delhi: Motilal Banarsidass, 1973. Aitareya Upanisad. Edit. and trans. into French by Lillian Silburn. Paris: Adrien Brhadāranyaka Upanisad. Ed. and trans. into French by Emile Sénart. Paris: Les Maisonneuve, 1950. Belles Lettres, 1934. Amanaskayoga of Goraksanātha. Ed. Ramlal Srivastav. Gorakhpur: Gorakhnath Brbat Sambitā of Varāhamihira. Varābamibira's Brbat Sambitā, 2 vols. Ed. and trans. Mandir, 1980. by M. Ramakrishna Bhat. Delhi: Motilal Banarsidass, 1981, 1982. Ed. Yognath Swami. Poona: Siddh Sahity Samsodhan Prakasan Mandal, Candakausika of Ksemīśvara. The Canda-Kausika. Ed. Sibani Das Gupta. Calcutta: 1967. The Asiatic Society, 1962. Amarakośa. Śrīmad Amarasimbaviracitam nāmalingānušāsanam Amarakośa, 2d ed. Caraka Samhitā, with the commentary of Cakrapānidatta. Ed. with Hindi transla- Ed. with Hindi and Sanskrit commentaries by Brahmananda Tripathi. Chau- tion by Yadavji Trikamji Acarya, Chaukhamba Ayurvijnan Granthamala, 34. khamba Surbharati Granthamala, no. 52. Benares: Chaukhamba Surbharati Bombay: 1941; reprint Benares: Chaukhamba Surbharati Prakashan, 1992. Prakashan, 1982. Caraka Sambitā. Caraka Sambitā: Agnivesa's Treatise Refined and Annotated by Caraka Amaraugha Prabodha of Goraksanātha. In Siddba Siddhānta Paddhati of Goraksa- and Redacted by Drdhabala: Text with English Translation. Ed. and trans. Priyavrat nātha. Sharma. Jaikrishnadas Ayurveda Series, no. 36. Benares: Chowkhamba Orien- Amaraugbaśāsana of Gorakşanātha. Amaraugbašāsanam. Ed. Mukundaram Shastri. talia, 1992. Kashmir Series of Texts and Studies no. 20. Bombay: Nirnay-sagar Press, 1918. Caryagiti. An Anthology of Buddbist Tantric Songs: A Study of the Caryagiti. Ed. and Ānandakanda. Ed. Sri S. V. Radhakrishna Sastri. Tanjore Sarasvati Mahal Series, trans. Per Kværne. Oslo: Universitetsforlaget, 1977; reprint Bangkok: White 15. Tanjore: TMSSM Library, 1952. Orchid, 1986. Åpastambha Dharma Sütra. With the commentary of Ujjwala. Ed. A. Chinnaswami Chändogya Upanisad. Ed. and trans. into French by Emile Sénart. Paris: Adrien Mai- Sastri and A. Ramanatha Sastri. Haridasa Sanskrit Granthamala, no. 93. Be- sonneuve, 1930. nares: Chowkhamba Sanskrit Series Office, 1932. Chos-bbyung of Bu-ston. History of Buddbism (Chos-hbyung) by Bu-ston, 2 vols. Trans- Arthašāstra of Kautilya. The Kautilya Arthašāstra, 2 vols. Ed. R. P. Kangle. Bombay: lated by Ernst Obermiller. Heidelberg: Harrassowitz, 1931-32; reprint Delhi: University of Bombay Press, 1960. Sri Satguru Publications, 1986. Astänga Hrdaya of Vagbhatta. Astangabrdaya with the commentary of Hemadri. Dam-pa'i-chos-rin-po-che-'phags-pa'i-yul-du-ji-ltar-dar-ba'i-tshul-gsal-bar-ston-pa- Bombay: Nirnaysagar Press, 1925. dgos-'dod-kun-bbyung of Taranatha. Taranatha's History of Buddhism in India. Ed. Astānga Samgraba. Śrīmadvāgbbattaviracita Astāngasamgraba, 2 vols. Ed. with a Debiprasad Chattopadhyaya. Trans. Lama Chimpa Alaka Chattopadhyaya. Hindi translation by Kaviraj Atrideva Gupta. Bombay: Nirnay Sagar Press, n.d. Simla: Indian Institute of Advanced Study, 1970. Atharva Veda. Atharvaveda Sambita, 2 vols. Trans. with a critical commentary by Darśana Upanisad. In Yoga Upanisads. William D. Whitney. Revised and edited by Charles R. Lanman. Harvard Ori- Dattātreya Tantra. In Indrajālavidyāsamgraba. Ed. Jivananda Bhattacharya. Calcutta: ental Series, vols. 7-8. Cambridge: Harvard University Press, 1905. V. V. Mukherji, 1925, pp. 132-79.

526 527 Selected Bibliography Selected Bibliograpby

Dbyānabindu Upanisad. In Yoga Upanisads. commentary of Brahmānanda. Ed. and trans. Srinivasa Iyengar. Madras: Ad- Ekalinga Mābātmyam. Premlata Sharma, Ekalingamāhātmyam (Ekling Mandir kā yar, 1972. Sthalapurāņa evam Mewād ke Rāj-Vamš kā Itibās. Delhi: Motilal Banarsidass, Himavat Khanda of the Skanda Purāna. Himavatkhandab. Ed. Yogi Narharinath. 1976 n.p .: 1957. Gheranda Sambita. Ed. and trans. Sisa Candra Vasu. Madras: Adyar, 1933. Hudud al-'Alam. Hudud al-'Alam, "The Regions of the World": A Persian Geography Goraksa Paddbati. Ed. with commentary by Mahidhara Sharma. Bombay: Khema- 372 A.H .- 982 A.H. Trans. and explained by V. Minorsky. Karachi: Indus Publica- raja Srikrishnadas, 1983. tions, 1980. Goraksa Sambitā. Goraksa Sambitā (Part One). Ed. Janardana Pandeya. Sarasvati- Jaiminīya Brābmana. Jaiminīya Brabmana of the Samaveda. Ed. Raghu Vira and bhavana Granthamala, vol. 11O. Benares: Sampurnananda Sanskrit Visvavidya- Lokesh Chandra. Sarasvati Vihara Series, vol. 31. Nagpur: Arya Bharati, 1954- laya, 1976. Jaiminīya Upanisad Brabmana. Ed. and trans. Hans Oertel in Journal of the American Goraksa Śataka. Das Goraksaśataka. Ed. Karl A. Nowotny. Wetterschlick b. Bonn: Oriental Society 16:1 (1894), pp. 79-260. Richard Schwarzbold, 1976. Jñānakārikā of Matsyendranātha. In Kaulajnānanirņaya of Matysyendranātha. Goraksa Siddbānta Samgraba. Ed. Gopinath Kaviraj. Benares: Vidya Vilas Press, Kādambarī of Bāņabhatta. Bāna's Kādambarī (Pūrvabbāga Complete). Ed. M. R. Kale, 1925 4th revised ed. Delhi: Motilal Banarsidass, 1968. -. Ed. Janardana Pandeya. Sarasvatibhavana Granthamala, no. I1O. Benares: -. Kadambarī: A Classical Sanskrit Story of Magical Transformations. Trans. with Varanaseya Sanskrit Visvavidyalaya Press, 1973. an introduction by Gwendolyn Layne. New York: Garland Publishing, 1991. Goraksa Vijay. Gorakba Vijaya of Bhīmasena Rāy. Ed. Pancanand Mandal. Calcutta: Kākacandīšvara Kalpatantra. Ed. Kailashpati Pandey. Kashi Sanskrit Granthamala, Visvabharati Granthalaya, 1949. no. 73. Benares: Chowkhamba Sanskrit Series Office, 1963. Goraksa Vijay. Kavikokil Vidyāpati-krt Goraksavijay. Ed. Harimohan Misra. Patna: Kaksapuța of Siddha Nāgārjuna. In Indrajālavidyāsamgrabab. Ed. Jivananda Bhatta- Bihar-Rashtrabhasha-Parishad, 1974. charya. Calcutta: V. V. Mukherji, 1925, pp. 264-390. Gorkhavamsavali. Ed. Yogi Narharinath. Benares: Rashtriya Press, 1964. Kaksaputa mchan kbung gi sbyor ba. In Sde-dge mtshal-par bstan 'gyur: A Fac- Granthāvali of Kabīr. Kabīr Granthāvali. Ed. Shyam Sundar Das, 5th ed. Benares: simile Edition of the 18th Century Redaction of Si-tu Chos-kyi-byun-gnas (Prepared Kasi Nagaripracarini Sabha, 1954. under the Direction of H.H. the 16th rgyal-dban Karma-pa). Delhi: Karmapae Kabir Granthavali (Doba). Trans. into French by Charlotte Vaudeville. Chodhey Gyalwae Sungrab Partun Khang, 1978. Toh 1609 (P 2480). Publications de l'Institut Francais d'Indologie, no. 12. Pondicherry: Institut Kālikā Purāna. Ed. English translation and introduction by B. N. Shastri, 2 vols. Français d'Indologie, 1957. Delhi: Chowkhambha Sanskrit Series, 1972. Reprint Delhi: Nag Publishers, Gubyasamāja Tantra. Ed. Benoytosh Bhattacharyya. Gaekwad's Oriental Series, no. 1992. 53. Baroda: Oriental Institute, 1931. Kātbaka Sambitā. 3 vols. Ed. Leopold von Schroeder. Leipzig: F. A. Brockhaus, Hārītasambitā. La Hārītasambitā: texte médical sanskrit. Ed. and trans. into French 1900-12. Reprint. Wiesbaden: Steiner Verlag, 1970-72. by Alix Raison. Pondicherry: Institut Français d'Indologie, 1974. Katbasaritsagara of Somadeva. The Ocean of Story Being C. H. Tawney's translation of Harivamsa. The Harivamsa Being the Khila or Supplement to the Mababbarata, 2 vols. Somadeva's Katha Sarit Sagara (or Ocean of Streams of Story), 10 vols. Ed. N. M. Ed. Parasuram Lakshman Vaidya. Poona: Bhandarkar Oriental Research Insti- Penzer. London: Chas. J. Sawyer, Ltd., 1924-28. tute, 1969, 1971. Kaulajnānanirņaya of Matsyendranātha. Kaulajnānanirnaya and Some Minor Texts of Harsacarita of Banabhatta. The Harsacarita of Banabbatta with Exbaustive Notes the School of Matysyendranatha. Ed. Prabodh Chandra Bagchi. Calcutta Sanskrit [Ucchvāsas I-VIII]. Ed. Pandurang Vamana Kane. Bombay: 1918; reprint New Series, no. 3. Calcutta: Metropolitan, 1934. Delhi: Motilal Banarsidass, 1973. Kausītaki Upanisad. Ed. and trans. into French by Louis Renou. Paris: Adrien Mai- .The Harsacarita, trans. E. B. Cowell and P. W. Thomas. London: Univer- sonneuve, 1948. sity Press, 1897. Krtyakalpataru of Bhatța Laksmīdhara. Tīrthavivecanakāņda. Ed. K. V. Rangas- Hathayogapradīpikā of Svātmarāman. Hatbayogapradīpikā of Svātmarāman. With the wamy Aiyangar. Baroda: Oriental Institute, 1942.

528 529 Selected Bibliograpby Selected Bibliograpby

Kubjikāmata. Kubjikāmata Tantra, Kulālikāmnāya Version. Ed. Teun Goudriaan and trans. into Hindi by Satyavrat Singh and Mahaprabhulal Goswami. Sitapur: In- Jan A. Schoterman. Orientalia Rheno-Traiectina, 30. Leiden: Brill, 1988. stitute for Puranic and Vedic Studies and Research, 1984-1986. Kulānanda Tantra of Matsyendranātha. In Kaulajānanirņaya of Matysyendranātha. Matrkabbeda Tantra. Ed. Chintamani Bhattacharya, Calcutta Sanskrit Series, 7. Kularnava Tantra. Ed. Taranatha Vidyaratna with an introduction by Sir John Calcutta: Metropolitan, 1933, 1958. Woodroffe. Madras: Ganesh and Company, 1965; reprint Delhi: Motilal Banar- Matrkabheda Tantra. Trans. Michael Magee [Loknath Maharaj]. Delhi: Indological sidass, 1975. Book House, 1989. Kulavīra Tantra of Matsyendranātha. In Kaulajnānanirņaya of Matysyendranātha. Mrcchakațikā of Šūdraka. Mrcchakațikā of Śrī Šūdraka. Ed. Jagdish Chandra Mishra. Lankāvatāra Sūtra. Edited by B. Nanjio. Kyoto: Otani University Press, 1923. Benares: Chaukhamba Surbharati, 1985. Lankavatāra Sūtra. Trans. Daisetz Teitaro Suzuki. London: Routledge and Kegan Mrgendragama. Mrgendragama, Section des Rites et Section du Comportement avec la Paul, 1932. Vrtti de Bhattanārāyanakantha. Ed. Hélène Brunner-Lachaux. Publications de Lobasarvasvam of Sureśvara. The Lohasarvasvam of Śrī Sureśwara. Ed. Brahmashankar l'Institut Française d'Indologie, no. 69. Pondicherry: Institut Français d'Indolo- Mishra with Hindi commentary by Pavani Prasad Sharma. Vidyabhavana Ay- gie, 1962. urveda Granthamala, no. 46. Benares: Chowkhamba Vidyabhavan, 1965. Navanäthacaritra of Gaurana. Ed. K. Ramakrishnaiah. Madras: Madras University Mababbarata, 21 vols. Ed. Visnu S. Sukthankar, et al. Poona: Bhandarkar Oriental Press, 1970. Research Institute, 1933-60. Netra Tantra. The Netra Tantram, with the commentary of Ksemarāja, 2 vols. Ed. Mabanirvana Tantra. With the commentary of Hariharananda Bharati. Ed. Arthur Madhusudan Kaul. Srinagar: Research Department, 1926, 1939. Avalon [John Woodroffe]. Madras: Ganesh, 1929. Reprint Delhi: Motilal Padmāvat of Muhammad Jāyasī. Lakshmi Dhar. Padumavati: A Linguistic Study of Banarsidass, 1977. the 16th Century Hindi (Avadhi). London: Luzac & Co., Ltd., 1949. Mabāprajñāparamitāšāstra of Nāgārjuna. Traité de la grande vertu de sagesse de Nāgār- Rajnath Sharma, Jayasī Granthavali (Padmāvat- Țīkā Sabit). Agra: Vinod juna (Mabāprajñāparamitāšāstra), 4 vols. Ed. and trans. into French by Etienne Pustak Mandir, 1965. Lamotte. Louvain: Institut Orientaliste, 1949-1976. . The Padumawati of Malik Muhammad Jaisi. Ed. with a Hindi commentary, Mabayāna Samgraba. La Somme du Grande Vebicule d'Asanga, 2 vols. Ed. and trans. translation, and critical notes by George Grierson and Sudhakara Dvivedi. Bib- into French by Etienne Lamotte. Bibliothèque du Muséon, nos. 6-7. Louvain: liotheca Indica, n.s. 877. Calcutta: Baptist Mission Press, 1896. Institut Orientaliste, 1938-39. -. A. G. Shirreff, Padmavati of Malik Mubammad Jaisi. Bibliotheca Indica, Maitrāyaņi Sambitā. Die Sambitā der Maitrāyanīya-Sākbā. Ed. Leopold von 267. Calcutta: Royal Asiatic Society of Bengal, 1944. Schroeder. Weisbaden: Franz Steiner Verlag, 1972. Parasurāmakalpasūtra. With the Sanskrit commentary of Rāmeśvara. Ed. A. M. Sas- Maitrī Upanisad. Ed. and trans. into French by Marie-Louise Esnoul. Paris: Adrien tri and S. Y. Dave. Gaekwad's Oriental Series, no. 22. Baroda: Oriental Insti- Maisonneuve, 1952. tute, 1950. Malati-Madhava of Bhavabhūti. Bhavabhūti's Malati-Madbava, with the Commentary Paratrimsika, with the Vivarana commentary of Abhinavagupta. Ed. M. K. Sastri. of Jagaddbara, 3d ed. Ed. M. R. Kale. Delhi: Motilal Banarsidass, 1967. Kashmir Series of Texts and Studies, vol. 18. Srinagar: Research Department, Manasollāsa of Bhūlokamalla Someśvara. Manasollāsa of King Bhūlokamalla Som- 1918. eśvara, 2d ed., 3 vols. Ed. G. K. Srigondekar. Gaekwad's Oriental Series, no. 28. Parātrīsika Laghuvrtti of Abhinavagupta. Ed. Jagaddhara Zadoo. Kashmir Series of Baroda: Oriental Institute, 1967. Texts and Studies, no. 68. Srinagar: Research Department, 1947 Mañjuśrīmūlakalpa. Ed. T. Ganapati Sastri, 3 vols. Trivandrum Sanskrit Series, nos. Parātrīšikā Vivarana of Abhinavagupta. A Trident of Wisdom: Translation of Parā- 70, 76, 84. Trivandrum: Superintendent, Government Press, 1920-22. trīšika-vivarana of Abbinavagupta. Trans. Jaideva Singh, with a foreword by Paul Manu Smrti. With the commentary of Kulluka Bhatta. Ed. Gopala Sastri Nene. Muller-Ortega. Albany: SUNY Press, 1989. Kashi Sanskrit Series, no. 114. Benares: Chowkhamba Sanskrit Series Office, Paśupata-Sūtram. With the commentary of Kaundinya. Ed. R. Ananthakrishna Sas- 1970 tri. Trivandrum: Oriental Manuscript Library of the University of Travancore, Mārkandeya Purāna. Śrīvyāsapranītam Śrīmārkandeya Purānam, 3 vols. Ed. and 1940.

531 530 Selected Bibliograpby Selected Bibliograpby

Prabodbacandrodaya of Krsnamiśra. Ed. and trans. Sita Krsna Nambiar. Delhi: Mot- Rasaratnākarāntargataś caturtho [sic] Rddbikbandah-Vadikbandab, ed. Jivaram Kal- ilal Banarsidass, 1971. idas Shastri. Rasashala Granthamala, no. 9. Gondal: Rasashala Aushadhash- Pranagnibotra Upanisad. In Paul Deussen, Sixty Upanisads (listed under Secondary ram, 1940. Sources), vol. 2, pp. 645-51. Rasaratnasamuccaya of Mānikyadeva Sūri. The Rasa-ratna-samuccaya of Mānikyadeva Praśna Upanisad. Ed. and trans. into French by J. Bousquet. Paris: Adrien Maison- Süri. Ed. and trans. J. C. Sikdar. Jaipur: Prakrta Bharati Academy, 1986. neuve, 1948. Rasaratnasamucchaya of Vāgbhatta. Rasaratnasamucchaya, 2 vols. Ed. with Marathi Rāja Mārtanda of Bhojaraja. Yoga-sutras of Patanjali with Bhojavrtti Called Rajamar- translation by Sadasiv Balwant Kulkarni. Kolhapur: Sivaji University, 1970, 1972. tanda (in English Translation). Delhi: Parimal Publications, 1990. -. Śrīvāgbbațācāryaviracitah Rasaratnasamucchaya. Ed. with a Hindi commen- Rājataramginī of Kalhana. Kalbana's Rajataramginī, Chronicle of the Kings of Kasbmir, tary by Dharmananda Sharma. Benares: Motilal Banarsidass, 1962. 2 vols. Ed. M. A. Stein. London: Constable, 1900. Reprint Delhi: Motilal Ba- Rasārņava. Rasārnava. Ed. P. C. Ray and Hariscandra Kaviratna. Bibliotheca Indica, narsidass, 1979 174. Calcutta: Baptist Mission Press, 1910. Rāmāyaņa of Vālmīki, 7 vols. Ed. G. H. Bhatt et al. Baroda: Oriental Institute, -. Rasārnavam nama Rasatantram. Ed. with Hindi translation by Indradeo 1960-75. Tripathi and notes by Taradatta Panta. Haridas Sanskrit Series, 88. Banares: Rasacintāmaņi of Anantadeva Sūri. Ed. Jivaram Kalidas. Bombay: Venkatesvara Chowkhamba, Sanskrit Series Office, 1978. Steam Press, 1967. Rasārņavakalpa. Rasārņavakalpa (Manifold Powers of the Ocean of Rasa). Ed. Mira Roy Rasabrdaya Tantra of Govinda. Ed. Yadavji Trikamji Acarya. Lahore: Motilal Banar- and B. V. Subbarayappa. New Delhi: Indian National Science Academy, 1976. sidass, 1927. Rasasanketakālikā of Camuņdakāyastha. Ed. with notes by Yadavji Trikamji Acarya -. Śrimadgovindabbagavatpādaviracita Rasbrdayatantram with the commentary and Hindi commentary by Indradeo Tripathi. Chaukhamba Ayurveda Grantha- of Caturbhuja Misra. Trans. into Hindi by Daultarama Rasasastri. Vidyavilas mala, 10. Benares: Chaukhamba Amarabharati Prakashan, 1984. Ayurved Series, 5. Benares: Chaukhamba Orientalia, 1989. Rasendra Bhāskara of Lakminārāyaņa. Jaipur: n.p., 1896. Rasakāmadbenu. Ed. Yadavaji Trikamji Acarya. Bombay, Nirnaysagar Press, 1925. Rasendracintāmani of Dhundhukanātha. Ed. Manirama Sastri. Bikaner: Manirama- Reprint Benares: Chaukhambha Orientalia, 1988. sarma, 1934. Rasakaumudi of Bhişagvāra Jnānacandra Sarman. The Rasakaumudi of Bhisagvāra Rasendracūdāmaņi of Somadeva. Rasendracūdāmaņi. Ed. Yadava Sharman. Lahore: Jñänacandra Sarman. Ed. Brahmasankar Mishra. Vidyabhavan Ayurveda Gran- Motilal Banarsidass, 1932. thamala, 47. Benares: Chowkhamba Vidyabhavan, 1966. Rasendracūdāmani by Acārya Somdeva. Ed. Siddhinandan Misra. Varanasi: Rasamañjari of Salinātha. Rasamañjari-vaidyaka. Ed. Raghunathji Srikrsnalal. Bom- Chowkhambha Orientalia, 1984. bay: Venkatesvara Press, 1906. Rasopanisat. Ed. Sambhasiva Sastri. Trivandrum: Superintendent, Government Rasapaddbati of Bindu. Ed. Yadavji Trikamji. Bombay: Nirnay Sagar Press, 1925. Press, 1928. Rasaprakāśa Sudhākara of Yaśodhara. Rasaprakaśa. Edited by Yadavji Trikamji Rasopanisat, 2 vols. Ed. with a Hindi translation and commentary by Badri- Acharya. Ayurvediya Granthamala, vol. 1. Bombay: Venkatesvara Steam Press, narayan Sarma. Ajmer: Krsna Gopal Ayurved Bhavan, 1959. 1910-II. Rg Veda. Rg Veda Sambita, Togetber with the Commentary of Sāyanāchārya, 4 vols. Ed. -. Rasaprakāśasudbākara by Acarya, Yasodbara. With a Hindi translation by Sid- F. Max Müller. 2d ed. London: Henry Frowde, 1890-92; first Indian ed. Chow- dhinandan Misra. Jaikrishnadas Ayurveda Series, no. 54. Benares: Chaukham- khamba Sanskrit Series no. 99. Varanasi: Chowkhamba Sanskrit Series Office, bha Orientalia, 1984- 1966. Rasaratnākara of Nityanātha, Rasāyanakbanda. Śrīnityanāthasiddbaviracitah (Rasarat- Rudrayāmala Tantra. Rudrayāmalam. Ed. the Yogatantra Department. Yogatantra nākarāntargataś caturthab) Rasāyanakbandab, ed. Yadavji Trikamji Acharya with a Granthamala, no. 7. Benares: Sampurnanand Sanskrit Vishvavidyalaya Press, Hindi translation by Indradeo Tripathi. Haridas Sansrit Series, no. 95. Benares: 1980. Chaukhamba Sanskrit Pustakalaya, 1939, 1982. Sādbanamālā. Sādhanamālā, 2 vols. Ed. Benoytosh Bhattacharya. Gaekwad's Orien- Rasaratnākara of Nityanātha, Rddhikbanda. Sriparvatīputranityanātbasiddbaviracitab tal Series nos. 26, 41. Baroda: Oriental Institute, 1925, 1928.

532 533 Selected Bibliograpby Selected Bibliograpby

Sakbi of Kabīr. Kabir, vol. 1. Trans. Charlotte Vaudeville. Oxford: Clarendon Śiva Sambitā. Ed. and trans. Sisa Chandra Vasu, 2d ed. Allahabad: Panini Office, Press, 1974- 1914; reprint New Delhi: Munshiram Manoharlal, 1975. Śankaradigvijaya of Vidyāraņya. Śrīmacchankaradigvijaya. Ed. K. S. Venkata Rama- Skanda Purāna. Ed. Nag Sharan Singh, 3 vols. Delhi: Nag Publishers, 1984. shastri. Srirangam: Srivanivilasamudralaya, 1972. Smrticandrikā of Devannabhatta. Smrticandrikā of Devamnabhatta, Āhnika Kāņda. Sankaravijaya of Ānandagiri. Śrīsankaravijaya of Anantānandagiri. Ed. N. Veezhina- Trans. with notes by J. R. Gharpure. Bombay: V.J. Gharpure, 1946. than, with an introduction by T. M. P. Mahadevan. Madras: University of Solab Somvār tathā Somvār aur Saumya Prados Vrat Katha. Hardwar: Ranadhir Book- Madras, 1971. sales, n.d. Śārada Tilaka of Laksmanadesikendra. The Śārdatilakam of Laksmanadesikendra with Śrī Mastnāth Carit (Śrī Mastnāth adbbut līlā prakās) of Śrī Šankarnāth Yogiśvar. Ed. the Padarthadarsa Commentary by Rāghavabhatta, 3d ed. Ed. Mukunda Jha Bakshi, Sankarnath Yogisvar. Delhi: Dehati Pustak Bhandar, 1969. Kashi Sanskrit Granthamala, 107. Varanasi: Chowkhamba Sanskrit Series Śrīnāthakathāsāra of Dwarkanāth. Šrīnātbakatbāsāra. Ed. Narharinath Yogi. Office, 1986. Benares: n.p., 1951. Sarngadbara Paddhati. The Paddhati of Sarngadbara: A Sanskrit Anthology. Ed. Peter Śrinath Tirthavali of Rājā Mān Singh. Churu, Rajasthan: n.p., 1951. Peterson. Bombay Sanskrit Series, no. 37. Bombay: Government Book Depot, Sūnyasampādane of Srī Gūļūra Siddhavīraņāryaru, 5 vols. Ed. and trans. S. C. Nan- 1888. dimath, L. M. A. Menezes, R. C. Hiremath et al. Dharwar: Karnataka Univer- Śarngadhara Sambitā. Ed. and trans. K. R. Srikanta Murthy. Benares: Chau- sity Press, 1965-72. khamba, 1984. Susruta Sambitā with the commentary of Atrideva. Ed. Bhaskara Govinda Gha- Sarvadarśanasamgraba of Mādhava. Sarvadarsana-samgraha of Sāyana-Mādhava. Ed. nekar. 5th ed. Delhi: Motilal Banarsidass, 1980. with Commentary by Vasudev Shastri Abhyankar. Government Oriental Series, Su yu ki of Hsuan-tsang. Buddhist Records of the Western World, Translated from the Class A, no. 1. Poona: Bhandarkar Oriental Research Institute, 1924. Reprint Chinese of Hiuen Tsiang (A.D. 629) by Samuel Beal, 2 vols. London: Trubner & Poona: Bhandarkar Oriental Research Institute, 1978. Co., 1884; reprint Delhi: Munshiram Manoharlal, 1969. The Sarva-darsana-samgraba or Review of the Different Systems of Hindu Phi- Svacchanda Tantra. Svacchandatantram, with the commentary of Ksemarājā, 2 vols. losopby by Madhava Acharya, trans. E. B. Cowell and A. E. Gough. London: Ed. Vrajavallabha Dvivedi. Delhi: Parimal Publications, 1985. Kegan Paul, Trench & Trubner, 1882. 7th reprint ed. Benares: Chaukhamba Śvetāśvatara Upanisad. Ed. and trans. into French by Aliette Silburn. Paris: Adrien Amarbharati, 1978. Maisonneuve, 1948. Satapatha Brābmana. The Satapatba Brābmana in the Madbyandina-Sakha, with Ex- Taittiriya Sambitā. With the commentary of Madhava. Ed. Rajendralal Misra. Cal- tracts from the Commentaries of Sayana, Harisvamin, and Dvivedaganga. Ed. Al- cutta: Bibliotheca Indica, 1860. brecht Weber. London-Berlin, 1855. Reprint ed. Chowkhamba Sanskrit Series, Taittirīya Upanisad. Ed. and trans. into French by Emile Lesimple. Paris: Adrien no. 96. Benares: Chowkhamba Sanskrit Series Office, 1964. Maisonneuve, 1948. The Satapatha Brabmana According to the Madbyandina School. Trans. Julius Tanjur. sde-dge mtshal-par bstan 'gyur: A Facsimile Edition of the 18th Century Redaction Eggeling. Sacred Books of the East, vols. 12, 26, 41, 43, 44. Oxford: Clarendon of Si-tu Chos-kyi-byun-gnas (Prepared under the Direction of H.H. the 16th Press, 1882-1900; reprint Delhi: Motilal Banarsidass, 1968. rgyal-dban Karma-pa). Delhi: Delhi Karmapae Chodhey Gyalwae Sungrab Par- Satsāhasra Sambitā. The Satsābasrasambitā, Chapters 1-5. Trans. J. A. Schoterman. tun Khang, 1978. Leiden: Brill, 1982. Tantrāloka of Abhinavagupta. The Tantrāloka of Abhinavagupta with Commentary Siddba Siddbanta Paddhati of Goraksanātha. Edited with an introduction by Kalyani by Rājānaka Jayaratha, ed. with notes by Mukund Ram Shastri, 12 vols. Allah- Mallik in Siddha Siddhanta Paddhati and Other Works of the Natha Yogis. Poona: abad: Indian Press, 1918-38. Reprinted with introduction and notes by Oriental Book House, 1954. R. C. Dwivedi and Navjivan Rastogi, 8 vols. Delhi: Motilal Banarsidass, Simbasanadvatrimsika. Vikrama's Adventures or the Thirty-two Tales of the Tbrone, 2 1987. vols. Trans. Franklin Edgerton. Harvard Oriental Series, 26. Cambridge: Har- . Luce delle sacre scritture (Tantraloka). Trans. Raniero Gnoli. Torino: Unione vard University Press, 1926. Tipografico-Editrice Torinese, 1972.

534 535 Selected Bibliograpby Selected Bibliograpby

Tantrasamgraba, 4 vols. Ed. Gopinath Kaviraj (vols. 1-3) and Ramaprasada Tripathi Yogaratnamālā of Nāgārjuna. Nāgārjuna's Yogaratnamālā with the Commentary of (vol. 4). Benares: Sampurnanand Sanskrit Visvavidyalaya, 1973-81. Śvetāmbara Bhiksu Gunākara. Ed. Priyavrat Sharma. Benares: Chaukhambha Tantrasāra of Abhinavagupta. The Tantrasāra of Abbinavagupta. Ed. with notes by Orientalia, 1977. Mukunda Ram Sastri. Bombay: Nirnaya Sagar, 1918; reprint Delhi: Bani Pra- Yoga Sāstra of Dattātreya. Ed. and trans. Amita Sharma. Delhi: Swami Keshawa- kashan, 1982. nanda Yoga Institute, 1985. Tantratattva. Principles of Tantra: The Tantratattva of Śrīyukta Siva Candra Vidyārņava Yogaśataka of Nāgārjuna. Yogašataka: texte médicale attribué à Nāgārjuna. Ed. and Bhattācārya, 2 vols. Ed. John Woodroffe. Madras: Ganesh, 1970. French trans. by Jean Filliozat. Pondicherry: Institut Français d'Indologie, Târă Tantra. Tara-Tantram, ed. Girish Chandra Vedantatirtha. With an introduc- 1979. tion by A. K. Maitra. Rajshahi: Varendra Research Society, 1914; reprint Delhi: Yoga Sūtras of Patañjali. Sāmkbya Yogadarśana or Yogadarśana of Patañjali. Ed. Gos- Bani Prakashan, 1983. vami Damodara Sastri et al. Benares: Chaukhambha Sanskrit Sansthan, 1990. Todala Tantra. Todala Tantram. Ed. Bhadrashil Sharma. Allahabad: Kalyan Mandir, -. Yoga Philosopby of Patañjali with the commentary of Vyasa. Ed. and trans. 1961. Swami Hariharananda Aranya. Calcutta: University of Calcutta, 1981. Vaikbānasa Smārta Sūtra. Ed. Wilhelm Caland. Calcutta: Baptist Mission Press, Yoga Upanisads. The Yoga Upanisads with the Commentary of Śri Upanisad- 1927. Brabmayogin. Ed. A. Mahadeva Sastri. Adyar Library Series, vol. 6. Madras: Vājasaneyi Sambitā. Vājasaneyi Sambitā, zd ed. With the commentaries of Uvata Adyar Library and Research Center, 1968. English translation by Srinivasa and Mahīdhara. Ed. Vasudeva Laksman Sastri Pansikar. Bombay: Pandurang Iyengar. Madras: Adyar, 1952. French translation by Jean Varenne. Upanishads Jawaji, 1929. du Yoga. Paris: Gallimard, 1971. Varņaratnākara of Kaviśekharācārya Jyotīśvara. Ed. S. K. Chatterji and Babua Yogatattva Upanisad. In Yoga Upanişads. Misra, Bibliotheca Indica, no. 262. Calcutta: Bibliotheca Indica, 1940. Yogavişaya of Matsyendranātha. In Siddba Siddbānta Paddbati of Gorakşanātba. Vasavadattā of Subandhu. The Vasavadattā of Subandhu. Ed. Fitz-Edward Hall. Yogimuni 1000 of Yogimuni. Ed. and trans. S. Raghunathan. Madurai: Madurai Calcutta: Asiatic Society of Bengal, 1859. University, 1982. Vayutattvam-bhavanopadesa of Ghoraksa. Esoteric Instruction on the Cultivation of the Yogisampradāyāviskrti of Jñāneśvara. Trans. into Hindi by Candranāth Yogī. Wind Principle (vāyutattvam-bhavanopadesa; rlung gi de nyid bsgom pa'i man ngag) Ahmedabad: Sivanath Yogi, 1924. in sde-dge mtshal-par bstan 'gyur: A Facsimile Edition of the 18th Century Redaction Yoni Tantra. Yonitantra. Ed. with an introduction by J. A. Schoterman. Delhi: Ma- of Si-tu Chos-kyi-byun-gnas (Prepared under the Direction of H.H. the 16th nohar, 1980. rgyal-dban Karma-pa). Delhi: Delhi Karmapae Chodhey Gyalwae Sungrab Partun Khang, 1978. Toh. 2377; P 3219. Secondary Sources Vimalaprabba. Portions trans. in Fenner, Edward Todd, Rasayan Siddbi (listed under Aiyar, C. S. Narayanaswami. "Ancient Indian Chemistry and Alchemy of the Secondary Sources). Chemico-Philosophical Siddhanta System of the Indian Mystics." In Proceedings Vinasikbatantra. Ed. with an introduction and a translation by Teun Goudriaan. and Transactions of the Third All-India Oriental Conference, Madras, 22-24 Decem- Benares: Motilal Banarsidass, 1985. ber 1924. Madras: Law Printing House, 1925. Visnudbarmottara Purāņa. Bombay: Venkatesvara Press, 1912. Alexander, Ernst Bruce. Statistical Description and Historical Account of the North- Yājñavalkya Smrti, 2d ed. Ed. T. Ganapati Sastri. New Delhi: Munshiram Mano- western Provinces of India. Gazetteer, Northwest Provinces, vol. 6. Allahabad: harlal, 1982. Northwest Provinces and Oudh Government Press, 1881. Yogabija of Goraksanätha. Ed. Ramlal Srivastav. Gorakhpur: Gorakhnath Mandir, Ali, S. M. The Geograpby of the Puranas, 2d ed. New Delhi: People's Publishing 1982. House, 1973. Yogadarśana Upanişad. In Yoga Upanişads. Alper, Harvey, ed. Mantra. Albany: SUNY Press, 1989. Yogakundali Upanisad. In Yoga Upanişads. Anantacharya, E. Rasayana and Ayurveda. Vishakapatnam: World Teacher Publi- Yogamārtanda of Gorakşanātha. In Siddba Siddbānta Paddhati of Goraksanātha. cations, 1935, 1982.

536 537 Selected Bibliograpby Selected Bibliography

Anantharaman, T. R. "Transformations-Metallurgical and Mental," Prof. N. P. Bhardwaj, Surinder Mohan. Hindu Places of Pilgrimage in India. Berkeley: Univer- Gandhi Memorial Lecture, Varanasi, 18 December 1973. sity of California Press, 1973. Apfel-Marglin, Fréderique. Wives of the God-King. New York: Oxford University Bhattacarya, Jivananda, ed. Indrajālavidyāsamgraba. Calcutta: V. V. Mukherji, 1925. Press, 1985. Bhattacharya, Benoytosh. An Introduction to Buddbist Esoterism. Oxford: Oxford Aung, Maung Htin. Folk Elements in Burmese Buddbism. London: Oxford University University Press, 1932. Press. 1962. Reprint Westport, Conn .: Greenwood Press, 1978. Bhattacharyya, Haridas, ed. Cultural Heritage of India. 5 vols. Calcutta: Rama- Babb, Lawrence. The Divine Hierarchy: Popular Hinduism in Central India. New krishna Mission, 1956. York: Columbia University Press, 1975. Bhattacharyya, Nagendranath. History of the Tantric Religion (A Historical, Ritualistic Bachelard, Gaston. La poétique de l'espace. Paris: Presses Universitaires de France, and Philosophical Study). Delhi: Munshiram Manoharlal, 1982. 1974. Biardeau, Madeleine. L'Hindouisme: Anthropologie d'une civilisation. Paris: Flam- Bagchi, Prabodh Chandra. "Compte-rendu de Benoytosh Bhattacharya, édition de marion, 1981. la Sadhanamala" Indian Historical Quarterly 6 (1930): 574-87. . Histoires de poteaux: Variations autour de la Déesse bindoue. Publications de "The Cult of the Buddhist Siddhacaryas." In Cultural Heritage of India, ed. l'Ecole Française d'Extrême Orient, 154. Paris: Ecole Française d'Extrême Ori- Haridas Bhattacharyya, vol. 4, pp. 273-79. ent, 1989. -. Studies in the Tantras. Calcutta: University of Calcutta, 1939, 1975. "Puranic Cosmogony." In Asian Mythologies, compiled by Yves Bonnefoy Bali, Krsna Kumar. Țillā Goraksanāth. Haridwar: Pir Kala Nath Trust, 1983. and translated under the direction of Wendy Doniger. Chicago: University of Bandyopadhyay, Pradip Kumar. Natha Cult and Mabanad. Delhi: B. R. Publishing Chicago Press, 1989, pp. 43-50. Company, 1992. Biardeau, Madeleine, and Charles Malamoud, Le Sacrifice dans l'Inde Ancienne. Banerjea, Aksaya Kumar. Philosopby of Gorakhnath with Goraksa-Vacana-Sangraba. Paris: Presses Universitaires de France, 1976. 2d ed. Delhi: Motilal Banarsidass, 1983. Blackburn, Stuart H., Peter J. Claus, Joyce B. Fluekiger, and Susan Wadley, eds. Banerji, R. D. The Haihayas of Tripuri and Their Monuments. Memoirs of the Archae- Oral Epics in India. Berkeley: University of California Press, 1989. ological Survey of India, no. 23. Calcutta: Government of India Central Publi- Bose, D. M., ed. A Concise History of Science in India. New Delhi: Indian National cation Branch, 1931. Science Academy, 1971. Barrow, H. W. "On Aghoris and Aghorapanthis." Anthropological Society of Bombay Bouillier, Véronique. "La caste sectaire des Kānphață Jogī dans le royaume du 3:4 (1893): 197-251. Nepal: l'exemple de Gorkha." Bulletin de l'Ecole Francaise d'Extreme-Orient 75 Barthwal, Pitambaradatta. Gorakb Bāņī. Allahabad: Hindi Sahitya Sammelan, 1955 (1986): 125-67. (Ist ed. 1942) -. "Growth and Decay of a Kanphata Yogi Monastery in South-west Nepal." -. Yog Pravab. Benares: Kashi Viyapith, 1946. The Indian Economic and Social History Review 28:2 (1991): 151-70. Bayly, C. A. Rulers, Townsmen and Bazaars: North Indian Society in the Age of British "The King and His Yogi: Prthivinarayan Sah, Bhagavantanath and the Expansion, 1770-1870. Cambridge: Cambridge University Press, 1983. Unification of Nepal in the Eighteenth Century." In John P. Neelsen, ed., Bedi, Rajesh and Ramesh. Sadhus: The Holy Men of India. New Delhi: Brijbasi Print- Gender, Caste and Power in South Asia: Social Status and Mobility in a Transitional ers Private Ltd., 1991. Society, pp. I-21. Bernier, François. Voyage dans les Etats du Grand Mogol. Ed. with an introduction by . "La violence des non-violents, ou les ascètes au combat." In Denis Vidal, France Bhattacharya. Paris: Fayard, 1981. Gilles Tarabout and Eric Meyer, eds., Violences et Non-Violences en Inde, pp. Beyer, Stephan. The Cult of Tara. Berkeley: University of California Press, 213-43 1973. Boulnois, Lucette. Poudre d'or et monnaies d'argent au Tibet. Paris: Editions du Bharati, Aghehananda. The Tantric Tradition. London: Rider and Co., 1965; New C.N.R.S., 1983. York: Grove Press, 1975. Briggs, George Weston. Gorakbnäth and the Kānphata Yogis. Calcutta: Y.M.C.A. Bharati, Dharmavir. Siddh Sabity. zd ed. Allahabad: Kitab Mahal, 1955, 1968. Press, 1938; New Delhi: Motilal Banarsidass, 1982.

538 Selected Bibliography 539 Selected Bibliograpby

Brooks, Douglas Renfrew. The Secret of the Three Cities: An Introduction to Hindu Śakta Tantrism. Chicago: University of Chicago Press, 1990. Dasgupta, Shashibhushan. An Introduction to Tantric Buddhism. 2d ed. Calcutta: Buoy, Christian. Les Natha-yogin et les Upanisads. Publications de l'Institut de Civi- University of Calcutta Press, 1958.

lisation Indienne, no. 62. Paris: De Boccard, 1994. Obscure Religious Cults. 3d ed. Calcutta: Firma KLM Limited, 1976. Burgess, James. Report on the Antiquities of Kâthiawad and Kacch, Being the Result of the Dasgupta, Surendranath. A History of Indian Philosopby, 5 vols. Cambridge: Cam- Second Season's Operations of the Archaeological Society of Western India, 1874-1875. bridge University Press, 1922; reprint Delhi: Motilal Banarsidass, 1951-55. London: India Museum, 1876; reprint Delhi: Indological Book House, 1971. Dash, Bhagwan. Alchemy and Metallic Medicines in Ayurveda. New Delhi: Concept Carstairs, G. Morris. The Twice Born. London: Hogarth Press, 1958. Publishing, 1986.

Chambard, Jean-Luc. "Les trois grands dieux aux enfers: Tradition orale et cycle . Fundamentals of Ayurvedic Medicine. Delhi: Bansal & Co., 1978.

des fêtes hindoues dans un village de l'Inde centrale (M.P.). In Catherine Cham- Davis, Richard H. Ritual in an Oscillating Universe: Worshipping Siva in Medieval pion, ed. Littérature Populaire et Tradition Orale en Inde (Purusārtha 17). Paris: India. Princeton: Princeton University Press, 1991

Éditions de l'E.H.E.S.S., 1994. Detienne, Marcel. L'invention de la mythologie. Paris: Gallimard, 1981. Chénu, Marie-Dominique. Nature, Man and Society in the Twelfth Century. Ed. and Deussen, Paul. Sixty Upanisads. Trans. V. M. Bedekar and G. B. Palsule. Delhi: trans. Jerome Taylor and Lester K. Little. Chicago: University of Chicago Motilal Banarsidass, 1980.

Press, 1969; reprint Chicago: Midway Books, 1979. Devi, Konduri Sarojini. Religion in the Vijayanagara Empire. New Delhi: Sterling, Cordier, Palmyr. Catalogue du fonds tibétain de la Bibliothèque nationale, vol. 3. Index 1990.

du Bstan-hgyur (Tibétain 180-332). Paris: Bibliothèque Nationale, 1915. Dhoundiyal, B. N. Rajasthan District Gazateers: Sirobi. Jaipur: Government Central Press, 1967. "Histoire de la medecine indienne: La phtisie pulmonaire." Annales d'bygiène et de médecine coloniales 15 (1912) 255-66, 535-48. Digby, Simon. "Encounters with Jogīs in Indian Sufi Hagiography." Typescript of . Nagārjuna et l'Uttaratantra de la Suçrutasambita. Antananarivo: n.p., 1896. paper read at the School of Oriental and African Studies, London, 27 January "Recentes decouvertes de mss. medicaux sanscrits dans l'Inde (1898- 1970.

1902)." Le Muséon, n.s. 4 (1903): 321-52. . "To Ride a Tiger or a Wall? Strategies of Prestige in Indian Sufi Legend." -. Vāgbhata et l'Astangabrdayasambitā. Besançon: n.p., 1896. In According to Tradition: Hagiographical Writing in India. Ed. Winand Cael- Cort, John. "Twelve Chapters from The Guidebook to Various Pilgrimage Places, the lewaert and Rupert Snell. Wiesbaden: Harrassowitz, 1994, pp. 99-129. Vividhatirthakalpa of Jīnaprabhāsūri," in Phyllis Granoff, ed., The Clever Adul- Dikshit, Rajesh. Śrī Navanäth Caritr Sagar. Delhi: Dehati Pustak Bhandar, 1969. teress and the Hungry Monk. Dimmitt, Cornelia, and J. A. Van Buitenen. Classical Hindu Mytbology. Philadelphia: Crooke, William. "Hinglãj." In Encyclopedia of Religion and Ethics, vol. 6, pp. 715-16. Temple University Press, 1978.

New York: Scribner's, 1914. Dimock, Edward C., Jr. The Place of the Hidden Moon: Erotic Mysticism in the Vais- Religion and Folklore of Nortbern India. London: Oxford University Press, nava-sabajiya Cult of Bengal. With a new Foreword by Wendy Doniger. Chicago: 1926; reprint Delhi: Munshiram Manoharlal, 1968. University of Chicago Press, 1966; reprint Delhi: Motilal Banarsidass, 1991. -. "A Version of the Guga Legend," Indian Antiquary 24 (1895): 49-56. Dobbs, Betty Jo Teeter. The Foundations of Newton's Alchemy or "The Hunting of the Ctesias. Indika. In Photios, Bibliotheca, 6 vols. Ed. and trans. into French by René Greene Lyon." London: Cambridge University Press, 1975. Henry. Paris: Les Belles Lettres, 1959. Dowman, Keith. Masters of Mabamudra: Songs and Histories of the Eighty-Four Bud- Baba Cunninathji. Tan-Prakāś. Gorakhpur: Eureka Press, 1981. dhist Siddbas. Albany: SUNY Press, 1985. Cunningham, Alexander. Archaeological Survey of India, Four Reports Made during the Dumézil, Georges. Mythe et épopée, 3 vols: Vol. 1. L'idéologie des trois fonctions dans les Years 1862-63-64-65. Simla: Government Central Press, 1871. épopées des peuples indo-européens, 5th ed. Paris: Gallimard, 1986. Vol. 2. Types -. A Report of a Tour in the Central Provinces in 1873-74 and 1874-75. Archaeo- épiques indo-européens: un béros, un sorcier, un roi, 4th ed. Paris: Gallimard, 1986. Vol. 3. Histoires romaines, 3d ed. Paris: Gallimard, 1981. logical Survey of India, vol. 9. Calcutta: Archaeological Society, 1879; reprint Dumont, Louis. Homo bierarchicus: Le système des castes et ses implications. Paris: Galli- Benares: Indological Book House, 1966. mard, 1978.

540 54

Selected Bibliography Selected Bibliograpby

Dutt, N. "Notes on the Nagarjunikonda Inscriptions." Indian Historical Quarterly La doctrine classique de la médecine indienne: Ses origines et ses parallèles grecs. 7:3 (September 1931): 633-53. 2d ed. Paris: Ecole Française d'Extrême Orient, 1949, 1975. Dvivedi, Hazariprasad. Nath Sampraday. 3d ed. Allahabad: Lokabharati Prakasan, -. Etude de démonologie indienne: Le Kumāratantra de Rāvana et les textes paral- 1981. lèles indiens, tibétains, chinois, cambodgien et arabe. Cahiers de la Société Asiatique, -, ed. Nāth Siddhom ki Bāniyam. 2d ed. Benares: Kasi Nagaripracarini Sabha, rè série, vol. 4. Paris: Imprimerie Nationale, 1937. 1980 (first ed. 1957). -. "Liste des manuscrits de la Collection Palmyr Cordier conservés à la Bibli- Dwarkanath, C. Introduction to Kāyacikitsa. Bombay: Popular Book Depot, 1959. othèque Nationale." Journal Asiatique 224 (1934): 155-73. Dyczkowski, Mark S. G. The Canon of the Saivāgama and the Kubjika Tantras of the "Le Kumāratantra de Ravana." Journal Asiatique 226 (1935):1-66. Western Kaula Tradition. Albany: SUNY Press, 1988. -. "Les mechanismes psychiques d'après les textes de yoga." Annuaire du Col- -. "Kundalinī the Erotic Goddess: Sexual Potency, Transformation and lège de France (Paris: 1970-71): 415-16; (Paris: 1971-72): 397-99. Reversal in the Heterodox Theophanies of the Kubjika Tantras." Typescript of -. "Taoïsme et yoga." Journal Asiatique 257 (1969): 41-87. paper read at American Academy of Religion Annual Meeting, San Francisco, Filliozat, Jean, and Louis Renou. L'Inde classique: Manuel des études indiennes. 3 vols. 23 November 1992. Paris: Imprimerie Nationale, 1947-53. Eck, Diana. Banaras: City of Light. Princeton: Princeton University Press, 1983. Fishlock, Trevor. India File. London: John Murray, 1983; reprint Calcutta: Rupa Eliade, Mircea. Forgerons et alchimistes. 2d rev. ed. Paris: Flammarion, 1977. and Co., 1984. -. The Sacred and the Profane: The Nature of Religion. Trans. Willard R. Trask. Forbes, Alexander Kinloch. Ras-mala: Hindu Annals of Western India, with particular New York: Harcourt Brace Jovanovich, 1959. Reference to Gujarat. London: Richardson, 1878; reprint New Delhi: Heritage Shamanism, Archaic Techniques of Ecstasy. Trans. Willard R. Trask. Publishers, 1973. Princeton: Princeton University Press, 1972. Forster, E. M. The Hill of Devi. Harmondsworth: Penguin, 1963. Yoga: Immortality and Freedom. 2d ed. Trans. Willard R. Trask. Princeton, Gairola, C. Krishna. "Les conditions sociales et religieuses à l'époque de Sata- N.J .: Bollingen, 1973. vāhana dans l'Inde (Ie siècle av. J .- C .- IIe siècle ap. J .- C)." Journal Asiatique 243 Encyclopedia of Religion and Etbics. 12 vols. Ed. James Hastings. New York: Scrib- (1955): 281-95 ner's, 1908-26. Gaston, Anne-Marie. Siva in Dance, Myth and Iconography. Delhi: Oxford University Encyclopedia of Religions. Ed. Mircea Eliade. Chicago: University of Chicago Press, Press, 1982. 986. Gautam, Camanlal. Śrī Gorakbnath Caritr. Bareilly: Samskrti Samsthan, 1981. Enthoven, R. E. The Tribes and Castes of Bombay. 3 vols. Bombay: Government Cen- Gellner, David N. Monk, Housebolder, and Tantric Priest: Newar Buddhism and Its tral Press, 1923. Hierarcby of Ritual. New Delhi: Cambridge University Press, 1992. Erndl, Kathleen M. Victory to the Mother: The Hindu Goddess of Northwest India in Gharote, M. L., and V. A. Bedekar, eds. Descriptive Catalogue of Yoga Manuscripts. Myth, Ritual, and Symbol. New York: Oxford University Press, 1993. Lonavla: Kaivalyadhama S.M.Y.M. Samiti, 1989. Ernst, Carl. "The Arabic Version of "The Pool of the Water of Life (Amrtakunda)." Ghimire, Yubaraj. "The Rise of the Sadhus." India Today (31 January 1993): 61-63. Typescript of paper read at American Academy of Religion, Annual Meeting, Giri, Sadananda. Society and Sannyāsin: A History of the Dasanāmī Sannyāsins. Be- Washington, D.C., 20 November 1993. nares: n.p., 1976. Fenner, Edward Todd. Rasayan Siddhi: Medicine and Alchemy in the Buddbist Tantras. Gold, Ann Grodzins. A Carnival of Parting: The Tales of King Bharthari and King Gopi Ph.D. dissertation, University of Wisconsin, 1979. Chand as Sung and Told by Madbu Natisar Nath of Ghatiyali, Rajasthan. Berkeley: Filliozat, Jean. "Al-Biruni and Indian Alchemy." In Studies in the History of Science in University of California Press, 1992. India. 2 vols. Ed. Debiprasad Chattopadhyaya. New Delhi: Educational Enter- -. "Gender and Illusion in a Rajasthani Yogic Tradition." In Tale, Text and prises, 1982, vol. 1, pp. 338-43. Time: Interpreting South Asian Expressive Traditions. Ed. Frank Korom, Arjun Ap- -. "La discipline psychosomatique du yoga et ses fondament théoriques." An- padurai, and Margaret Mills. Philadelphia: University of Pennsylvania Press, nuaire du Collège de France (1965-66): 383-87. 1990, pp. 102-35.

542 543 Selected Bibliograpby Selected Bibliograpby

Gold, Daniel. "The Instability of the King: Magical Insanity and the Yogi's Power Heimann, Betty. Facets of Indian Thought. London: Allen and Unwin, 1964- in the Politics of Jodhpur, 1803-1843." In David N. Lorenzen, ed., Bhakti Reli- Helffer, Mireille and Gaborieau, Marc. "A propos d'un tambour du Kumaon et de gion in North India: Community Identity and Political Action. Albany: SUNY Press, l'ouest du Népal: Remarques sur l'utilisation des tambours-sabliers dans le 1995, pp. 120-32. monde indien, le Népal et le Tibet." In Studia instrumentorum musicae populares, Gonda, Jan. Change and Continuity in Indian Tradition. The Hague: Mouton, 1965. Festschrift to E. Emsheimer on the occasion of his 7oth birthday. Stockholm: . Medieval Religious Literature in Sanskrit. History of Indian Literature, II, I. Musikhistorika Museet, 1974. Wiesbaden: Harassowitz, 1977. Hiltebeitel, Alf, ed. Criminal Gods and Demon Devotees: Essays on the Guardians of -. Triads in the Veda. Amsterdam: North-Holland, 1976. Popular Hinduism. Albany: SUNY Press, 1989. . Visnuism and Sivaism, A Comparison. London: Athlone, 1970. Hou, Ching-lang. Monnaies d'offrande et la notion de Trésorerie dans la religion chinoise. Goswamy, B. N., and J. S. Grewal. The Mugbals and the Jogis of Jakbbar: Some Paris: Mémoires de l'Institut des Hautes Etudes Chinoises, 1975. Madad-i-Ma'ash and Other Documents. Simla: Indian Institute of Advanced Hulin, Michel. La face cachée du temps: L'imaginaire de l'au-delà. Paris: Fayard, 1985 Study, 1967. I-Ching. A Record of the Buddhist Religion as Practiced in India and the Malay Arbci- Goudriaan, Teun, ed. The Sanskrit Tradition and Tantrism. Panels of the VIIth pelago (AD 671-695). Trans. J. Takakusu. London: Clarendon Press, 1896; re- World Sanskrit Conference, Kern Institute, Leiden, 3-29 August 1987. Leiden: print Delhi: Munshiram Manoharlal, 1966. Brill, 1990. Jacobi, H. "Abode of the Blest (Hindu)." In Encyclopedia of Religion and Ethics, vol. Goudriaan, Teun, and Sanjukta Gupta. Hindu Tantric and Sakta Literature. History I, pp. 698-700. of Indian Literature, II, 2. Wiesbaden: Harrassowitz, 1981. Jaggi, O. P. Yogic and Tantric Medicine. History of Science and Technology in India, Granoff, Phyllis, ed. The Clever Adulteress and the Hungry Monk: A Treasury of Jain vol. 5. Delhi: Atma Ram and Sons, 1973. Literature. New York, Mosaic Press, 1990. Jamison, Stephanie W. The Ravenous Hyenas and the Wounded Sun: Myth and Ritual Granoff, Phyllis, and Koichi Shinohara, eds. Monks and Magicians: Religious Bio- in Ancient India. Ithaca, N.Y .: Cornell University Press, 1991. graphies in Asia. Oakville, Ontario: Mosaic Press, 1988; reprint Delhi: Motilal Joshi, Damodar. "Mercury in Indian Medicine." Studies in History of Medicine 3 Banarsidass, 1994. (1979):234-97. . "The Song of Manik Candra." Journal of the Royal Asiatic Society of Bengal Joshi, Rasik Vihari. "Notes on Guru, Dīksā and Mantra." Ethnos 37 (1972): 103-12. 47:3, pt. I (1878): 135-238. Kaelber, Walter O. Tapta-Marga: Asceticism and Initiation in Vedic India. Albany: . "Two Versions of the Song of Gopicand." Journal of the Royal Asiatic Society SUNY Press, 1989. of Bengal 55 (1885): 35-55. Kakar, Sudhir. The Inner World: A Psychoanalytic Study of Childhood and Society in Gupta, Mataprasad. Prācīn Bhāsā Natak Sangrab. Agra: Agra University Press, 1970. India. New York: Oxford University Press, 1978. Gupta, Sanjukta. "The Mandala as an Image of Man." In Richard Gombrich, ed., .Shamans, Mystics and Doctors: A Psychological Inquiry into India and Its Healing Indian Ritual and Its Exegesis. Delhi: Oxford University Press, 1988, pp. 32-41. Traditions. New York: Alfred A. Knopf, 1982; New Delhi: Oxford India Paper- Gupta, Sanjukta, Dirk Jan Hoens, and Teun Goudriaan. Hindu Tantrism. Hand- backs, 1990 buch der Orientalistik, 2.4.2. Leiden: Brill, 1979. Kalyān Sakti Ank. Gorakhpur: Gita Press, 1934. Harzer, Edeltraud. "Samkhya," In Encyclopedia of Religion, vol. 13, pp. 47-51. Kamat, Ashok Prabhakar. Mabarastr ke Natbpanthiy Kaviyom ka Hindi Kavy. Math- Hawking, Stephen W. A Brief History of Time: From the Big Bang to Black Holes. New ura: Jawahar Pustakalaya, 1976. York: Bantam Books, 1988. Kane, Pandurang Vaman. History of Dharmasāstra. 5 vols. 2d ed. Poona: Bhandarkar Hazra, R. C. Studies in the Puranic Records on Hindu Rites and Customs. 2d ed. Delhi: Oriental Research Institute, 1968-75. Motilal Banarsidass, 1975. Kantawala, S. G. Cultural History from the Matsya Purāna. Baroda: M. S. Univer- Heesterman, Jan. The Broken World of Sacrifice. Chicago: University of Chicago sity, 1964. Press, 1993. Kapani, Lakshmi. La notion de samskara. 2 vols. Publications de l'Institut de Civi- . The Inner Conflict of Tradition. Chicago: University of Chicago Press, 1985. lisation Indienne, no. 59. Paris: De Boccard, 1991, 1993.

544 545 Selected Bibliograpby Selected Bibliography

Karambelkar, V. W. "Matsyendranath and His Yogini Cult." Indian Historical Quar- Larson, Gerald James. Classical Samkbya: An Interpretation of its History and Meaning. terly 31 (1955): 362-74- 2d ed. Delhi: Motilal Banarsidass, 1979. Kardaswamy, Thiru N. History of Siddha Medicine. Madras: n.p., 1979. Lecomte-Tilouine, Marie. "Des dieux aux sommets (Népal)." In Véronique Bouil- Kaul, Jayalal. Lal Ded. New Delhi: Sahitya Akademi, 1973. lier and Gérard Toffin, eds. Classer les Dieux? Des Panthéons en Asie du Sud [Puru- Kaviraj, Gopinath. Aspects of Indian Thought. Calcutta: University of Burdwan, sārtha 15]. Paris: EHESS, 1993, pp. 153-72. 1966. Lester, Robert C. Theravada Buddbism in Southeast Asia. Ann Arbor: University of -. Bharatiy Samskrti aur sādbanā. 2 vols. Patna: Bihar Rashtrabhasha Par- Michigan Press, 1973. ishad, 1962-64 Lévi, Sylvain. "Kanişka et Satavahana: deux figures symboliques de l'Inde au pre- Tāntrik Sādhana aur Samskrti. Patna: Bihar Rashtrabhasha Parishad, 1979. mier siècle." Journal Asiatique 228 (1936):61-121. Tantrik Vanmay mem Sākta Drsti. Patna: Bihar Rashtrabhasha Parishad, . "Manimekhala, divinité de la mer." In Mémorial Sylvain Lévi. Paris: P. Hart- 1963. Khakhar, Dalpatram Pranjivan. "History of the Kanphatas of Kacch." Indian Anti- mann, 1937, pp. 371-83. . Le Népal. 3 vols. Paris: Ernest Leroux, 1905; reprint Paris: Toit du quary 7 (February 1878):47-53 Monde & Errance, 1985. Khan, Dominique-Sila. "L'Origine ismaélienne du culte hindou de Ramdeo Pīr." Lincoln, Bruce. Death, War, and Sacrifice: Studies in Ideology and Practice. Chicago: Revue de l'Histoire des Religions 210 (1993):27-47 University of Chicago Press, 1991. -. "Ramdeo Pir and the Kamadiya Panth." In N. K. Singhi and Rajendra Locke, John K. Karunamaya: The Cult of Avalokitesvara-Matsyendranath in the Valley Joshi, eds. Folk, Faitb, and Feudalism. Jaipur: Rawat Publications, 1995, pp of Nepal. Kathmandu: Sahayogi Press, 1980. 295-327. Lorenzen, David. The Kapalikas and Kalamukbas: Two Lost Saivite Sects. New Delhi: Knipe, David M. "The Heroic Theft: Myths from Rgveda IV and the Ancient Near Thomson Press, 1972. East." History of Religions 6 (1967):328-60. -. "Śaivism: Pāśupatas." In Encyclopedia of Religion, vol. 13, pp. 18-19. "Night of the Growing Dead: A Cult of Virabhadra in Coastal Andhra." -. "Warrior Ascetics in Indian History." Journal of the American Oriental Soci- In Alf Hiltebeitel, ed. Criminal Gods and Demon Devotees: Essays on the Guardians ety 98 (1978):61-75. of Popular Hinduism. Mabbett, I. W. "The Symbolism of Mount Meru." History of Religions 23 -. "One Fire, Three Fires, Five Fires: Vedic Symbols in Transformation," (1983):64-83. History of Religions 12 (1972):28-41 MacDonald, Ariane. Le mandala du Mañjusrīmūlakalpa. Paris: Adrien Maison- Kolff, Dirk H. A. Naukar, Rajput and Sepoy: The Ethnobistory of the Military Labour neuve, 1962. Market in Hindustan, 1450-1850. Cambridge: Cambridge University Press, Madhavan, V. R. Siddba Medical Manuscripts in Tamil. Madras: International Insti- 1990 tute of Tamil Studies, 1984. Kramrisch, Stella. The Hindu Temple. 2 vols. Calcutta: University of Calcutta, 1946; Madhihassan, S. Indian Alchemy or Rasāyana. New Delhi: Vikas, 1979. reprint Benares: Motilal Banarsidass, 1976. Mahapatra, Piyush K. "The Nath Cult of Bengal." Folklore (Calcutta) 12:10 (Octo- -. The Presence of Siva. Princeton: Princeton University Press, 1981. ber 1971): 376-96. Kraus, Paul. Jabir ibn Hayyan: Contribution à l'étude des idées scientifiques dans l'Islam: Majumdar, M. R. et al. Historical and Cultural Chronology of Gujarat. 2 vols. Baroda: Jabir et la science grecque. Paris: Les Belles Lettres, 1986. University of Baroda, 1960. Lamb, Alastair. British India and Tibet, 1766-1910. 2d ed. London: Routledge and Malamoud, Charles. "Cosmologie prescriptive: Observations sur le monde et le Kegan Paul, 1986. non-monde dans l'Inde ancienne." Le temps de la réflexion 10 (1989):303- 25. Lamotte, Etienne. Histoire du Bouddhisme indien, des origines à l'ère Saka. Biblio- . Cuire le monde: Mythe et pensee dans l'Inde ancienne. Paris: Editions de la thèque du Muséon, vol. 43. Louvain: Université de Louvain, 1958. Découverte, 1989. Lapoint, E. C. "The Epic of Guga: A North Indian Oral Tradition." In S. Vatuk, Mallmann, Marie-Therese de. "Divinites hindoues dans le tantrisme bouddhique." ed., American Studies in the Anthropology of India. New Delhi: Manohar, 1978. Arts Asiatiques 1O (1964):67-86.

546 547 Selected Bibliograpby Selected Bibliograpby

Mariadassou, Paramananda. Médecine traditionelle de l'inde. 4 vols. Pondicherry: Neelsen, John P., ed. Gender, Caste and Power in South Asia: Social Status and Mobility Sainte Anne, 1934-36. in a Transitional Society. Delhi: Munshiram Manoharlal, 1991. Maspero, Henri. Le Taoïsme et les religions chinoises. With a preface by Maxime Kal- Newman, John. "A Brief History of the Kalacakra." In The Wheel of Time: The Kala- tenmark. Paris: Gallimard, 1971. cakra in Context. Madison, Wis .: Deer Park Books, 1985. McDaniel, June. The Madness of the Saints: Ecstatic Religion in Bengal. Chicago: Uni- Obermiller, Ernst. History of Buddbism (Chos-bbyung) by Bu-ston. 2 vols. Heidelberg: versity of Chicago Press, 1989. Harrassowitz, 1931-32; reprint New Delhi, Sri Satguru Publications, 1986. McGregor, Ronald Stuart. Hindi Literature from Its Beginnings to the Nineteenth Cen- O'Flaherty, Wendy Doniger. Dreams, Illusion and Otber Realities. Chicago: Univer- tury. Wiesbaden: Harrassowitz, 1984. sity of Chicago Press, 1984. Meister, Michael, ed. Discourses on Siva: Proceedings of a Symposium on the Nature of The Rig Veda. Harmondsworth: Penguin, 1981. Religious Imagery. Philadelphia: University of Pennsylvania Press, 1984. Śiva: The Erotic Ascetic. New York: Oxford University Press, 1981. Mely, M. F. de. "L'alchimie chez les chinois et l'alchimie grecque." Journal Asia- -. Tales of Sex and Violence: Folklore, Sacrifice, and Danger in the Jaiminiya Brab- tique, 9è série, vol. 6 (September-October 1895): 314-40. mana. Chicago: University of Chicago Press, 1985. Meulenbeld, Jan G. History of Indian Medical Literature. Leiden: Brill, 1996. . Women, Androgynes and Other Mythical Beasts. Chicago: University of Chi- Michaels, Axel. "On 12th-13th Century Relations between Nepal and South cago Press, 1980. India." Journal of the Nepal Research Center 7 (1985):69-72. Oguibenine, Boris. "Sur le terme yoga, le verbe yuj- et quelques-uns de leurs dérivés Misra, Sarat Chandra. "On the Cult of Goraksanath in Eastern Bengal." Journal of dans les hymnes védiques." Indo-Iranian Journal 27 (1984):85-101. the Department of Letters (University of Calcutta) 14 (1927):1-41. Ojha, Gaurishankar H. "A Gold Coin of Bappa Rawal." Journal of the Asiatic Society Misra, Shiva Sheikhar. Fine Arts and Technical Sciences in Ancient India with Special of Bengal, new series, 23:6, numismatic supplement no. 40 (1926-27):14-18. Reference to Somesvara's Manasollasa. Benares: Krishnadas Academy, 1982. Oman, John Campbell. Cults, Customs, and Superstitions of India. London: T. Fisher Misra, Siddhinandan. Āyurvedīya Rasašāstra. Jaikrishnadas Ayurveda Series, 35. Unwin, 1908. Benares: Chowkhamba Orientalia, 1981. The Mystics, Ascetics and Saints of India. London: 1903; reprint Delhi: Orien- Mitra, Debala. Buddhist Monuments. Calcutta: Sahitya Samsad, 1971. tal Publishers, 1973. Monier-Williams, Monier. A Sanskrit-English Dictionary. London: Oxford Univer- Oosten, Jarich G. The War of the Gods: The Social Code in Indo-European Mythology. sity Press, 1899; reprint Delhi: Motilal Banarsidass, 1984. London: Routledge & Kegan Paul, 1985. Mookerji, Bhudeb. Rasa-jala-nidhi or Ocean of Indian Chemistry and Alchemy. 5 vols. Padoux, Andre, ed. L'Image divine: Culte et meditation dans l'bindouisme. Paris: Edi- Calcutta: K. C. Neogi, 1926-38. tions du CNRS, 1990. Mukherji, Tarapada. Gopicandra Nātaka. Calcutta: University of Calcutta Press, -, ed. Mantras et diagrammes rituels dans l'hindouisme. Table Ronde, Paris, 1970. 21-22 June 1984. Paris: Editions du CNRS, 1986. Muller-Ortega, Paul Eduardo. The Triadic Heart of Siva: Kaula Tantricism of Abbina- - "The Body in Tantric Ritual: The Case of the Mudras." In Teun Goud- vagupta in the Non-dual Shaivism of Kasbmir. Albany: SUNY Press, 1989. riaan, ed., The Sanskrit Tradition and Tantrism, pp. 65-77. Murty, K. Satchidananda. Nagārjuna. New Delhi: National Book Trust, 1971. Vac: The Concept of the Word in Selected Hindu Tantras. Albany: SUNY Mus, Paul. Barabudur. Hanoi: Imprimerie d'Extrême Orient, 1935. Press, 1989. -. "La notion de temps réversible dans la mythologie bouddhique." Ecole Pra- Pal, Pratapaditya. Hindu Religion and Iconology According to the Tantrasara. Los tique des Hautes Etudes. Annuaire 1938-1939. Melun: Imprimerie Administra- Angeles: Vichitra Press, 1981. tive, 1938 Pandey, Divakar. Gorakhnātb evam unkī Paramparā kā Sābity. Gorakhpur: Gorakh- Yogi Narharinath. Nav Nath Caurāsi Siddb. Benares: Rashtriya Press, 1968. pur University, 1980. Needham, Joseph, et al. Science and Civilisation in Ancient China. 6 vols. in 17 tomes. Pandey, Raj Bali. Hindu Samskaras (Socio-Religious Study of the Hindu Sacraments). 2d Cambridge: Cambridge University Press, 1954-88. ed. Delhi: Motilal Banarsidass, 1969.

548 549 Selected Bibliography Selected Bibliograpby

Parry, Jonathan. "Sacrificial Death and the Necrophagous Ascetic." In Maurice Renou, Louis. "Études védiques," Journal Asiatique 243 (1955):405-38. Bloch and Jonathan Parry, eds. Death and the Regeneration of Life. Cambridge: Rhys-Davids, Caroline A. F. "Original Buddhism and Amrta." Mélanges chinois et Cambridge University Press, 1982. bouddhiques [Brussels], vol. 6 (1938-39), pp. 371-82. Philostratus. Life of Apollonius. Trans. C. P. Jones and edited, abridged, and intro- Robinson, James B. Buddha's Lions: The Lives of the Eighty-four Siddhas. Berkeley: duced by G. W. Bowersock. Harmondsworth: Penguin, 1970. Dharma Publishing, 1979 Prakash, Satya. Prācīn Bhārat mem Rasāyan kā Vikās. Allahabad: Prayag Visvavidya- Rose, H. A. A Glossary of the Tribes and Castes of the Punjab and Northrwest Frontier laya, 1960. Province, 3 vols. Lahore: Superintendent, Government Printing, Punjab, Pranke, Patrick. "On Becoming a Buddhist Wizard." In Don Lopez, ed., Buddbism 1911-19; reprint Delhi: Asian Educational Services, 1990. in Practice. Princeton: Princeton University Press, 1995, pp. 343-58. Roșu, Arion. "A la recherche d'un tirtha énigmatique du Deccan médiéval." Bulletin Pir Premnath, Siva-Goraksa. New Delhi: Vijnana Prakashan, 1982. de l'Ecole Francaise d'Extrême Orient 60 (1969):23-57 Przyluski, Jean. "Les Vidyārāja," Bulletin de l'Ecole Française d'Extrême Orient 23 -. "Alchemy and Sacred Geography in the Medieval Deccan." Journal of the (1923): 301-18. European Ayurvedic Society 2 (1992):151-56. Puhvel, Jaan. Comparative Mythology. Baltimore: Johns Hopkins University Press, -. "Considerations sur une technique du Rasayana Ayurvedique." Indo-Iranian 1986. Journal 17 (1975):1-29. Punjab States Gazetteer, vol 22A. Chamba State with Maps, 1904. Lahore: Civil and . Gustave Lietard et Palmyr Cordier: Travaux sur l'Histoire de la médecine in- Military Gazette Press, 1910. dienne. Publications de l'Institut de Civilisation Indienne, no. 56. Paris: De Boc- Rama Rao, B. and Reddy, M. V. "A Note on Goraksanätha and His Work Yogadi- card, 1989. pika." Bulletin of the Indian Institute of the History of Medicine (Hyderabad) 12 "Mantra et Yantra dans la medecine et l'alchimie indiennes." Journal Asia- (1982):34-38. tique 274:3-4 (1986):205-68. Ramachandra Rao, S. K. Śrī-Cakra. Delhi: Sri Satguru Publications, 1989. "Le renouveau contemporain de l'Ayurveda." Wiener Zeitschrift für die Raman Shastri, V. V. "The Doctrinal Culture and Tradition of the Siddhas." In Kunde Sidasiens 26 (1982):59-82. Cultural Heritage of India, ed. Haridas Bhattacharyya, vol. 4, pp. 300-8. "Yoga et alchimie." Zeitschrift der Deutschen Morgenländischen Gesellschaft Ramanujan, A. K. The Inner Landscape: Love Poems from a Classical Tamil Anthology. 132 (1982):363-79. Bloomington: Indiana University Press, 1975. Roux, Jean-Paul. "Turkish and Mongolian Shamanism." In Asian Mythologies, com- "The Myths of Bhakti: Images of Siva in Saiva Poetry. In Michael Meister, piled by Yves Bonnefoy and trans. under the direction of Wendy Doniger. Chi- ed., Discourses on Siva: Proceedings of a Symposium on the Nature of Religious Im- cago: University of Chicago Press, 1989, pp. 329-30. agery. Ruben, Walter. Eisenschmiede und Dämonen in Indien. Internationales Archiv Für -. Speaking of Siva. Harmondsworth: Penguin, 1979. Ethnographie, 37, suppl. Leiden: Brill, 1939. Rao, T. A. Gopinath. Elements of Hindu Iconography. 2 vols. Madras: Law Road Ruegg, David Seyfort. "Sur les rapports entre le Bouddhisme et le 'substrat reli- Printing House, 1914-16. gieux' indien et tibétain." Journal Asiatique 252 (1964):77-95. Rastogi, Navjivan. Introduction to the Tantrāloka: A Study in Structure. Albany: Sachau, Edward, ed. Alberuni's India. 2 vols. London: Kegan Paul, Trench, and SUNY Press, 1987. Trubner, Inc., 1910; reprint Delhi: Munshiram Manoharlal, 1983. -. The Krama Tantrism of Kasbmir. Vol. 1. Delhi: Motilal Banarsidass, 1979. Sahagal, Sarasvati. Gorakb Darsan. Hyderabad: Madhu Printers, 1979. Ray, Prafulla Chandra. A History of Hindu Chemistry from the Earliest Times to the Saletore, Bhasker Anand. Ancient Karnataka. 2 vols. Poona: Oriental Books Middle of the 16th Century A.D. 2 vols. Calcutta: Prithwis Chandra Ray, 1902, Agency, 1936. 1909. "The Kanaphata Jogis in Southern History." Poona Orientalist 1 (January Ray, Priyanaranjan. History of Chemistry in Ancient and Medieval India Incorporating 1937):16-22. the History of Hindu Chemistry by Acharya Prafulla Chandra Ray. Calcutta: Indian Social and Political Life in the Vijayanagara Empire (A.D. 1346-A.D. 1646). 2 Chemical Society, 1956. vols. Madras: B. G. Paul & Co., 1934.

550 551 Selected Bibliograpby Selected Bibliograpby

Samksipt Hindi Sabd Sagar. Edited by Ramcandra Varma. Benares: Nagaripracarini Sharma, Priyavrat. Ayurved ka Vaijnānik Itibās. Jaikrishnadas Ayurveda Series, no. Sabha, 1981. 1, 2d ed. Benares: Chowkhamba Orientalia, 1982. Sanderson, Alexis. "Mandala and Agamic Identity in the Trika of Kashmir." In An- Indian Medicine in the Classical Age. Chowkhamba Sanskrit Series, 85. dré Padoux, ed., Mantras et diagrammes rituels dans l'hindouisme, pp. 169-207. Benares: Chowhkamba Sanskrit Series Office, 1972. -. "Meaning in Tantric Ritual." In Ann-Marie Blondeau, ed., Le Rituel. vol. Sharpe, Elizabeth. An Eight-Hundred Year Old Book of Indian Medicine and Formulas, 3. Paris: Ecole Pratique des Hautes Etudes, 5è section, 1996, pp. 15-95. Translated from the Original Very Old Hindi into Gujarati Character and Thence into -. "Purity and power among the Brahmins of Kashmir." In The category of the Englisb. London: Luzac & Co, 1937; reprint New Delhi: Asian Educational Ser- Person: Anthropology, Philosophy, History. Ed. Michael Carrithers, Steven Collins, vices, 1979. and Steven Lukes. Cambridge: Cambridge University Press, 1985, pp. 191- Shastri, Haraprasad. A Catalogue of Palm-Leaf and Selected Paper Manuscripts Belong- 216 ing to the Durbar Library, Nepal. 2 vols. Calcutta: n.p., 1905, 1915. "Saivism and the Tantric Tradition." In The World's Religions. Ed. S. Suther- Shorter Oxford Economic Atlas of the World. 2d ed. Prepared by the Economist Intelli- land et al. London: Routledge and Kegan Paul, 1988, pp. 660-704. gence Unit. London: Oxford University Press, 1959. -. "Saivism: Krama Saivism"; "Saivism: Śaivism in Kashmir"; "Saivism: Trika Siegel, Lee. Net of Magic: Wonders and Deceptions in India. Chicago: University of Śaivism." In Encyclopedia of Religion, vol. 13, pp. 14-17. Chicago Press, 1991. -. "The Visualization of the Deities of the Trika." In André Padoux, ed., L'i- Silburn, Lilian. Kundalini: Energy of the Depths. Trans. Jacques Gontier. Albany: mage divine: Culte et méditation dans l'hindouisme. Paris: CNRS, 1990. SUNY Press, 1988. Sankrtyayana, Rahula. "(Recherches Bouddhiques II) L'Origine du Vajrayāna et "Le vide, le rien, l'abîme." In Le Vide: Expérience spirituelle. Paris: Hermes, les quatre-vingt siddhas." Journal Asiatique 225:2 (October-December 1934): 1969. 209-30. Singh, Moti. Nirgun Sabity: Samskrtik Prsthabbūmi. Benares: Nagaripracarini Sastri, Devadatt. Āgneya tīrth Hinglaj. Bombay: Lokaloka Prakashan, 1978. Sabha, 1963. Schmidt, Toni. "Fünfundachtzig Mahasiddhas." Ethnos 2-3 (1955):103-21. Singh, Raya Bahadur Munshi Haradayal. Riport Mardumasumārī Rāja Mārvād Bābat Schwartzberg, Joseph E., ed. A Historical Atlas of South Asia. Chicago: University san[vat] 1891 Isvī, Tīsarābissā. 2 vols. Jodhpur: Vidyasal, 1895. of Chicago Press, 1978; second impression, with additional material, New York: Sinha, S., and B. Saraswati. Ascetics of Kashi: An Anthropological Exploration. Varanasi: Oxford University Press, 1992. N. K. Bose Memorial Foundation, 1978. Sen, Gananath. Åyurved Paricay. Shantiniketan: Visva Vidya Samgraha, 1943. Sircar, D. C. Cosmograpby and Geograpby in Early Indian Literature. Calcutta: Indian Sen, Sukumar. History of Bengali Literature. New Delhi: Sahitya Academy, 1960. Studies: Past and Present, 1967. Sethi, V. K., ed. Kabir: The Weaver of God's Name. Beas, Punjab: Radha Soami Sat- The Gubilas of Kiskindba. Calcutta: Sanskrit College, 1965. sang Press, 1984. Tbe Sākta Pitbas. 2d rev. ed. Delhi: Motilal Banarsidass, 1973. Sethu Raghunathan, N. "Contribution of Yoogi Munivar for Siddha System of Studies in the Geography of Ancient and Medieval India. 2d rev. ed. Delhi: Medicine." In Heritage of the Tamils: Siddba Medicine. Ed. S. V. Subrahmanian Motilal Banarsidass, 1977, 1990. and V. R. Madhavan. Sivalal, Sridhar, ed. Gopīcand kā Akbyān. Bombay: Jnan Sagar, 1890. Shanmuga Velan, A. Siddhar's Science of Longevity and Kalpa Medicine of India. Ma- Smith, Brian K. Reflections on Resemblance, Ritual, and Religion. New York: Oxford dras: Sakthi Nilayam, 1963. University Press, 1989. Shah, Shersingh. Śrī Kedārnāth Mābātmya (Yātrā Gaīd). Kedarnath: Shersingh Snellgrove, David. Buddhist Himalaya: Travels and Studies in Quest of the Origins and Shah, n.d. Nature of Tibetan Religion. Oxford: Cassirer, 1957. Sharma, Brijendra Nath. Iconograpby of Sadasiva. New Delhi: Abhinava Publica- Indo-Tibetan Buddbism: Indian Buddbists and Their Tibetan Successors. 2 vols. tions, 1976. Boston: Shambhala, 1987. Sharma, Padmaja. Mabaraja Man Singh of Jodhpur and His Times (1803-1843 A.D.). Spiro, Melford. Buddhism and Society: A Great Tradition and Its Burmese Vicissitudes. Agra: Shiva Lal Agarwala & Company, 1972. New York: Harper & Row, 1970.

552 553 Selected Bibliograpby Selected Bibliograpby

Srivastav, Ramlal, ed. "Gorakb" Visesānk [Yog Vanī special issue no. I for 1977]. Gor- Upadhyaya, A. N. "On Some Undercurrents of the Nätha-sampradāya or the Car- akhpur: Gorakhnath Mandir, 1977. pata-śataka" Journal of the Oriental Institute (Baroda) 48:3 (1968-69): 198-206. -, ed. "Gorakb Bāni" Višesānk [Yog Vanī, special issue no. I for 1979]. Gorak- Vaidyaraj, Vansidhar Sukul. Vāmamārg. Allahabad: Kalyan Mandir, 1951. hpur: Gorakhnäth Mandir, 1979. Vaidyaraja, Yogiraja. "Yoga Research on Mercury and the Dhatus." Cakra: A Journal , ed. "Nath Siddh Carit" Visesānk [Yog Vanī, special issue no. I for 1984]. of Tantra and Yoga [New Delhi] 4 (March 1972):186. Gorakhpur: Gorakhnath Mandir, 1984. Vaisya, Syamasundaracarya. Rasāyan Sār. 2 vols. 5th ed. Benares: Syamasundar Ra- Stein, R. A. Grottes-Matrices et Lieux Saints de la déesse en Asie orientale. Paris: Ecole sayansala, 1971. Francaise d'Extrême Orient, 1988. Vallée Poussin, Louis de la. "Le Bouddhisme et le Yoga de Patañjali." Mélanges Stein, Rolf. "Architecture et pensée en Extrême-Orient." Arts Asiatiques 4:3 chinois et bouddhiques [Brussels] 5 (1936-37): 223-42. (1957): 163-86. Van Buitenen, J. A. "The Indian Hero as Vidyadhara." In Traditional India: Structure "Jardins en miniature d'Extrême Orient." Bulletin de l'Ecole Française d'Ex- and Change. Ed. Milton Singer. Philadelphia, American Folklore Society, 1959, trême Orient 52 (1942):1-104. pp. 99-105. Stern, Henri. "Le temple d'Eklingji et le royaume de Mewar (Rajasthan) (rapport Van der Veer, Peter. Gods on Earth: The Management of Religious Experience and Iden- au divin, royauté et territoire: sources d'une maîtrise)." In Jean-Claude Galey, tity in a North Indian Pilgrimage Centre, London School of Economics Mono- ed., L'Espace du Temple (Purusartha 10) (1986), pp. 15-30. graphs on Social Anthropology, 59. London: Athlone Press, 1988. Subrahmanian, S. V., and V. R. Madhavan, eds. Heritage of the Tamils: Siddba Medi- Van Kooij, Karel R. Worship of the Goddess According to the Kalika Purana, part 1: A cine. Madras: International Institute of Tamil Studies, 1983. Translation with an Introduction and Notes of Chapters 54-69. Leiden: Brill, Sukul, Kubernath. Varanasi Down the Ages. Benares: Bhargava Bhushan Press, 1974. 1972. Sutherland, Gail Hinich. "Bija (Seed) and Ksetra (Field): Male Surrogacy or Niyoga Varenne, Jean. Cosmogonies védiques. Milan: Archè, 1982. in the Mahabharata." Contributions to Indian Sociology, n.s. 24 (1990): 77-103. . Upanisbads du yoga. Paris: Gallimard, 1971. Svoboda, Robert. Aghora: At the Left Hand of God. Albuquerque: Brotherhood of Vaudeville, Charlotte. Kabir; vol. 1. Oxford: Clarendon Press, 1974. Life, Inc., 1986. Venkata Ramanan, K. Nagarjuna's Philosopby: As Presented in the Maba- Swynnerton, Charles. Romantic Tales from the Panjab. Westminster: Archibald Con- Prajñapāramitā-Sāstra. Delhi: Motilal Banarsidass, 1975. stable & Co., 1903. Vidal, Denis, Gilles Tarabout, and Eric Meyer, eds. Violences et non-violences en Inde Temple, Richard Carnac. The Legends of the Panjab. 3 vols. London: Turner and (Purusārtha 16). Paris: Editions de l'EHESS, 1994. Company, 1884-86; reprint Patiala: Department of Languages, Punjab, 1963. Vostrikov, A. I. Tibetan Historical Literature. Trans. Harish Chandra Gupta. Cal- Thapar, Romila. "The Image of the Barbarian in Early India." Comparative Studies cutta: R. D. Press, 1970. in Society and History 13:3 (July 1971):408-36. Waley, Arthur. "Notes on Chinese Alchemy," Bulletin of the School of Oriental Studies Tod, James. Annals and Antiquities of Rajasthan. Ed. with an introduction by William (1930-32):22-24. Crooke. 3 vols. London: H. Milford, 1920; reprint Delhi: Low Price Publica- Walker, Benjamin. Hindu World: An Encyclopedic Survey of Hinduism, 2 vols. New tions, 1990. York: Praeger Books, 1968. Treloar, Francis E. "The Use of Mercury in Metal Ritual Objects as a Symbol of Walleser, Max. "The Life of Nagarjuna from Tibetan and Chinese Sources." In Śiva." Artibus Asiae 34 (1972):232-40. Bruno Schindler, ed., Asia Major: Hirth Anniversary Volume. London: 1923. Tucci, Giuseppe. "Animadversiones Indicae." Journal of the Royal Asiatic Society of Walter, Michael Lee. "Preliminary Results from a Study of Two Rasayana Systems Bengal n.s. 26 (1930):125-60. in Indo-Tibetan Esoterism." In Tibetan Studies in Honor of Hugh Richardson, ed. -. Tibetan Painted Scrolls. 3 vols. Rome: Libreria dello Stato, 1949. Michael Aris and Aung San Sun Kyi. Warminster, Eng .: Aris and Phillips, 1979. Tulpole, Shankar Gopal. Mysticism in Medieval India. Wiesbaden: Harrassowitz, . The Role of Alchemy and Medicine in Indo-Tibetan Tantrism: Ph.D. disserta- 1984. tion, Indiana University, 1980.

554 Selected Bibliograpby

Watt, George. Dictionary of the Economic Products of India. 9 vols. Calcutta: Superin- INDEX tendent of Government Printing, 1889-96. Second reprint in 10 vols. Delhi: Periodical Exports, 1972. Watters, Thomas. On Yuan-Chwang's Travels in India, 629-645 A.D. 2 vols. London: Royal Asiatic Society, 1904-05. Wayman, Alex. The Buddbist Tantras: Light on Indo-Tibetan Esotericism. Delhi: Moti- lal Banarsidass, 1973. -. Yoga of the Gubyasamajatantra: The Arcane Lore of Forty Verses. New York: Samuel Weiser, 1977. Weinberger-Thomas, Catherine. Cendres d'immortalité, la crémation des veuves en Inde. Paris: Editions du Seuil, 1996. White, David Gordon. "Dakkbina and Agnicayana: An Extended Application of Paul Mus's Typology." History of Religions 26:2 (November 1986): 188-213. Abhinavagupta, 4, 7, 73, 88-89, 97, 212, 214, 246, 259, 269, 271, 283,

. Myths of the Dog-Man. Chicago: University of Chicago Press, 1991. 135-38, 142, 172, 178, 213, 222, 285-86, 441, 447, 448, 463, 493,

-. "Why Gurus Are Heavy." Numen 33 (1984), pp. 40-73. 224, 228, 255, 312, 329, 330, 385, 499, 512. See also fire; rasa-agni-

Wink, Andre. Al-Hind: The Making of the Indo-Islamic World. Vol. 1: Early Medieval 389, 416, 423, 461, 464, 465, 470, vāyu

India and the Expansion of Islam, Seventh to Eleventh Centuries. Delhi: 496, 508. See also Parātrimšikāvi- agnicayana (piling of the fire altar), 11, varaņa; Tantrāloka Oxford University Press, 1990. 32, 269

Woodroffe, John. The Serpent Power. 9th ed. Madras: Ganesh, 1973. Abu (Mount), 118-20, 158, 331, 349- ājñā (the "control" cakra), 40, 226, 284,

Tantrarāja Tantra: A Short Analysis. 3d ed. Madras: Ganesh, 1971. 50, 389, 410, 412-13, 434 291, 293, 469, 480; Benares identi- fied as, 231 Tantra of the Great Liberation. London: Luzac & Co., 1913; reprint New Acharya, A. C., 337

York: Dover Publications, 1972. Acintanāth, 96, 397, 500. See also Mats- Akulavīra Tantra of Matsyendra, 141,

Wujastyk, Dominik. "An Alchemical Ghost: The Rasaratnākara by Nāgārjuna." yendra 158, 513

Ambix 31:2 (July 1984):70-83. Adams, Douglas, 318 Alberuni, 49, 53-54, 76, 105, 116, 125,

-. "Ravigupta and Vagbhata." Bulletin of the School of Oriental and African Stud- adhopātana yantra. See pātana 149, 165, 199, 299, 300, 311, 372, adhoretas, 233, 240, 473. See also ürd- 429 ies 48:1 (1985):74-78. dhvaretas Yazdani, G., ed. The Early History of the Deccan. 2 vols. London: Oxford University alchemical allegory, 294-300, 505. See adhovaktra. See mouth also Padmävat; Püran Bhagat Press, 1960-61. Zbavitel, Dušan. Bengali Literature. History of Indian Literature, vol. 9, fasc. 3. adbyātman-adbiyajña-adbidevatā. See mi- alchemical canon, 144-70

Leiden: Brill, 1976. crocosm-macrocosm-mesocosm Alchemical Man, 149-51, 245, 301,

Zimmerman, Francis. Le discours des remèdes au pays des épices: Enquête sur la médecine Ādi[-nāth], 2, 71, 73, 82, 84, 88, 91-93, 320; in western traditions, 295 alchemists, Indian. See Rasa Siddhas bindoue. Paris: Payot, 1989. 95-96, 98, 124, 133, 140-41, 169,

-, "Rtu-Sātmya: Le cycle des saisons et le principe d'appropriation." Puru- 230, 295, 390, 392, 415, 421, 460, alchemy, historical etymology of the 473 term, 204, 376 sārtha 2 (1975):87-105. (English translation: "Rtu-Satmya: The Seasonal Cycle Advaita Vedanta. See Sańkara Alexander the Great (Iskandar), 238 and the Principle of Appropriateness." In Social Science and Medicine 14B Afghanistan, 63, 66, 98, 121, 204, 398, Alha Khand, 238, 476 [198o]:94-106.) Zvelebil, Kamil V. The Poets of the Powers. London: Rider and Company, 1973. 449 Allama Prabhu, 83, 102-3, III, 129,

Zysk, Kenneth. Religious Healing in the Veda. Transactions of the American Philo- Aghoris, 201, 313, 341, 399, 412, 454 237-38, 375, 406-7, 475

sophical Society, vol. 75, part 7. Philadelphia: The Society, 1985. āgneya. See year, solar alternation of generations. See guru Agni, 11, 13, 16, 189, 190, 191, 203, amalgamation. See gold, amalgamation

555

556 557

Index Index

Amanaska Yoga of Gorakhnāth, 141, apotheosis. See Siddhas (Perfecti), ac- asuddhādbvan. See māyā Babb, Lawrence, 364 249, 314, 316, 319, 481 ceding to the world of Asuras (antigods), 10, 58-59, 190, 192, Babylonian mythology, 212, 462 Amarakośa, 57 Ardhanārīśvara, 252 278, 285, 323, 324, 374, 426, 500, Bachelard, Gaston, x Amaraugha Prabodha of Gorakhnāth, arivarga (class of "enemy" metals), 154 512 badavā (doomsday mare) and bādava 141, 225, 228, 264 Arjun Nāga/Nāga Arjan, 92-93, 121, Aśvaghośa, 35 (doomsday fire), 193, 223, 231-34, Amaraughaśasana of Gorakhnāth, 141, 132. See also Nāgārjuna Aśvins, 359 414, 472. See also Vādava 231, 254, 484 arsenic, 189, 195, 198. See also red, ar- Ataka. See caves Bagchi, Prabodh Chandra, 467, 470 amāvāsyā. See month, lunar senic Atharva Veda, 12-14, 23, 359, 373, 447 Bahl, Kali Charan, xi Ambā, 118 Artba Sāstra of Kauțilya, 370, 380, 382, Atma Bodh. See Gorakb Bodh Bāil Bhādãi, 92-93 amnāyas (tantric Transmissions), 79- 446 atman, 18-20, 33-34, 36-38, 46, 102, Bālātripurasundarī, name of the God- 80, 181, 210, 388, 423. See also in- Aryadeva, 69, 81. See also Karņari-pā 155, 185, 207, 211, 359, 360, 459. dess in the Śrīvidyā Kaula, 79, dividual names of āmnāyas Aryal, Mukunda Raj, 438 See also brábman 178, 181, 392, 443 Amrt Kund. See waters, magical āsanas (yogic postures), 220, 265, 274, Aughars, 100, 399 Baluchistan. See Pakistan amrta. See nectar 294, 422. See also matsyendrāsana; Aurangzeb, I, 52, 106, 405 bābya-druti (external flux), 268, 492 Amrtakunda, hathayogic text attributed viparītakaraņi aurifaction. See transmutation Bāņa[bhațta], 49, 52-53, 60, III, 118, to Goraksa, 106, 108 Asańga, 43 ausadbi. See herbs 321, 330, 377, 382, 409. See also anagogy, 354 ashes: and purification, 444, 445; ash- avadbūta, 426, 497; Avadhūta Gītā, Kādambarī; Harsacarita anābata: bandba, 497, 502; cakra, 293, fruits, 286; curative powers of, at 141, 426; Avadhūtī, in Buddhist bandhana (alchemical and hathayogic 452; nāda, 107, 293 Bogar shrine, 376; golas of, 287; subtle physiology, 252, 454; form binding, fixation) and bandbas Ānandakanda, 80, 146, 154, 160, 167- identified with nectar, 285, 349; in of yoga taught by Dattātreya, 141; (hathayogic locks), 135, 164, 166, 69, 191, 192, 325, 391, 405, 416, Näth Siddha legend, 287-88, 503; sampradāya, 99, 395 197, 199, 202, 220, 249, 262, 266, 418, 437; authorship of, 126; list initiation with, 289, 299, 499; re- Avalokiteśvara, 70, 94, 224, 396, 415; 273, 275, 276-81, 290, 292, 293, of Rasa Siddhas in, 85-88, 90, duction of mercury to (see mer- Siddhas in the entourage of, 512 299, 441, 470, 490, 491, 495, 496, 100 cury, bhasmasūtaka; mrtasūtaka); re- Avedyanāth, xii, 346-47, 398, 493, 519 498; jālandbara bandba, 249, 277, Ananta (name of a Serpent), 213-215, duction of metals to, 244, 267, Avicenna, 189 319, 496; mabābandba, 279; mūla 374, 463 268, 297, 441; remains of pralaya, Ayas Dev Nath. See Deonath bandha, 221, 277; sabīja bandba, Anantharaman, T. R., 338 213-25, 232 (see also Seșa); ritual Ayodhya, 346-47, 398 294; uddiyāna bandha, 277. See also Andhra Pradesh, 60-61, 77, 103, 112, use of, 180; "theology" of, 284- Āyurveda, 13, 19-29, 37, 52-53, 55, stabilization and stability; 121, 161, 286, 356, 375, 382, 406, 85, 499; worn by Saiva sectarians, 87, 117, 160, 167, 184-88, 206-7, swooned, killed, and bound 408. See also Srisailam; Telugu lit- 9,97,305,500 241, 245, 258, 270, 271, 287, 324, mercury erature Asi River. See Benares 339, 340, 343, 346, 347, 356, 377, bānīs of Gorakh, xi, 121, 202, 235, 242, añjana (black antimony; invisibility Assam, 62-63, 115, 119, 138-39, 195, 411, 427, 430, 462; definition of 250, 278, 281, 293, 321, 400, 477, salve), 49, 60, 164-65, 166, 305, 235, 278, 379, 473. See also Kāma- human being, 20. See also bhūta- 488, 502; language of, 96, 132; 329, 330, 331, 436, 514 rūpa vidyā, kuțī-praveša; rasa šāstra; rasā- names of specific poem cycles, anthropic principle, 426 astabandba, 194 yana, Ayurvedic; rogavāda; ṛtu- 140. See also Gorakh Bodh Anup Sanskrit Library, Bikaner, 127, Astānga Hrdaya of Vāgbhatta the sātmya; vaidya; vājīkaraņa banka nāla (the "curved duct"), 261, 421, 431, 435 Younger, 390 485, 489 aphrodisiacs, 145, 153, 160, 166, 370, Astänga Samgraba of Vägbhatta the El- Bābā Kinnarām, founder of the Aghori Bappa Rawal, king of Mewar, 120, 122, 430 der, 52, 64-65, 125, 149, 371, 390 sect, 412 311, 313, 413, 508

558 559 Index Index

Barthwal, Pitambaradatta, 132, 399 Mahābhairava, 126, 517; Pañcāli bhūtaśuddbi, 150, 180, 183, 270-72, Bose, Jagadis, 338 Basava[nna], 46, 393; Bāsava Purāņa, Bhairava, 442; Rasānkuśa Bhair- 319; Bhūtaśuddbi Tantra, 493 Bouillier, Véronique, 508, 518 314 ava, 440; Vajrabhairava, 465; Ya- Bhūtiprakarana, 44, 128, 130, 140, 145, Brahmā, 46, 87, 124, 141, 189-90, 228, Bațukanātha, 430; Bațukanātha Bhair- māntaka Bhairava, 65. See also 155-58, 162, 163, 167, 176, 197, 294, 295, 297, 315, 316, 317, 318, ava, 152; Bațukeśvara, 430 Manthänabhairava; Rasabhairava; 198, 214, 274, 321, 372, 416, 426, 326, 392, 415, 495, 501 Bāuls, 109, 228-29, 403, 471 Svacchandabhairava 430, 432, 433, 437, 440, 474, 512. brábman, 18-20, 33-35, 38, 43, 102, bcud len. See extraction of the essence Bhairavācārya: alchemical author, See also Goraksa Sambitā; 147, 185, 208, 212, 357, 359, 360, bCud len gyi man ngag bshad pa of Bo 126, 128, 158, 430; Saiva ascetic Śrīkantha 470, 477, 482, 484. See also ātman dong, 71, 385, 449 in the Harsacarita, 307, 310, 321, Bihar, 61, 64, 147, 364, 391, 398. See Brāhmaņas: Gopatha, 23; Kausītaki, Bdud-rtsi-bam-po-brgyad-pa, of Vima- 330 also Nalanda 503; Satapatha, I1, 32, 362 lamitra, 453 Bhairavānanda Yogi, 127, 148, 417 bījas: metallic, 6, 197, 369, 493; sabīja brabmadvara. See doors Benares, 225-34 passim, 245, 305, 308, Bhairavī, 71, 127, 142, 145, 152-53, and nirbija initiation, 271; seed Brahmaguphā. See Kedārnāth 336, 346, 347, 398, 412, 435, 469, 167, 430. See also Candabhairavī; mantras, 177, 180, 271, 443, 454; brabmānda. See cosmic egg 471, 487, 518; mystic etymology Rasāṅkuśī subtle matrix of sound, 291, 368; brabmarandbra (cleft of brábman; fonta- of name, 226 Bhāluki, 158, 390 rasa-bīja (mercurial seed), 271; yo- nelle), 112, 135, 178, 196, 230, Benares Hindu University, 336-38, Bhambhulnāth, xii, 349 gin's seed, 251 241, 247, 255, 477, 485 514 Bhandarināth, 349-51 binding. See bandhana Brabmayāmala Tantra, 79, 136, 145, Bengal, 2, 60, 65, 89, 98, 107-9, 162, Bharati, Dharmavir, 455 bindu: as nasalization, 43; bindu- 404, 419 237, 260, 324, 345, 393, 399, Bharatīya Janata Party (Indian People's sādhana, 139, 425; bindu-vedha, breast milk, 21, 256, 340, 452 402-5, 434, 435, 470, 472, 476, Party), 347 312; in subtle body, 178, 185, 202, breath, 22, 45, 46-47, 174, 211, 212, 502. See also Bauls; eastern India Bhartrhari, 92-94, 98, 109, 117, 132, 312, 368, 455, 483; mabābindu, 215, 227, 243, 278, 279, 281, 284, Bengali literature, 92-94, 223, 244, 333, 396, 412, 497; called Vic- 185, 202, 455; tejobindu, 316. See 290, 294, 301, 302, 313, 318, 319, 295, 333, 395, 403, 408, 426, 454, ārnāth, 288, 500 also phonemes 320, 369, 461, 468, 495 (see also 466, 475, 476, 502, 508. See also Bhāskara, 82, 106, 112, 114, 125-26, Binduknäth, name of a son of Matsyen- prāna); as basis for units of time, Caryāpadas; Goraksa Vijay 158, 408, 415 dra, 237 511 (see also mātra; pala); control, Bernier, François, 9, 50-52, 56, IO1, bbedābbeda (identity in difference), 222, Bindunätha, name of a Rasa Siddha, 83 45-47, 135, 220, 226, 250, 253, 170, 349 241, 252 bipolarity, 17, 23, 28, 38, 42, 46-47, 254, 264, 265, 274, 281, 290, 294, bhagat, etymology of, 298. See also bbedana. See piercing 143, 241, 251-52, 360 319, 369, 384, 501 (see also Work Püran Bhagat Bhīls, 120, 413, 451 Birendra, king of Nepal, 339 in Two Parts) Bhagavantnāth, 343, 344, 347, 348 bboga, 220, 234, 263, 465; bhogavatī a black holes, 320, 511 Brhaspati, 285, 323, 324, 374 Bhairava[s], 9, 44, 54, 71, 89, 127-28, name for kundalini, 219, 465 bodhana, awakening of mercury, 267, Brhat Sambitā of Varāhamihira, 65, 135, 137, 142, 145, 148, 152, 157, Bhoja[deva]: King of Mālava, 49, 54, 290; Bodhinĩ, 291; in yogic prac- 370, 375, 378 160, 167, 169, 171, 180, 228, 284, 373; Bhojarājīyamu of Ananta 369, tice, 501 Briggs, George Weston, 396, 399, 414, 286, 316, 397, 414, 416, 429, 440, 429. See also Rājamārtanda Bodhgaya. See Bihar 418, 464 443, 444, 476; Ānandabhairava, Bhutan, 104-5 bodbisattvas, 35; Mañjuśrī, 60, 89; Nā- Broach (Bhrgukaccha). See Gujarat 142, 417; Bațukanātha Bhairava, Bhutas, demonic Beings, 374, 494; gārjuna, 67-68. See also Avalo- Brooks, Douglas, xi, 142, 364, 389 152; called leader of Siddha Bhūta Tantras, 419, 493, 494; Bhũ- kiteśvara; Buddhist divinities Buddha, 109, 268; serpent couch of, Kaula, 133; in Siddha lists, 83-84, tanātha, 98, 305; bbūtavidyā (de- Bogar, 87, 373, 376 59; teachings, 20, 32-33, 35, 375, 86, 91; Kālabhairava-nātha, 98; monology), 58, 165, 373, 374 Book of Machabees, 205 422

560 561 Index Index

Buddhism and Hinduism. See syncre- Cakravarman, king of Kashmir, 308, Carpați, 2, 71, 91, 98, 106, 109, 121, Chou 1 Tshan Thung Chbi of Wei Po- tism, Hindu-Buddhist 309, 313, 322, 507 125, 143, 369; "fathered" by Gor- Yang, 372 Buddhism, Hinayana. See Hinayana calcination. See jārana; mercury: bhas- akh, 289, 296, 501; in Siddha lists, Cikitsāsamgraba of Cakrapānidatta, 76, Buddhism masūtaka, mrtasūtaka 81-84, 86, 92-93; works attrib- 388 Buddhism, Mahāyāna. See Mahāyāna Cālukyas of Kālyāņī, 128, 131, 393, uted to, 126, 132, 160, 415, 417, Cikitsāsārasamgraba of Vangasena, Buddhism 408, 409, 413 434 388 Buddhism, Siddha traditions of. See Cambodia. See Southeast Asia Caryāpadas, 81, 89, 107, 117, 224, 252, ciñcini (tamarind): name of Kubjikā, Mahāsiddhas Camuņda, 152, 163, 412 Buddhism, tantric. See Vajrayāna Bud- Candabhairavī, 181; Caņdaghaņțā, 394, 404, 422, 507 88, 194, 450; Ciñcinimatasārasau- Caturasitisiddha Pravrtti of Abhaya- mucchaya, 88, 231, 356, 386, 388; dhism 181; Caņdakāpālinī, 181, 444 datta. See Grub thob Ciñcinīśvara, 86-88 Buddhism, Theravāda. See Hinayana Candakausika of Kşemīśvara, 11I, Caturbhujamiśra, 83 cinnabar, 62, 64-66, 117, 192, 193, Buddhism 305-7, 309, 325, 329, 513 Catuspīthatantra, 470 195-96, 198, 206, 247-48, 377, Buddhism, Tibetan. See Vajrayāna Bud- Cāndālī: alternate name for kundalinī, Caurangi[nāth], 94, 106-7, 109, 132- 380, 387, 446, 449, 450, 480, 501; dhism 367; name of a goddess, 152; 33, 239, 294, 296, 297, 298, 351, in Chinese alchemy, 194, 250, Buddhist alchemy, 71-73, 78, 103, 105, name of a plant, 154, 199; out- 391, 403, 404, 413, 460, 503-4; in 482. See also darada 110-11, 382, 384 caste woman used as tantric con- Siddha lists, 80-81, 84, 86, 91-93. clan nectar, 79, 101, 137-38, 172, 195, Buddhist divinities, 221, 247; fidibud- sort, 453, 454, 508. See also Avad- See also "Prāņ Sankalī"; Pūraņ 199, 200, 435, 473. See also yoni- dha, 71, 365, 394; Aksobhya, 35, hūtī; Ņombī Bhagat tattva 70; Amitabhā, 35, 109; Amogha- Caņdī, Candikā, 49, 142, 163, 280, caves, 115, 195, 196, 240, 294, 310, clans, Siddha and tantric, 78, 88. See siddhi, 35; as divinizations of tatt- 357,438 333, 397, 412, 451, 480, 500, 515; also kula vas, 35; Cakra Samvara, 196; Rat- Candra. See King Moon Ațaka, 246-47; Ca-ri, 216, 247, Clémentin-Ojha, Catherine, 450 nasambhava, 35; the Buddha's Candradvīpa and Candragiri, 73, 89, 452 code language, 173 dharmakāya, 35; Vairocana, 35; 113, 134, 224, 225, 229-33, 236, Central Asia, exchanges with India, 2, Congress Party, 348 Vajrapāņi, 516; Vajravarāhī, 196, 240, 253, 255, 386, 422, 467, 468, 62, 66, 251, 298, 377-80, 385, cooking, 20, 186, 207, 221, 232, 283, 247. See also bodbisattvas; Prajñā- 470, 471 471 339, 446, 499 paramitā; Vajrasattva Candrakūpa, 196, 205, 246, 457 Chambard, Jean-Luc, 362 Cordier, Palmyr, 363, 380, 387, 401 Buffalo Demon, 190 Candrapītha, 73, 230 charismatic impersonation, 387 Corpus Hermeticum, 447 Burma. See Southeast Asia Candrapurī, 73, 95-96, 230, 330, 422 China, exchanges with India, 2, 53-55, corresponding hierarchies, 139, 184- Candrasena, 82, 125 60-65, 68, 71, 194, 203-4, 250, 217 passim, 240 cakras, 5, 36, 40, 163, 170, 190, 207, Candrāvalocana of Matsyendra, 42 1 376-79, 381-82, 385, 409, 482, Cort, John, 479 208, 230-31, 240, 246, 247, 260, cannabis, 119; called korakkar-muli, 501. See also Hsuan-tsang cosmic egg, 189, 190, 215, 216, 218, 291-93, 303, 308, 318-21, 389, 412 Chinese alchemy, 53-54, 56, 63-64, 295, 326, 450, 463, 510. See also 423, 452, 468, 489, 490, 511; of Caraka Sambitā, 13, 20, 52, 76, 117, 72, 189, 194, 244, 250, 327, 328, Lokāloka; Meru Ataka, 246; earliest Indian refer- 186, 215, 327, 356, 373, 459, 490; 333, 363, 372, 376, 378, 384, 455, cranial vault, 28, 39, 41, 45, 150, 170, ences to six, 73, 111, 134; four dating of, 361, 370 470, 476, 479, 482. See also Immor- 2O1, 202, 211, 216, 218, 230, 231, Buddhist, 422 (see also pīthas, Bud- carana (coursing, chewing of metals by tals (China); Ko Hung dhist); nine, 90, 367, 423, 485, mercury), 268, 492 Chinese literature, 68-69 233, 240-43, 245-49, 252-55,

495, 496. See also ājñā; anābata; Chos bbyung of Bu-ston, 70, 231, 381, 277, 293, 301, 308, 312, 319, 320, Cāranas (atmospheric cohorts of Sid- 333, 341, 367, 469, 471, 477, 481, mūlādbāra; sabasrāra dhas and Vidyādharas), 514 383, 471 506. See also ājñā; brabmarandhra;

562 563

Index Index

cranial vault (continued) 295; rites, 162, 194, 283, 287, 444, body, 72, 102, 202, 271, 281, 303, Dūtī (tantric consort), 168, 177, 256, kbecarī mudrā; lower abdomen; 461, 496, 497. See also kāla- 315, 316, 407, 456; bound mer- 441. See also Avadhūtī; Cāņdālī; sabasrāra; trikuta vañcana; Yama cury, 149-50, 429; Diamond- Ņombī; Yoginīs

cremation ground: favorite haunt of Deccan, 2, 60, 89, 97, 103, 106, III, limbed Beauty pill, 430; Diamond Dvivedi, Hazari Prasad, 121, 132, 391, tāntrikas, 286, 305-8, 329-30; in 374, 393, 408 Maiden, 442 475 subtle body, 40, 282 debavāda (elixir alchemy), 52-54, 71- diaphragmatic retention, 135, 225, Dyczkowski, Mark, 255, 394, 429 cycles: food, 30, 32, 367; menstrual, 72, 130, 145-46, 155, 161, 170, 228, 231, 233, 248, 273-75, 301,

37, 197, 229, 366, 454, 473, 500; 187, 198, 204, 267, 342, 371, 319, 422, 470, 495. See also matsy- ear-boring, 120, 318-22, 487, 512. See temporal, 17-18, 23-32, 41-42, 384 odara also kuņdalas; mudrās; Nāth Sid- 45-47, 215-16, 317-18, 367. See Deonāth, 344-48, 518 Digby, Simon, 387, 405 dhas, called Kānphatas also kalpa; mabāyuga; month, lu- Deoraj, Rawal king of Jaisalmer, 122, digestion, 138; in fiyurvedic theory, eastern India, 107-110, 134, 163, 385, nar; transmigration; year, solar 309 21-22, 186, 207, 339-40; yogic 403, 405, 435. See also Bengal, Detha, Inder Dan, 350 power of, 286. See also jāraņa Bihar, Uttar Pradesh

daksiņāyana. See year, solar Detienne, Marcel, 123 Digvijaynāth, 346 Eastern Transmission. See Trika Kaula Dâmara Tantra, 428 Devagiri, Yādava capital, 104-6, 112, dīksā. See initiation Egg of Brahman. See cosmic egg damaru, 250-51, 482-83; damaru yan- 114, 393 Dinnāga, 35 Eighteen Purusas, 180-81 tra, 248, 250, 482; made of human devayāna. See year, solar dīpana (kindling of mercury), 267, 290, Eighty-four Siddhas, 69-70, 74, 80- skulls, 251 Devī, 142, 145, 148, 171, 357 292 81, 85, 123, 231, 243, 385, 387, darada (name for cinnabar), 247; Dar- Dhamma. See Buddha, teachings dismemberment, 295-300 391. See also Grub thob

ada-deśa, 66, 168, 192, 202, 247, Dhank, 114, 116, 165, 409-10 dissolution, cosmic. See pralaya Ekalinga Mābātmya, 120, 508 Dhanvantari, 87, 192, 504 Dobbs, Betty, xi Eliade, Mircea, xiii, 104, 295, 402 449, 457 Dārika, 231-35, 472; in Siddha lists, Dharamnāth, 96, 116, 295, 348, 410 dogs, 422 elixir alchemy. See debavāda

81,84 Dhātukalpa, 130 Dombi, 308-309, 313, 322, 367, 453, elixirs, 49, 58, 62, 67-68, 71, 114-15, Dāriyanāth, 468 Dhātukriyā or Dhātumañjarī, 417, 419 454, 507. See also Avadhūtī; Cān- 119, 122, 145, 153, 155, 166, 186, darsanas, earrings worn by Nāth Sid- dbätus: bodily constituents in fiy- dālī, Dūtī; bamsa; kuņdalinī 188, 194, 198, 213, 222, 238, 239, dhas (see kundalas; mudrās) urveda, 21-22, 25, 35, 37, 186, Doniger, Wendy, 447 241, 259, 311, 381, 385, 409, 427, Dasgupta, Shashibhushan, 295 207, 213, 316, 361, 363, 394, 446, doors (in subtle body), 252-58; brab- 454; sexual fluids as, 137, 453; Dasnāmis, 99, 254, 399, 487. See also 458, 462, 477, 499 (see also rasa madvāra, 219, 252, 276, 464, 501; three types of, 327 Nāgas [vital fluid] as chyle; ojas); term for paścima dvāra, 255. See also nine emanation, 17, 206, 219, 241, 263, Datta Patala, 42 1 metal, 145, 189, 207, 210, 212, doors; tenth door 291, 368, 441, 445, 511. See also Dattagoraksasamvāda, 425 458, 503 (see also metals) dosas ("morbid states," three humors), srsti; withdrawal and return, yogic Dattātreya, 2, 92, 99, 113, 132-33, dhātuvāda (transmutational alchemy), 21-23, 27, 29, 170, 283, 362. See embryo, 201

141, 335, 395, 396, 412; as incarna- 51-55, 71, 130, 145-46, 153, 155, also mercury, dosas of eon, cosmic. See kalpa tion of Visnu, 92, 411; at Abu and 161, 187, 204, 305, 393, 409, 446; double mountains. See mountains Ephthalite Huns, 504 Girnar, 117-18, 332-33, 515 Dhātuvādašāstra, 383 Dowman, Keith, 467 Ernst, Carl, 402

Dattātreya Tantra, 131, 133, 155, 159, Dhinodar, 96, 410, 482 Dragpa jets'en, 371 erotico-mystical practice, 6, 54, 71-72, 419-20 dhūnī, 288-89, 412, 500 dravya. See power substances 135-39, 152-54, 172-73, 176, death, 10, 23-25, 28, 32, 42, 187, 222, dbyana. See meditation Dumézil, Georges, 286, 512 198, 199, 201, 207-11, 215, 216, 232, 233, 254, 274, 279-82, 293, diamond, 177, 190, 268, 269, 431, 517; Durgā, 181, 190, 357 223, 228, 240, 253, 256, 303, 330,

564 565 Index Index

erotico-mystical practice (continued) of Sadāśiva, 35, 255, 440, 457; nine heavenly bodies, 159. See also 199, 207, 208, 210, 213-14, 216, 335, 405, 424, 456, 460. See also fires doctrine, 367; kbandhas, 33; diamond 244, 258-59, 268, 284, 292, 297, bindu-sādhana; vajrolī mudrā nectars (see extraction); sets of five geometric progressions, 294, 315-19, 298, 300, 301, 309, 314, 315, 316, ether, 150, 189, 210-12, 241-43, 272, tattvas, 34. See also extraction, of 369, 502 319, 320, 322, 337, 357, 368, 369, 293, 317, 319, 320, 444, 459, 461, pañcāmrta; pañcakarmāņi; pañca- geothermal phenomena, 121, 193, 196, 410, 413, 427, 431, 446, 447, 448, 477; as substrate of sound (śabda), makāra 234, 377, 409, 449, 450, 479. See 450-53, 458, 461, 490, 503, 505, 241. See also śūnya; vyoma flight: power of, 73, 119, 132, 145, also Gaurī Kuņd; Jwālamukhī; wa- 506; amalgamation with mercury, extraction: of pancamrta in Buddhist ters, magical; volcanos and Hindu tantra, 73, 150-51, 147, 151, 169, 187, 211, 257, 274, 370, 438, 451, 504 (see also 279, 317, 326, 330, 333, 349, 431, Germano, David, 384 vedha); equated with immortality, 438, 460, 472; of the essence (bcud 461; mercury possessed of, 211- Gheranda Sambitā, 444 6, 13, 189, 269, 492; golden body, len) in Tibetan alchemy, 71, 73, 12, 461. See also khecari Ghodacoli, 83, 86, 129, 132, 397, 399, 102, 271, 281, 303, 315, 456; mak- 201, 384, 386 funerary ritual. See death, rites 418 ing (see dbātuvāda); man with a Girnar, 117-19, 196, 245, 260, 331- golden finger, 300-301; origin Fa-hsien, 68, 381, 382 Gadamer, Hans Georg, 353 333, 396, 411, 412, 479, 515. See myths of, 12, 189-90; smithy of femininity, in tantric theory and prac- Gadanigraha of Sodhala, 434 also Gomanta; Raivata Gorakh, 301 tice, 139, 199-200, 214, 218, 222, Gahanināth, 92-93, 113, 157, 160, Goa. See Konkaņa Gold, Ann Grodzins, 476 235, 240, 252, 341, 456, 465. See 162, 396, 399, 403 Godavari River, 112-14, 134, 393, 408 Gold, Daniel, 388 also śakti Gambhīrnāth, 398 Goddess: clan nectar streams from Gomanta, 331-33. See also Girnar; Filliozat, Jean, 20, 104 ganas (Siva's host), 164, 166, 285; Gaņa- womb of, 1O1, 138, 146; cult of, 4, Raivata fire, 11, 25, 38, 182, 282; agnisāra, 495; nātha (see Gaņeśa) 49 (see also Sākta sects and sectari- Gopīcand, 92-93, 109, 117-18, 132, brabmāgni, 282; crematory, 498; di- Gandavyūba, 60 anism; worship, tantric); in Vedic 238, 244, 261, 275, 280, 283, 294- gestive (dhātvagni), 186; of desire Gandbakakalpa, 130, 419 sources, 364; linga-like images of, 96, 297, 299, 300, 301, 369, 374, (kāmāgni), 284, 463; of time (kā- Gandharvas, 10, 153, 331, 462 500; menses of in August, 473; re- 397, 411, 412, 471, 476, 499, 502, lāgni), 40, 221, 232-33, 282, 339, Gandhi, Mohandas, 348 lationship to Matsyendra/Mīna, 503, 505, 516, 517 362, 463, 472; of yoga (yogāgni), Ganeśa, 157, 159, 325, 397, 417, 513 223, 225; relationship to Siva, 5, Gorakh, 2, 9, 57, 74, 94, 106, 113, 119, 39, 220, 232, 233, 240, 282, 284, Ganges River (Gangā), 111, 113, 189, 35, 54, 128, 143, 145, 158, 173, 128, 134, 143, 156-57, 160, 162, 289, 445, 463, 499, 505; three 190, 191, 225-34 passim, 240, 176, 211-12, 220, 224, 235, 247, 195, 219, 223, 228, 231, 235-39, types of sacrificial, 362. See also 246, 277, 468, 470, 480, 490 251, 253, 256, 367, 421, 457, 462, 242-45, 249, 251, 255, 261, 264, Agni; Kālāgnirudra; sacrifice, garbhadruti (internal flux of mercury in 464, 466, 493 (see also Siva, rela- 275, 278, 280, 286, 287, 289, 295, Vedic metals), 268, 492 tionship to the Goddess); sexual 296, 299, 300, 301, 302, 310, 313, fish, 91, 229-31, 466, 471; Benares has Garuda, 190, 213, 462; Garuda Tan- emissions become arsenic, mica, 319, 320, 321, 343, 347, 391, 397, shape of, 225. See also Kaivartta; tras, 419 sulfur, 187, 189, 192-93, 198, 210, 400, 404, 406, 414, 429, 451, 453, Luī-pă; Matsyendra; matsyodara; gas vents. See geothermal phenomena 212, 234, 246, 258, 351; universe 456, 460, 465, 468, 475, 476, 488, Mīn gatis. See mercury, gatis of as subtle body of, 255; womb- 497, 500, 503, 504, 506, 512, 517; five, 16, 32-36; breaths, 22, 38, 489; el- Gaurī, 146, 197, 256, 296, 345, 440; caves of (see caves). See also kun- as a Siddha demigod, 107-9, 335, ements, 18-20, 32, 34-35, 150, Gaurī Kund, 196, 245 dalinī; śakti; and names of individual 339; as a Vidyādhara, 99, 333; aț 170, 176, 180, 185, 190, 206-10, Geber, 447 goddesses Abu and Girnar, 117, 332, 333, 240, 272, 293, 317, 319, 320, 444, gemstones, 145, 148, 190, 194, 209, gold, 6, 9, 51, 68, 137, 146, 150, 177, 412, 515; birth of, 288; founder of 445, 458, 459, 497, 510, 517; faces 289, 394, 409, 417; identified with 183, 188, 190-91, 194, 195, 198, Nāth sampradāya, 95-100; Gorakh

566 567 Index Index

Gorakh (continued) Goras Ganes Gusti, 460 gutikās (mercurial pills, globules), 147, Himachal Pradesh, 121, 126, 193, 234. Țilla/Țileti, 106, 118, 299, 346, Goudriaan, Teun, 432 153, 164, 166, 194, 315, 320, 370, See also Jwālamukhī 412, 500; in Siddha lists, 80-81, Govinda, 125-26, 158; in Siddha lists, 450, 500; gutikā bandba, 211, 258, Himalayas, 63, 112, 143, 157, 163, 84-85, 87, 91-93; in Tibetan tradi- 82, 84, 86, 91. See also Rasabrdaya 277, 287, 300, 322, 352, 487, 496; 191, 193, 225, 238, 303, 333, 335, tions, 107-8, 403; in Western Tantra siddhi-gutikā, 262, 487, 496; 339, 351, 405, 475, 480 Transmission sources, 386; liter- grace. See prasāda vyāgbrī-gutikā, 452 Hinayāna Buddhism, 270, 355, 375, ary works in Sanskrit, 131, 139- granthis (knots obstructing medial 494 41, 418 (see also individual titles of channel), 279, 496, 501 Hādi-pā, 92, 241, 280, 295-99. See also Hinglāj Devī, 66, 121-22, 196, 205, Gorakb's works); metaphysical sys- grāsa (swallowing of metals, minerals Jālandhara[-nāth] 235-46, 255, 260, 311, 414, 451, tem of, 511; on the Nepali one ru- by mercury), 6, 256, 268, 292, hailstorms, controlled by Näth Sid- 457, 464, 479, 480, 488; Nāth Sid- pee coin 311; smithy of, 301-2; 452, 492 dhas, 121 dhas custodians of, 205, 451, 464. south Indian traditions of, 87, Greer, Patricia, 482 Hall, Fitz-Edward, 418 See also nummelite 102-3, III, 131, 396, 401, 407, gross body, 18-19, 207 hamsa, 211, 218, 242, 278, 308, 461, bingula. See cinnabar 412; (spurious) guru-disciple rela- Grub thob, 74, 107-9, 126, 224, 383, 496; bamsa-ga (mercury lost hot springs. See geothermal phe- tionship with Matsyendra, 85, 386, 390, 397, 403, 467, 503; list through evaporation), 209, 444, nomena 139, 467; theories concerning of Mahāsiddhas in, 80-81, 84-85, 461; Hamsī (Goosie), 308, 322; Hsuan-tsang, 62, 67, 69, 75-76, 118, Buddhist origins of, 107-9; vernac- 87, 470 Hamseśvarī, 260. See also mantras 165, 331, 377, 515 ular works, 140, 242, 260 (see also Grünwedel, Albert, 67 Hanuman, 475 Hudūd al- 'Alam, 380 bānis of Gorakh); worshipped as Gũgā Chauhan (Gūgă Pīr; Guru haratāla. See orpiment deity, 94, 107-9, 335, 339, 350 Gūgā), 59, 243, 289, 333, 374, Hardwar, 98, 345, 349-50, 435, 512 Ibn Battûta, 370 Gorakh bānīs. See bānis of Gorakh 465, 500 Hariścandra, king of Ayodhya, 304-7, I-Ching, 62, 387 Gorakb Bodb, 485, 484, 488 Gubyasamāja Tantra, 385, 422 325 idā. See nādīs Gorakh Upanisad, 91, 140 Gujarat, 62-65, 69, 96-98, 104, 106, Hārīta Rāśi, 120-22, 31I, 413 immobilization. See stabilization and Gorakhnāth. See Gorakh 109-I10, 114-18, 158, 159, 165, Harivamśa, 332-33, 515 stability Gorakhpur, 109, 333, 420; Nāth mon- 225, 245, 331-32, 336, 337, 410- Hariyana, 121 immortality, 7, 24, 38, 45, 52-55, 106, astery at, 98, 100, 196, 346, 347, 13, 434, 505, 515; Gujarat Harsacarita of Bāņabhatta, 67, 310, 119, 138, 143, 145, 167, 173, 174, 398, 500 Ayurved University, 391, 437. See 325, 515 184, 188, 192, 195, 200, 202, 222, Goraksa Paddbati, 141, 319, 321 also Dhank; Girnar; Pattan Hatbayogapradīpikā of Svātmarāman, 239, 241, 243, 252-55, 257, 264, Goraksa Sambitā, 126, 139-41, 156-57, gunas, 21, 27, 190, 210, 283, 445, 450, 80, 140, 170, 173, 202, 220, 245, 274, 279, 281, 282, 283, 285, 286, 312, 321, 426, 432, 433. See also 461 273-77, 281, 293, 369, 391, 418, 292, 293, 294, 296, 301, 308, 319, Bbūtiprakaraņa Gupta, Sanjukta, 337 421, 423; list of Mahāsiddhas in, 321, 324, 327, 330, 315, 319, 431, Goraksa Siddbānta Samgraba, 98, 141, gupta rog, 26, 339-43 83-86, 129, 160, 470 439, 453, 481; in the Vedas, 10- 394,418 guru, 143, 145, 162, 257, 283, 285, Heesterman, Jan, 269, 348 II, 13. See also death; debavāda Goraksa Vijay of Sheikh Fayzullā, 106, 286, 306, 312, 313, 321, 322, 324, herbs, 58, 145, 153, 154, 184, 194, Immortals (China), 63, 295, 327-28, 108, 236, 295, 395, 460, 466, 502 332, 446, 487; "alternation of gen- 199, 238, 257, 259, 266, 267, 277, 334 Goraksanātha. See Gorakh erations" with disciple, 269, 286, 300, 350, 351, 371, 420, 429, 433, Indika of Ctesias, 205, 246 Gorakşasabasranāmastotra, 422, 472 466, 493, 512; Four Gurus, 181, 443, 446, 453, 454, 461, 505; Indo-European mythology, 205, 285, Goraksaśataka, 85, 141, 169, 228, 250, 324, 352, 394. See also param- sixty-four divine, 158, 160, 168, 457, 462, 472, 497, 512 253, 254, 324, 467 parās; yuganāthas 429 Indra, 46, 189, 259, 278-79, 417, 441,

568 569 Index Index

Indra (continued) jarana (digestion of metals by mercury; Jwālamukhī, 234, 414; jvālamukhī-bīda, 161, 164; name of a work in the 462, 466, 499; tin produced from calcination of mercury), 149, 268, 449; jvālamukhī-rasa, 436 Tanjur, 383 seed of, 190 282, 292, 293, 299, 300, 492. See jyotirlingas. See linga Kāla, God of Death. See Yama Indus River, 468, 491. See also Sind also kbecari; mercury, mrtabbasma/ Kalacakra Tantra, 71, 105, 148, 380, initiation, 27, 37, 143, 145, 162, 169, mrtasūtaka 384, 402, 438, 454, 488, 508 173, 175, 179, 182, 239, 245, 257, Jarāsandha, mythic king of Kathiawar, Kabīr, 85, 233, 238, 242, 243, 404, Kalacuris of Tripura, 147-48, 428. See 270, 271, 273, 280, 281, 283, 284, 332, 515 456, 503 also Madana 285, 286, 289, 303, 304, 306, 322, Jātakamālā of Aryaśūra, 58 Kacch. See Gujarat kālāgni. See fire 323, 344, 363, 391, 437, 444, 453, Jayadrathayāmala Tantra, 136, 150-51, Kadalī (Plantain): City, 113; Forest, Kālāgnirudra, 40, 215, 220, 232-33, 455, 465, 481, 484, 486, 489, 490, 404, 422 113, 162, 223, 234, 238-39, 240, 240, 367, 472; Siva Kālaghna, 440 493, 500, 508, 509, 513; and spit- Jhelum, 62, 106, 239, 287, 412, 500, 261, 466, 474, 489, 502, 504 (see Kālamukhas, 97-98, 102, 110; -rāśi ting, 120, 210-13, 321, 343, 413, 504 also kajjali); Kingdom, 236-37, name endings of, 413 (see also Pāśu- 508; by penetration (see vedha, ved- jboli (yogin's wallet), 289, 299, 351, 296, 475. See also Kingdom of patas) ba[mayī dīksā]); role of sublimated 520 Women. kalās (lunar digits), 18, 25, 36-43, 170, sexual fluids in, 138, 309, 313; jīvanmukti (bodily liberation, immortal- Kadalīmanjunātha Māhātmya, 113 367-68, 475, 477; amrta kalā, sabīja and nirbīja forms, 271 ity), 2, 5, 102, 143, 145, 147, 155, Kādambarī of Banabhatta, 53, 62 37-38; nivrtti kalā, 39 internalization of the sacrifice, 13, 18, 162, 173-74, 185, 303, 427, 514. Kadri, 86, 94-97, 109-10, 396, 404, kāla-vañcana (skewing Time), 42, 145, 184, 281; prānāgnibotra, 281, 498 See also immortality; moksa; parā- 406, 474 154, 167, 222, 255, 280, 475, 497 Īsāna, 178, 259, 495 mukti Kailash, Mount, 152, 225, 260, 284, Kālī, 79, 89, 152, 286, 357, 416, 438, Īśvara, 127, 152, 229, 315, 416, 501 Jñānakārika of Matsyendra, 386, 421, 298, 475, 488 444; circle of Kālīs, 291; Guhya- 429, 511 Kaivartta, subcaste of fishermen, 467, kālī, 152; Kālī Tantra, 163; mantra Jabirians (Persian alchemical school Jnanasiddbi, 35 475 (see mantras) founded by Jabir ibn Hāyyan), 54, Jñānesvarī of Jñāneśvara, 95, 113, 160, kajjalī (black mercuric sulphide) 267, kalpa, 3, 17, 42, 46-47, 215-16, 251, 204, 447, 456 395, 396, 399, 408, 466 300, 388, 420, 449; Forest of, 318, 326 (see also cycles, temporal); Jain alchemy, 114-19, 376 Jodhpur, 346, 350, 391, 416, 418. See 238-39, 299-300, 476, 505 kalpa sādbana (see reversal, yogic). Jainism, tantric and siddha traditions, also Mahāmandir; Mān Singh Li- Kākacaņdeśvarī, 127, 152, 163, 416; in Kāma, 232, 284, 289, 463 2, 53, 57, 66, 114-19, 298, 331- brary Siddha lists, 86 Kāmākhyā/Kāmāksī. See Kāmarūpa 34, 371, 376, 475, 498, 515. See Jogis, 8-9, 50, 52, 56, 59, 93, 95, 98- Kākacandeśvarīmata, 127, 130-31, 145, Kamala, Queen of the Plantain Forest, also syncretism, Hindu-Jain 99, I01, 107-9, 116, 122, 130, 152-55, 159, 160, 163, 164, 166, 236-37, 297, 474 Jālandara[-nāth]/Jālandhari[-pa], 98- 170, 283, 309, 311; Muslim, 311, 167, 176, 209, 324, 371, 416, 420, Kāmarūpa (Assam), 62, 73, 89, 108, 100, 106-8, 244, 261, 295, 296, 345, 349, 350, 370, 393, 399, 406, 428, 430, 431, 453, 481, 512 414, 448, 489, 504; subcastes of, Kākacaņdīśvara, 83, 127 115, 134, 195-96, 223, 224, 234- 374, 396, 397, 403, 425, 460, 473, 476, 502, 503, 508; in Siddha 399, 404 Kākacandīśvara Kalpa Tantra, 152, 416, 36, 240, 244-45, 260, 386, 435, 451, 460, 466, 467, 473, 475, 479, lists, 81, 84, 86, 91, 92-93; jālan- Joshi, Damodar, xiii, 436, 446, 458, 428, 430 480, 482, 488, 495; location in sub- dhara bandha (see bandhana); Jālan- 491, 502 Kaksaputa Tantra of Siddha Nāgārjuna, tle body, 112, 163, 239, 389, 467; dhara pītha, 260, 496; literary out- Joshi, Shantilal Pranjivan, 410 76, 104, 112, 114, 116, 155, 419, name of a siddbi, 317, 330, 514 put, 132, 140; name of a demon, Jūna Akhāda. See Nāgas 435; adapted from Siddha Khaņda Kambali, name of a Siddha, 82, 87, 93, 116. See also Hādi-pā Junagadh. See Gujarat of the Rasaratnākara, 129, 131, 124, 390

570 Index 571 Index

Kanci. See Tamil Nadu Kathmandu [Valley], 62, 89, 96, 112, kańcukas, 213-14, 283, 462 311, 339, 343, 350-51, 378, 392, kbānepāne (vernacular term for siddbi), 160, 162, 165, 308, 397, 409, 418-

Kaņerī. See Kānha-pā 393, 396, 407, 422, 438, 442, 500, 158, 433 20, 428, 431, 432, 442, 496. See

Kānha-pā, 106-9, 236, 296, 297, 299, 519. See also Pasupatinath kbecara, 253; Khecarī (name of a Dūtī also Kautukacintāmaņi of Pratāpa-

396, 397, 399, 400, 460, 473, 503, Kaula Mata, 151 in alchemical mandalas), 441; kbec- deva; Kriyākālagunottara

507; in Siddha lists, 81, 83-84, Kaulajñānanirņaya of Matsyendra, 73, ari cakra, 198, 253, 484; kbecarī gut- Kriyākālagunottara, 129, 131, 151

88, 112, 133-36, 138, 145, 166, ikā, 153, 316, 427, 430, 487, 496; Kriyātantrasamucchayī, in Rasa Siddha 92-93 Kanjur, 383, 472 223, 228, 230, 232, 233, 313, 324- kbecari jārana, 292, 326; kbecari lists, 83, 129

Kānphatas. See Nāth Siddhas 25, 371, 421, 422-23, 432, 471, kula, 443, 461; kbecari mantra, Krsna, 410, 515

Kanthadī, 98, 400; in Siddha lists, 81, 513; early description of six cakras 170, 253; kbecarī mudrā, 135, 146, Ksāntaśīla, name of a tāntrika in the

83, 86, 91-92, 498 in, 134-35 150, 158-59, 170, 173, 230, 252, Vetālapañcavimśati, 306, 307

Kapāli[-pā], in Siddha lists, 81-82, 84, Kaulas and Kaulism, 79, 88, 90, 135- 254, 257, 258, 276, 481, 483, 487, Ksemarāja, disciple of Abhinavagupta,

86, 127, 390 37, 143, 158, 171-73, 181-82, 512; khecarī śakti, 487; kbecari sid- 136, 138, 142, 178, 418, 465

Kāpālika: alchemical method, 148, 166, 199, 228, 229, 239, 256, 291, 325, dhi (see flight, power of) ksetrīkarana (preparation of the body in

173, 417, 427, 428; Buddhist Kā- 332, 388, 432, 437, 440, 442, 447, Khecarī Vidyā, 146, 169-70, 253, 324, alchemy and fiyurveda), 266,

pālika-Yoginī cult, 136; name of a 473, 481, 486, 489, 494. See also 370, 484, 509 270-71

Rasa Siddha, 82, 127, 417 Krama; kula; Siddha Kaula; Śrīvi- Kingdom of Women, 119, 139, 219, Kubera, 58, 259, 415

Kāpālikas, 7-9, 73, 97-99, 102, 107-8, dyā; Trika Kaula; Western Trans- 223, 236, 251, 466, 489. See also Kubjikā, goddess of Western Transmis-

IIO-II, 127, 135-36, 142-43, mission; Yoginī Kaula Kadalī sion, 2, 79, 88-89, 91, 134-35,

169, 228, 238, 305, 306, 307, 310, kaula-sadbbāva. See clan nectar Kiranāgama, 213, 397 151, 168, 180, 194, 255, 390, 443,

Kaulāvalīnirņaya of Jñānānandagiri Kirātas, 147-48 321, 325, 334, 375, 386, 399, 404, 481; kubjikā mudrā, 429; Kubjika

4°5, 477, 507, 513. See also Vidyā Paramahamsa, 291, 501 Ko Hung, Chinese alchemist, 64 the name of Buddhist Mahasid-

Pītha Kautukacintāmaņi of Pratāpadeva, 131 koans, 501 dha, 74, 387. See also mantras;

karma, 3, 29, 42, 182, 185, 286, 497, kavi (alchemical wizard), 13, 154, Kolff, Dirk, 345 Western Transmission

159, 286, 323-24, 359, 374, 499, Konganar, name of a Siddha, 61, 87, Kubjikāmata, 88, 90, 105, 134-35, 145, 517 Karņari[-pā], 69, 71, 81, 91-92, 132. III; works attributed to, 202 180, 230, 233, 260, 386, 392, 422, 512 See also Aryadeva; Kānha-pā Kaviraj, Gopinath, 139, 418, 425, Konkana, consort of Macchanda in Sid- 423, 471

Karnataka, 87, 94-97, 103, 110, 113, 48I dha Cakra, 89, 135 Kubjikānityābnikatilaka, 73, 88, 91, 181,

134, 147, 392, 393, 396, 401, 406, Kāvya Uśanas, 285-86, 323, 374, 499, Konkana, west coast of India, 73, 88, 230, 330, 386, 392, 422

407, 408, 422. See also Kadri; Vijay- 512 91, 94, 230, 393 Kubjikāpūjāpaddbati, 421

anagara (Hampi) Kāyanāth, 287-88, 500 Końkaneśvara, name of a Siddha, 86- kula, 200; as tantric clan or lineage, 35,

Kārttikeya. See Skanda Kedārakbanda, 396 89,168 137-38, 172, 210, 330, 357, 443,

Kashmir, 60, 66, 97, 103, 106, 114, Kedārnāth, 175, 193, 196, 245-46, Korakkar. See Gorakh, south Indian tra- 460, 461 (see also kbecari kula);

179, 192, 306, 308, 381, 397, 424, 254, 260, 331, 351, 468, 480, 484, ditions of kula-parvata, 112; kula-siddhas, Kosinski, Jerzy, 455 431, 444, 514 488, 490 324, 444; kulayāga, 255; Kuleśvara

Kathāsaritsāgara of Somadeva, 306, Kerala, 112, 155 Krama, 79, 181, 291, 416 and Kuleśvarī, 135; term for early

325, 355, 407, 413, 415, 456, 513. Ketakī, 295-96, 503 krāmana (progression of mercury), tantrism of the Vidya Pitha, 73,

See also Vetālapañcavimšati kba. See ether 149, 268, 492 88, 90, 101, 136, 224, 230, 438,

Kathiawar. See Gujarat Khan, Dominique-Sila, 468 Kraus, Paul, 457 517 Kriyā Tantras, 129, 131, 146, 152-54, kuladravya. See clan nectar

572 573 Index Index

kulāmrta. See clan nectar Lakulīśa, Pāśupata founder, 97, 126, lobavāda. See dhātuvāda mabārasas (primary alchemical re- Kulānanda Tantra of Matsyendra, 152, 474 Lokāloka, 510 421, 430 Lalitāvistara, 152 Lokanātha, 70, 94, 128, 142 agents), 145, 147-48, 154, 155, 158, 160, 164, 166, 177, 187, 384, Kulārnava Tantra, 92, 145, 312, 314, Lallã (Lāl Ded), Kashmiri mystic, 250, lower abdomen, 27, 40, 201, 207, 218, 441, 446 332, 356, 367, 389, 415, 438 482, 505 221, 230-31, 234, 240, 249, 253, Maharashtra, 104, 110-14, 141, 147, Kulavīra Tantra of Matsyendra, 421 laya. See resorption 308, 312, 367, 454, 481. See also 160, 178, 391, 393, 408, 477, 515. Kumāra. See Skanda lead (nāga), 189, 190, 207, 208, 214, cranial vault; linga, western; mū- See also Devagiri; Godavari River Kumāra Tantra of Rāvaņa, 58, 125, 216, 284, 431, 447, 452, 458, 459, lādbāra Mahāsiddhas, 2, 69, 71, 74, 80-82, 84, 150-51, 373 463. See also Serpents Luĩ-pā, 107-9, 397, 403, 467; in Sid- 87, 89, 92, 99, 101, 106-8, 224, kumbbaka. See diaphragmatic retention left side of body, 230-32, 251-58. See dha lists, 80, 84-85, 91, 224, 231; 251, 367, 386, 395, 418, 477; list kundagolaka, 154, 198, 424, 453 also linga, western Tibetan translations of name, 224 of in Hatbayogapradīpikā, 82, 92. kundalas (earrings worn by Nāth Sid- Lévi, Sylvain, 104, 393 Lūņā (Lonā; Nonā) Chamāri, 298- See also Eighty-four Siddhas dhas and other Saiva sectarians), Life of Apollonius of Philostratus, 205, 300, 403, 453, 484, 504 Mahāvidyā goddesses, 516 99-100, 107, 400, 403. See also 457 Mahāyāna Buddhism, 35, 43, 53, 60, mudrās light: body of, 72; drop of (tejobindu), Macchanda. See Matsyendra 316; metaphysics, 385; prakāśa, Madana, Kalacuri king, 147-48, 428, 63, 66-70, 72, 105, 151, 201, 455, kundalinī, 4, 79, 163, 170, 195, 198- 501 200, 208, 218-22, 225, 233, 234, 263, 427; vimarśa, 220, 263, 427 437 Mahāyāna literature, 67, 76, 201, 371, 235, 239, 243, 247, 251, 252, 254, lineages. See paramparās Madanavinodanighantu of Rāmarāja, 373, 383, 381, 382, 430, 516 255, 256, 263, 276, 289, 290, 293, linga, 4, 120, 194, 287, 375, 393, 397, 436 mabāyuga (unit of time), 46, 215, 308, 312, 313, 320, 322, 367, 450, 440, 444, 464; calcinated mercury Madhya Pradesh, 29, 54, 64, 97, 147- 318 464, 465, 468, 470, 471, 485, 486, takes shape of, 194, 292; earliest 49, 298, 364, 370, 383, 397. See Maheśvara, IO1, 162, 196, 325, 495 489, 501; waking and sleeping of, excavated, 356; Goddess's pinda also Malava; Ujjain Māheśvara Siddhas, 2, 77, 99, 101-3, 216-22, 232, 234, 240, 291, 319, images have form of, 500; internal Madras. See Tamil Nadu 11I-12, 145, 147, 155, 372, 392, 367, 476, 51I to subtle body, 170, 231, 218; jyo- Magee, Michael, 449 401 kundas. See waters, magical tirlingas, 11I, 161, 245, 398; mer- magical alchemy, 52-53, 57-66, 72, 74, Mabodadbi, 127, 155, 157, 416. See kūrpa. See samskāras curial (rasa-linga), 103, 131, 157, 76-77, 499. See also añjana; pāda- also Rasopanisat Kuruksetra, 91 163, 174, 176, 194, 258, 260, 440, lepa; rasa-rasāyana Maithili literature, 108, 391, 474. See kutīpraveśa ("entering the hut"; sum- 441, 450, 487; moveable, 486; Mabābbārata, 43, 60, 121, 238, 248, also Varņaratnākara mum of Ayurvedic rejuvenation term for cinnabar in south Indian 285, 331, 367, 371, 375, 410, 414, Malabar coast, 9, 50 therapy), 26-27, 169, 245, 258, languages, 194; transmutational, 422, 459, 481, 499, 515 Mālatī-Mādbava of Bhavabhūti, 73, 273, 364, 366, 378, 479, 484, 491; III, 407; "western," 231, 225-56, Mahādeva: name of a Rasa Siddha, 83; III, 134, 423 structure of the kuti, 364 464, 471, 476, 486. See also Siva, name of Siva, 155, 398, 414 Malava, 49, 416, 428 cult of Mahākāla, 49, 71, 98; Mabākāla Sam- Mālinī, 176-77, 441, 442. See also Trika laboratory, alchemical, 140, 145, 158, liquification, 148, 164, 166, 291, 417, bitā of Ādinātha 42 1; Mabākālayo- Kaula 176, 180, 182, 194, 196, 222, 245, 439. See also garbhadruti; bābya- gašāstra, 169; Mabā-kāla Tantra- Mallik, Kalyani, 395 258-60, 487 truti rāja, 471 Mallikārjuna. See Srisailam laboratory assistant, 150, 158, 162, Lohasarvasvam of Sureśvara 191, 418 Mahāmandir, 344, 346, 508. See also Mãn Singh, king of Marwar, 244, 310, 167, 172, 196, 198, 452 Lobaśastra, 76 Jodhpur 344-45, 348, 391, 412, 479, 501.

574 575 Index Index

Mãn Singh (continued) 179, 180, 444; kālī, 179, 181; khec- 229, 231, 324-25, 393, 396, 399, menstrual blood. See cycles, menstrual; See also Natba Caritra; Śrī Nath ari, 170; kubjika, 180, 429, 444, 447, 473, 512, 513; as honorific ti- sexual fluids Tirthāvalī 461; kulakbecari, 444; mālā, 179; tle, 75, 166, 335, 387, 466; Bhai- Mãn Singh Library, Jodhpur, 416, 418, Mantreśvaras, 134; mūla, 179; pra- rava takes form of, 224, 467; cult mercuric sulphide. See cinnabar; kajjalī

421 nava (Om), 178, 179, 291, 444, of, 94, 224, 414; fourth yuganātha mercury, 9, 11, 16, 50-52, 70, 143,

464, 471; rasānkuśa, 179, 278, 443 of Siddha Cakra, 135, 181; in Sid- 147-50, 154, 158-59, 162, 166, manas. See mind 170, 172, 174-75, 222, 234, 235, Mānasollāsa of Someśvara III "Bhuloka- (see also Rasānkuśī); sakala and dha lists, 80, 84-85, 87, 91-93, niskala, 442; tripura bhairavī, 444; 400; legends of, 119, 134, 222-40 248-52, 256, 257, 259, 369, 377, malla," 131, 393, 409 384, 409, 420, 428-33, 481, 490, mandalas, 54, 78-80, 124, 145, 158, vāyu, 445; yoginī, 440 passim, 244, 251, 261, 296, 297, 168, 175-79, 182-83, 192, 258- Manu Smrti, 368 298, 299, 412, 451, 475, 476, 503, 491, 517; as Siva's semen, vital

504; literary works, 133-36, 42 1 fluid, 5, 14, 25, 36, 104, 146, 188, 59, 384, 389, 426, 441-43, 445, mārana: "fifth state" of consciousness, (see also Kaulajñānanirņaya and ti- 190-91, 194, 198, 203, 208, 212- 450, 483, 487 498; killing of mercury, metals, tles of individual works); matsyendrā- 14, 221, 246, 248, 265, 284, 289, Māndavya, name of a Siddha, 82, 125, 153, 164, 266, 273, 275, 281, 290, sana, 220; names of, 467; praised 300, 316, 326, 439, 440, 442, 443, 390 299, 300, 491, 496 (see also mer- 448, 449, 450, 463, 491, 504; bhas- Maniknāth (Manikcandra), name of a cury: bbasmasūtaka, mrtabbasma/ by Abhinavagupta, 88, 393, 393, Siddha, 289, 295, 333 mrtasūtaka; swooned, killed and 423; (spurious) guru-disciple rela- masūtaka, 275, 284; chemical be- havior of, 6, 277, 314, 357, 370, Maņimekbalā, 456 bound mercury) tionship with Gorakh, 85, 139, 286, 288-89, 321, 467, 489. See 417, 453, 454, 461; clipping the Mañjunātha, 94 Marathi literature, 87, 92, 95-96, 332, wings of (paksa-cchedana), 211, Mañjuśrī. See bodhisattvas 391, 395, 408, 411. See also also Acintanātha; fidināth; charis- 278, 452, 461, 496, 506; comple- Mañjuśrī Mūlakalpa. See Mahāyāna lit- Jñāneśvarī matic impersonation; Luĩ-pā; mentarity with breath (see Work erature Marco Polo, 9, 50, 52, 56, 95, 101, 370 matsyodara; Mīna; Siddhanātha in Two Parts); dosas of, 162; earli- Mankaņaka, 284 mardana (rubbing, trituration of mer- Matsyendra Sambitā, 421 est prescription for internal use, Manthānabhairava, 83, 126, 129, 143, cury), 148, 266, 273, 299, 300, Matsyendranāth. See Matsyendra 148, 151, 156, 158, 162, 386, 416, 409 matsyodara (fish belly), 88, 135, 138, 52; extraction of, 203, 247, 325, 480, 494; five names for, 167, 186, 417, 418, 430; mantbānabbairava mātra (unit of time), 46, 294, 318, 369, 222-29 passim, 231-33, 235, 240, 187, 188, 190, 191, 194, 197, rasa, 126, 151, 160, 436. See also 502 467-70, 490 200-3, 205-8, 211, 212, 213 (see Bhairava Mātrkabbeda Tantra, 131, 146, 154, māyā, 220, 222-23, 278, 279, 297, 426, Manthanabbairava Tantra, 88, 150-51, 162-63, 176, 193, 234, 255, 256, 438; viśuddha and aśuddha, 102, also miśraka; pārada; rasa; rasendra;

271, 325, 405, 420, 449 178, 210, 213-14 süta); five states or phases (avas- 164, 256, 386, 392, 425, 429, 432 mātrkas. See mother goddesses Māyana[-matī], mother of Gopīcand, thã), 459; gatis of, 209, 459; gods adulterate with impurities, 192, Mantra Pītha, 79, 156, 176, 181, 417. Mastnāth, 244, 501, 518 92-93, 280, 295-96, 517 213, 264, 449, 490; identified with See also Svacchanda Tantra Matsyendra, 2, 9, 74-75, 95, 100, McDaniel, June, 229 atman in Rasopanisat, 155; in the mantras, 49, 54, 59, 139, 143, 152, 154, 106-7, 109, 129, 133, 138, 157, medial channel, 28, 39, 41, 45, 201, subtle body, 209; maritime trans- 155, 157, 163, 169, 174, 176, 179- 171, 172, 200, 278, 386, 418, 460, 211, 218, 219, 221, 226-28, 231- 33, 235, 241, 243, 247, 252, 254- port of, 63-66; mrtabbasma/ 83, 209, 220, 271, 280, 287, 293, 468, 500; as fisherman, 224, 447, 467, 513 (see also Kaivartta); as 57, 276, 277, 279, 282, 290, 291, mrtasūtaka, 207, 289, 300, 267, 299, 310, 417, 423, 439, 443, 490, 282, 329, 330, 331, 491, 495 (see 517; aghora, 178-79, 181, 417, founder or revealer of Siddha and 319, 464, 469, 470, 477, 480, 481, also māraņa); ores (see cinnabar; 433, 440, 459; agni, 445; dāmara, tantric lineage traditions, 5, 73, 485, 490 kajjalī; sindūra); origin myth of, 444; bamsa, 179, 444; kālapāśa, 88, 91, 99, 136, 142, 177, 223-24, meditation, 265 159, 168, 191, 212, 234, 247, 265,

576 577 Index Index

mercury (continued) 298, 397, 423, 466, 470, 473, 474; 283, 472; "six-times killed," 105, guru-disciple relationship with 367, 393. See also Candradvīpa; mūrcchana (swooning), of mercury, Candragiri; Candrapuri; cranial 166, 292, 402, 502; trans- Gorakh, 85, 139, 228, 467; in Sid- 266, 273-75, 290, 492, 495; yogic, vault; kalās; month, lunar; sun Himalayan sources of, 62-66, 69; dha lists, 80, 84-85, 91-93, 251; 228, 257, 274-75, 495, 498. See mother goddesses, 43, 135, 157, 176, receives Kaula revelation from also swooned, killed, and bound treated with sexual fluids, 153, 163, 452, 493; used to fortify tem- Śiva, 225, 468. See also Matsyendra 324, 361, 424, 457; Seven Sisters, mercury

mind, 34, 38, 45, 239, 242, 274, 275, 29, 499 Murthy, K. R. Srikanta, 420, 437 ple sculpture, 103, 401. See also Motināth, 350 Murugan. See Skanda bandhana; gold, amalgamation 277, 278, 279, 291, 293, 302, 338, with mercury; gutikā; linga, mercu- 340, 477, 483, 496, 497, 501, 505, mountains, sacred, 114-19, 327-34. See Mus, Paul, 360

rial; māraņa; pārada; swooned, also Abu (Mount); Girnar; Himala- Muslim conquest of India, 49, 70-71, 517 killed, and bound mercury; Work minerals, 145, 148, 177, 266, 268, 322, yas; Kedārnāth; Meru (Mount); 103-7, 116, 305, 345, 394, 402, Srisailam 476 in Two Parts 351, 352, 370, 417, 420, 441, 442; Meru (Mount), 190, 230, 328, 514, origin myths of, 188; essences of mouth, 193, 200, 212, 252-58, 471, Muslims and Hindus. See syncretism,

(sattvas), 147, 268, 481. See also ar- 482, 487; mercury becomes pos- Hindu-Muslim 516; meru-danda, 328 sessed of, 6, 256, 267 (see also metals, 145, 155, 158, 162, 168, 188- senic; cinnabar; kajjalī; nummel- 91, 206, 208, 214, 216, 224, 240, ite; orpiment; red arsenic; mahā- grāsa); passage of sublimated sex- nāda (resonance), 43, 202, 211, 236, ual fluids from mouth to mouth, 259, 264, 267, 268, 314, 315, 316, rasas; sindūra; uparasas 251, 274, 280, 291, 301, 444, 456,

319, 326, 376, 384, 440, 441, 446, Misra, Siddhinandan, xiii, 127, 337, 137, 200, 255, 312, 313, 321, 322, 484, 496, 501; location within sub-

447, 458, 459, 490, 503, 506; iden- 440, 446, 452, 462, 487, 493, 498, 439; sealing of (see samputa); tle body, 178, 368; reversing the,

tified with celestial bodies, 338- Yogini's upper and lower, 234-35, 293. See also anābata; phonemes 516 39. See also dhātus miśraka (name for mercury), 208 240, 254, 256, 454, 464, 467, 476, nādīs (breath and energy channels),

metaphysical categories. See tattvas Mitra, Rajendralal, 418 482, 485, 486 (see also kundalinī; 11I, 170, 226-28, 232, 247, 251-

Meulenbeld, Jan, xiii, 390, 426, 434 moksa, 102, 143, 145-46, 155, 173, śankinī; Yoginī; yoni) 54, 257, 260, 273, 276, 277, 282,

Mewar, 120, 345, 412. See also Bāppā Mrcchakatikā of Śūdraka, 60, 422 185, 220, 265, 279, 304, 465 Mrgendrāgama, 134, 491, 501 301, 319, 321, 464, 469, 480, 495,

Rāwal Mongolia. See Central Asia 506; three portrayed as primal Sid-

mica, 6, 153, 154, 187, 189, 197, 198, month, lunar, 25-26, 36-44, 46, 248, mudrās: alchemical seals, 256; ankušā dhas Sun, Moon, and Fire (see Sid- dhas, three primal). See also banka 210, 268, 292, 294, 301, 316, 421, 277, 318, 340; amāvāsyā (new mudrā, 432, 496; earrings worn by Näth Siddhas, 281, 299, 310, 321 moon), 36, 38, 197, 229, 244, 250, nāla; medial channel; śankbinī 429, 446, 452, 459, 460, 463, 498, 501; as Goddess's sexual emission, 306, 307, 366; full, 197 (see also kundalas); one of the pañ- Nāgabodhi, 69, 81-82, 104, 124-25,

Mookerji, Bhudeb, 402 camakāra, 234; šāmbbavī, 248, 481; 390. See also Nāgārjuna 5, 146, 210, 212, 442. See also vyoma moon, 11, 18, 21, 23-29, 36-44, 178, yogic seals, 220, 265, 274, 276 Nāgārjuna, 60, 62, 64-70, 73, 75-76, (see also kbecarī; kubjikā; vajroli) 103, 106, 110-II, 113-14, 117, microcosm-macrocosm-mesocosm, 1I, 184, 185, 189, 197, 202, 205, 226, muktivāda. See moksa 229-32, 242-44, 246, 250-52, 121, 125-26, 131, 133, 143, 150- 15-23, 28-29, 33, 38, 46, 78, 139, 184, 186, 202, 218-62 passim, 277, 280, 282, 285, 296, 301, 308, mūla bandba. See bandhana and bandbas 51, 158, 160, 165, 215, 218, 312, mūlādbāra (the root cakra), 40, 163, 263, 269, 272, 280, 283, 291, 293, 313, 315, 323, 445, 455, 456, 457, 327, 333, 339, 378, 380-82, 387,

294, 327-34, 349, 360, 367, 447, 468, 471, 478, 482, 514; King 208, 234, 241, 243, 256, 293, 301, 410, 411, 414, 415, 422, 430, 464,

463, 465, 476, 488, 511, 519 Moon, 24-27, 44, 184, 190, 339, 319, 469, 482, 490

Mina[-nāth], 95, 107, 223, 231, 296, Mūlanātha, 85, 87-88, 168, 392 516; as Ayurvedic author, 370,

341, 359, 517; Siva and 37, 366, Muller-Ortega, Paul, 468 372, 381, 388, 431 (see also Rasa- vaiśesika Sūtra); as honorific title,

578 579

Index Index

Nāgārjuna (continued) -ānanda, dās, pā, 107, 394; Saiva nectar, 5, 76, 173, 184, 190, 192, 199, Nityānanda Siddha, 390, 418 75, 166, 335, 466; in Siddha lists, and Gorakhnathi lineages, 99 207, 212, 285, 299, 300, 309, 313, Nityanātha Siddha, 83, 129, 160, 407, 81-82, 84, 86, 91-93, 164, 397; 399, 474. See also Aughars; twelve 320, 328, 349, 355, 437, 455, 462, 408, 417; "son of Pārvatī," 129-30, Jain traditions of, 114-19, 480, panths; Räwal; Vairāg panth 481-84, 496, 498, 506; in the 160, 418, 435. See also Rasaratnā- 505; Nāgārjunanāth, 11I, 165; nā- Nāth Siddhas, xi, 2-4, 6, 9-10, 43, 56- Vedas, 10-11, 358; "red-white," kara; Śabaracintāmaņi gārjuna-varta (eye salve), 76; Sid- 57, 63, 66, 73, 86, 88, 102-22, 202; three types, 327; yogic gener- Nityasodasika Tantra, 163, 394 dha Nāgārjuna, 69, 74-76, 11I, 131, 168, 185, 191, 196, 198, 202, ation of, in cranial vault, 28, 38- nivrtti. See withdrawal and return, 165, 408; Śrī[man] Nāgārjuna, 76, 210, 216, 222, 223, 228, 230, 235, 39, 41, 218, 221-22, 242-43, 247- yogic

155, 164, 435; the bodhisattva, 67- 238, 241, 242, 244, 251, 254, 258, 50, 252-54, 257, 258, 272, 277, Nivrttināth, 113, 396 68, 381; the Mādhyamika philoso- 261, 280, 287, 288, 289, 294, 295, 282. See also waters, magical; ex- niyamana (regulation, restraint of mer- pher, 66-67, 76, 104, 165, 380; Ra- 297, 298, 318, 349, 352, 385, 391, traction; clan nectar; soma cury), 267, 290, 461 saratnākara spuriously attributed 393, 395, 420, 425, 465, 478, 481, Needham, Joseph, 63, 204 numbers and number symbolism, to, 104-5, 160; portions of Rudray- 506; as alchemical authors, 130; as Nemināth/Nimnāth, 93, 119, 298, 15-47 passim. See also five; geo- amala Tantra spuriously attributed guardian spirits of sacred moun- 331-32, 399, 475 metric progressions; six; sixteen; to, 130, 419. See also Arjun Nāga; tains, 332-33, 335, 339; as power Neoplatonism, 189, 490 three; twelve

charismatic impersonation; Na- brokers, 8, 309-11, 343-49, 398, Nepal, 2, 65, 89, 98, 105, 109, 119, nummelite, 196. gnāth; and individual works by Nā- 508, 519; called Gorakhnāthis, 99; 134, 152, 224, 251, 278, 288, nyāsa. See worship, tantric

gārjuna. called Jogis, 9, 99, 130; called Kän- 310-12, 321, 333, 349-51, 380, Nyingma. See Vajrayāna Buddhism Nāgas, Saiva order, 99, 254, 278, 345, phatas, 9, 99-100, 321, 399, 413; 389, 391, 393, 431, 435, 465, 467, 381, 398, 484, 500. See also Ser- cult of, 374, 414; definition of 473, 502, 508; Gurkhas in, 310- Obermiller, Ernst, 471

pents 11, 335, 343-44, 357 (see also Prthi- Obeyesekere, Gananath, 341, 517 Nāgnāth, 2, 92-93, 98, 414. See also Nā- yoga, 456; four or five original, 210, 236, 296, 460; give sons to vinārāyan Sāh); National Archives, Ocean of Milk, 190, 192, 221, 234, gārjuna barren women, 196; lists of, 418, 421, 423, 426, 431, 434, 472. 289, 447, 466, 475 Nalanda, 64-65, 68-71, 76, 103-5, 83-86; 91-100, 123; regalia of, See also Kathmandu [Valley] ogbas ("streams" of Siddhas), 3, 92, 382, 401 122, 195, 280, 310, 349, 413, 489, Nepali literature, 108, 465, 474 135, 327, 332, 356, 386, 394, 415 Nandi[keśvara], name of a Siddha or 501 (see also kundalas; mudrās); Netra Tantra, 178, 418, 419, 442, 501 ojas (the "eighth dhatu"), 184, 341, 363, alchemical god, 61, 76, 83, 87, theories concerning Buddhist ori- Newman, John, 384 477, 506, 517 126, 129, 143, 157, 158, 160, 375, gins of, 104, 106-8. See also Jogis; Nicholas, Ralph, 475 Old Bengali language, 81, 107, 236,

440 Nāth sampradāya; Nine Nāths; Nīlakaņtha, 97, 126, 142, 221, 447 390. See also Caryāpadas Nārāyan Nāth, 93-94, 396, 399 twelve, Nath panths; and names of nine doors, 163, 254, 489, 504. See also Old Kannada language, 140, 407. See Narharināth, xii, 118, 349, 391, 517 individual Nath Siddbas tenth door also Yogadīpikā Näth, name for Siva, 99, 166; title Natha Caritra of Mān Singh, 411, 451, Nine Nāths, 85-88, 90-100, 109, 112, Old Punjabi language, 96, 132 meaning "guru" or "lama" in Ben- 474, 516 123, 135, 169, 181, 260, 330, 332, Old Rajasthani language, 96, 132 gali Vajrayāna, 399 Nathābbyudaya Tantra, 231, 471 339, 356, 391, 394, 395, 396, 408, Om. See mantras, pranava

Nath sampradāya, 7,90-100, 107, 121, National Museum of Denmark, 483 484, 488; identified with Nine Oman, John Campbell, 51-52, 349, 161, 162, 172, 287, 309, 310, 333, Navanāthacaritra of Gaurana, 92, III, Nārāyaņas in Marathi traditions, 370

344, 346, 347, 398, 399,400,473, 165, 407, 415 395, 411. See also Nāth Siddhas, opium, 159, 167, 434 487, 500, 504, 519; householder navel, divides upper and lower body, lists of oral transmission, 2, 145, 161. See also Näths, 466; name endings in 243 nirvāna, 33, 35, 201, 375, 482 written transmission

58o 581 Index Index

Orissa, 98, 147-48, 224, 345, 391, 392, parāmukti, 102 Peshawar. See Pakistan Prajñā[-paramitā], 67, 72, 104, 152, 434 Parāparaprakāšika, 314 phonemes, 43, 179-80, 274, 367, 368, 164, 378, 382; and Upaya, 247, orpiment (baratāla), 153, 170, 320, paras pattbar (alchemical touchstone), 452, 471, 486; identified with 251 329, 331, 429, 441, 453, 514 407, 413 Pārasnāth/Parśvanāth, 93, 114, 119, mātrkas, 291. See also bindu; nāda prakāśa. See light piercing, 289, 303, 313, 320, 463; of Prakrti (Nature, feminine principle of pādalepa, 187, 305 298, 399, 475 Padmavat of Mallik Muhammad Jayasi, cakras, 40, 135, 293, 314-18, 319, materiality), 20-21, 33-34, 36, Parasurāma Kalpa Sūtra, 496 Parātrimšikāvivarana of Abhinavagupta, 489, 511. See also ear-boring; vedha 195, 210, 215, 218, 473, 492 198, 238, 260-62, 281, 391, 453, Pietro della Valle, 94-95 pralaya (cosmic dissolution), 3, 232-33, 489, 505 423, 439, Paduba-doba of Muni Rāmasimha, 117 Pārvatī, 155, 195, 284, 357, 451, 468, Pigalle, Paris red-light district, 341 289, 296, 297, 317, 326, 463, 513;

Pakistan, 62-63, 66, 106, 116, 121-22, pindas, 287, 500 as withdrawal of cosmic yogin, 33, 473, 499. See also Nityanātha Sid- dha, "son of Pārvati" pingalā. See nādīs pītbas: Buddhist, 72, 260, 385, 389, 482 41, 46-47, 221, 251, 263, 294, 204, 311, 374, 380. See also Hin- glāj Devī; Punjab; Sind Pascal, Blaise, 318 463 (see also Visņu Nārāyaņa); pro- (see also Jālandhara; Kāmarūpa; Ud- voked by dance of Śiva and Kālī, paksa-ccbeda. See mercury, clipping the paścimāmnāya. See Western Trans- diyãna); chasing in which a Siva 286; within subtle body, 40. See wings of mission linga housed, 258; sākta pīthas, sa- also samtbāra; srsti pala (measure of time and mass), 294, paśu, 240, 279, 323, 465. See also vīra Paśupata Sūtras, 360, 397, 513 cred sites of the Goddess, 91, 112, "Prāņ Sānkali" of Caurangi[nāth], 133, 316-19, 502, 510 Pala dynasty. See Bengal Pāśupatas, 7, 9, 49, 97-99, 102, 110, 195, 210, 260, 245, 246, 467, 488 260, 404, 416, 504 (see also worship, tantric); tantric pañcakarmāni (five treatments of Āy- 118-19, 121-22, 126, 284, 285, prāna, 22, 39, 184, 185, 200, 211, 226, literary classification system, 79- 316, 359, 445, 454. See also breath urveda), 270 310, 325, 334, 375, 393, 397, 413, 80, 181, 389, 424 (see also Mantra 418, 499, 515; definition of yoga, prāņāyāma. See breath, control pañca-makāra, 163, 172, 180, 210, 220, Pītha; Vidyā Pītha) prasāda (gift of grace from god or Sid- 234, 323, 424, 460 397; -rāsi name endings of, 121 dha), 174, 288, 311, 512 pañcāmrta. See extraction (see also Kālamukhas). See also pitryāna. See year, solar

Pandeya, Janardana, 156, 356, 390, Plantain Forest. See Kadalī pravrtti. See withdrawal and return, Lakulīśa Paśupatināth, 89, 351, 392, 519 poison; 164-65, 190, 209, 221, 222, yogic 432, 490, 493 Prthivīnārāyan Sāh, Gurkha king of panths. See twelve, Nath panths pātana (sublimation, evaporation of 239, 289, 329, 338, 376, 447, 465,

Parā-Aparā-Parāparā, 79, 91, 157, 476; Tantras, 131, 419, 465 (see Nepal, 310, 311, 313, 321, 335, mercury), 212, 267, 290; pātana also Garuda Tantras; Yogaratnāvali) 176-77, 441, 442. See also Trika yantras (three types of sublimation 343-44, 348 power substances, 4, 135, 137, 153, Pseudo-Zosimus, 203-5, 447 Kaula apparatus), 203, 247-49, 258, 480, 158, 163, 167, 173, 183, 198-200, Ptolemy, 205 pārada (name for mercury), 188, 208, 483 323, 430, 433, 454. See also clan Punjab, Indian state of, 1, 98, 398; Paki- 370, 373, 381, 437, 448, 459, 462, Patañjali, 87, 216. See also Yoga Sūtras Patna, 61-62, 64, 76, 104, 378, 382 nectar; yonitattva stani state of 62, 239, 298-99, 498; Pārada-deśa, 65, 204, 380; Pāradakalpa, 130, 419; Pārada Sam- Pattan, 116, 295, 348, 411 Prabandbacintāmani of Merutunga, 114, 333, 404, 479, 500 (see also Jhelum) bitā, 415 Pavannāth, 289 409, 504 Prabandbakośa, 114, 119 Pāradeśvara, 157, 439 Punjabi literature, 92-93, 239, 244, penetration. See piercing; vedba parakāya-praveśa, 236, 474 Persia, exchanges with India, 2, 54, Prabhudeva. See Allama Prabhu 298, 333, 476, 500, 502, 503

Parameśvara, 330 Prabbulingalilā of Pidapatti Somanātha Pūraņ Bhagat, 92, 239, 244, 261, 289, 205, 246, 247, 378, 409, 446, 449 paramparās, 78, 92-93, 100, 123, 142, Persian alchemy, 54, 204. See also Jabi- Kavi, 111 298-301, 333, 403, 476, 484, 489, Prajāpati, 12-13, 17, 23, 32-33, 189, 171, 390. See also clans; kula 500, 503, 504. See also Carangi[- rian 271, 272, 283, 286, 351, 365, 492 nāth]; wells

582 583 Index Index

Purāņas: Agni, 396; Bhāgavata, 58, 317, Rajatarangini of Kalhana, 308 rasacikitsā (mercurial medicine, chemia- Rasaratnasamuccaya of [Jain author] 373, 374, 510, 514, 516; Bhavisya, rājayaksma (royal consumption), 24- try), 55, 159, 162, 175, 372, 447 Māņikyadeva Sūri, 411 410, 406, 411; Brabmāvaivarta, 25, 36, 44, 341, 363 Rasacintāmaņi of Anantadeva Sūri, 130 Rasaratnasamucchaya of Vāgbhatta II, 396; Buddha, 496, 500; eighteen, Rāma, 87, 347 Rasadarpana, 421 124-130, 146, 158, 162, 167-68, 163; Garuda, 125; Kālikā, 195, Rāmānandīs, 121 Rasādbyāya of Kankāla Yogī, 130 451, 476; Linga, 116, 227, 410, Rāmāyana, 58, 189, 214, 359, 415; Ben- Rasabrdaya Tantra of Govinda, 55, 83, 247, 274, 294, 372, 376, 390, 427, 429, 492; lists of Rasa Siddhas in, 470, 494; Mārkandeya, 141, 425, gali recension, 355; Irāmāvatāram 102, 105, 125, 130, 146-49, 155, 81-84, 113, 160, 190-92, 197, 429; Matsya, 331, 515; Nārada, of Kampan, 58, 125 167, 198, 208, 274, 314, 372, 402, 210, 212 386; Padma, 375; Skanda, 117, Rānā Samarsi, king of Mewār, 345 426, 431, 435, 437, 492 Rasaratnāvali, of Guru Datta Siddha, 227, 233, 331, 448, 466, 469; Vis- rañjana (coloration of metals by mer- Rasakalpa, 128, 417, 130 133 nudbarmottara, 103 cury), 148-49, 259, 268, 384, 417, Rasakāmadbenu, 156, 190, 210, 214 Rasārnava, xi, 5, 53-57, 59, 71, 101-2, Puri (Jagannāth). See Orissa 441, 451, 492 Rasakaumudi, 109, 415 104, 113-14, 133, 137, 145, 148- purification, 22, 148, 180, 237, 264, rasa (vital fluid), 4, 11, 24-25, 27-28, Rasakautaka, 448 55, 158-59, 160, 162, 163, 164, 265, 266, 269-73, 285, 286, 428, 184-88, 229, 275, 303, 361, 445; Rasālu, son of Püran Bhagat, 299-300, 167-68, 171-83 passim, 188, 190, 457, 459, 495. See also bbūtaśuddbi as chyle, 21, 185, 207, 340, 436; es- 413 194, 196, 198, 206, 207, 209, 210, Purusa: Vedic god of sacrifice, 16, 37, sential compound, 439, 446; name Rasānkuśa (Elephant Goad of Mer- 245, 260, 264, 274, 275, 278, 282, 207, 216, 269-70; transcendent for mercury, 11, 172, 186, 187, cury), 155, 182; mantra (see man- 284, 297, 301, 315, 316, 320, 323, soul in Sāmkhya, 20, 33-34, 36, 271, 274, 283, 303, 327, 370; of tras); name of a Rasa Siddha, 83, 324, 326, 327, 328, 363, 374, 396, 492, 510. See also Prajāpati; creator god, 269; smelted ore, 128. See also Bhairava 398, 401, 413, 426, 434, 437, 442, Prakrti 446; taste, 185, 362; theory of aes- Rasānkusī, 128, 155, 159, 168, 174-77, 487, 510; authorship of, 127, Puspabhūti, name of a king in the Harsa- thetics, 186; yogic mahārasa, 200. 179-80, 258, 443, 444 429-30; mention of Siddha Cakra carita, 307 See also mabārasas; uparasas Rasapaddbati of Bindu, 130, 481 in, 133; possible influence of putas. See sampuța rasa-agni-vāyu. See sacrifice, Vedic Rasapradīpa of Prāņanātha, 130 KJiN on, 133; tantrism in the, rasa śāstra (Ayurvedic pharmacy), 52- Rasaprakāśa Sudbākara of Yasodhara 171-83 passim quicksilver. See mercury 53, 55, 105, 117, 145-46, 188, Bhatta, 117, 128, 130, 146, 158- Rasārņava Kalpa, 160, 428 292, 361, 371, 372, 388, 400, 415 60, 375, 434 Rasasanketakālikā, of Camuņdakāyas- Raghunathan, N. Sethu, xii, 373, 407, Rasa Siddhas, 2-5, 9-10, 14, 99, 110- Rasarājalaksmī of Visnudeva, 130, 157, tha, 130, 400, 448 415, 447, 456, 458 22, 143-44, 146, 168, 238, 274, 164, 400 Rasasāra of Govindācārya, 130, 155, Raivata (medieval name for Girnar), 297, 327, 334, 356, 390, 418 (see Rasarājaśiromaņi of Paraśurāma, 130 384, 431 329-32, 515 also names of individual Rasa Sid- rasa-rasāyana (magical elixir siddbi), 53, Rasasiddbanta, 373 Rajamārtanda of King Bhoja of Malava, dhas); lists of, 74, 78, 80-87, 90, 58, 70, 187, 324, 371, 420 Rasasiddbišāstra of Vyādi, 383, 415 131, 429 92, 100, 123-31, 159, 167, 187, Rasaratnadīpikā of Rāmarāja, 130, 436 Rasasindhu of Visnudeva, 130, 448 rajas. See gunas 429, 433, 441; paucity of evidence Rasaratnākara of Nityanātha Siddha, Rasavaisesika Sūtra of Bhadant Nāgār- Rajasthan, 93, 98, 103, 110, 118-22, for lineages of, 101; theories con- 104-5, 111, 113, 129-31, 145, juna, 381, 431 147, 225, 309, 310, 344, 349, 350, cerning Buddhist origins of, 155, 157, 158, 160-62, 165, 167, Rasāvatāra, 491 363, 374, 396, 398, 399, 401, 412, 104-6; worshipped, 169 (see also 168, 188, 372, 400, 405, 417, 434, rasaviddhi. See vedba 413, 434, 468, 476, 503, 518. See Siddhas, cult of) 435, 48o; list of Rasa Siddhas in, rasāyana: alchemical elixir, 49, 58, 70, also Abu (Mount); Jodhpur; Rasabhairava, 128-29, 159, 168, 174- 81-83, 124, 160; Mantra or Sid- 79, 258, 417, 440, 442, 443 dha Khanda of, 129, 161, 164 76, 137, 305, 327, 374, 393, 427, Mewar 436, 446 (see also elixirs, rasa-

584 585 Index Index

rasāyana (continued) Rāthaduņdā. See Rajasthan rogavāda (therapeutic alchemy), 52, 55, Sādhanamālā, 65, 81, 314, 371, 385 rasāyana); Ayurvedic elixir or reju- Ratnaghośa, 82, 84, 104, 124-25, 152, 130, 145-46, 161, 267, 271, 431 sabasrāra (the "thousand-petaled" venation therapy, 13, 25-26, 131, 164 Rohini: name of a star-maiden, 24, 44, cakra), 41, 242-43, 245, 293, 145, 168, 169, 186, 188, 266, 270, Rattannāth, 278-79, 400, 448 363; red cow, 36 489 271, 273, 303, 327, 342, 356, 371, Rāvaņa, 58, 82, 125, 150-51, 373, 374, Rosu, Arion, xiii, 370, 401 śailodaka. See waters, magical, Rock or 378, 479, 491; "coming forth of 415. See also Kumāra Tantra royal consecration, 304, 309, 322 Peak Water the essence," 73, 386; name of a Räwal: kingdom, 122; Nāth panth, 98, royal consumption. See rājayaksma Śaiva sects and sectarianism, 7, 9, 60, physician in the Harsacarita, 377; 121-22, 358, 399, 414. See also Řsyaśrnga, 203 80, 87, 90, 93, 96-102, 104, 111- Tibetan Buddhist inner alchemy, Bāppā Rāwal Rtu Kund. See waters, magical 12, 119, 171-83 passim, 193, 213, 71, 385; term for alchemy, 14, Ray, Prafulla Candra, 104-5, 152, 160, rtu-sātmya, 23 282, 284, 286, 305, 306, 309, 310, 188 164, 372, 401 Rudra-Siva. See Siva 321, 325, 330, 333, 340, 345, 347, Rasāyana Šāstroddbrti of Vyādi, 415 realgar. See red arsenic Rudrayāmala Tantra, 130, 145, 417, 357, 393, 398, 445, 496, 513, 515. rasendra, name of mercury, 188, 208; rebirth. See transmigration 428, 452. See also Dhātukalpa; See also Aghoris; Dasnāmis; Kāpāli- name of Śiva-Bhairava, 180 red: arsenic (manabsila), 170, 189, 195, Gandhakakalpa; Pāradakalpa; Rasa- kas; Pāśupatas; Šaiva Siddhānta Rasendra Bhāskara, 191, 210, 214, 408 205, 284, 326, 451, 453; color of kalpa Śaiva Siddhānta, 37, 60, 95-98, 102, Rasendra Mangala of Śrīman Nāgār- female elements, 192, 193, 194, 134, 146, 162, 172-73, 177, 182, juna, 76, 104-5, 123-27, 130, 198, 199, 202, 209; Matsyendra Śabara Tantra, 98 213, 397, 416, 424, 426, 439, 440, 145-46, 149-51, 154, 155, 158, as, 224, 396, 468 Śabaracintāmaņi of Nityanātha Siddha, 462. See also Mrgendrāgama 160, 161, 162, 164-67, 274, 325, religious alchemy, typologized, 52-57 129 šākta pīthas. See pītbas 327, 390, 401, 415, 428, 430, 435; resorption, 17, 19, 33, 185, 206, 208, Śābara-pā, 84 Śākta-Śaiva devotionalism. See wor- list of Rasa Siddhas in, 81-84 2II, 215, 219-23, 233, 263, 272, sacrifice: blood, 29, 89; Vedic, 10-11, ship, tantric Rasendracintāmaņi, of Dhuņdhukanā- 293, 441, 458, 463; time of (laya- 15-20, 23-24, 28, 36, 184, 189, Śakta sects and sectarianism, 2, 79, 99, tha, 130, 248 kāla), 316-19. See also emanation; 215, 245, 269-71, 281, 283, 466 101, 11O-I1, 140, 146, 171-83 pas- Rasendracūdāmaņi of Somadeva, 117, pralaya; withdrawal and return, (see also agnicayana); into mouth sim, 235, 260, 355, 357, 405. See 125-26, 128, 130, 146, 154, 157, yogic of kundalinī, 220, 234; rasa-agni- also Krama; Śrīvidyā; Trika Kaula; 158-60, 167, 168, 169, 187, 190, Ret Kund. See waters, magical vāyu as three elements of, 11-13, Western Transmission; Yoginī 191, 212, 274, 375, 394, 415, 429, Revana, 86-87, 92-93, 133, 168, 391, 21-26, 32, 45, 184-85, 187, 283, Kaula 434, 439, 487, 492 401 362, 445. See also diksā; internal- śakti (feminine energy), 35, 61, 177, Rasendrakalpadruma of Rāmakrsņa reversal, yogic, 29, 39, 41-42, 201-2, ization of the sacrifice 218, 222, 230, 243, 257, 312, 355, Bhatta, 130 227, 248, 262-64, 274, 290, 313, Sadāśiva, 152, 162, 206, 416, 438, 450, 367, 423; visarga-śakti, 256; in left Raseśvara: name of mercury, 187; 321, 508. See also ürddhvaretas 496, 501; and divinization of tatt- side of subtle body, 485 (see also name of Siva as god of alchemy, Rg Veda, 10, 17, 37, 211, 216, 242, 278, vas, 35; portrayed as mabāpreta in linga, western); rasa-śaktis, 442. 103, 157, 179, 258, 439; Raseśvara 478; earliest use of term yoga in, Kaula iconography, 176-78, 442. See also femininity; kundalinī Siddhas, 111, 390 19-20 See also five, faces; Saiva Siddhānta Sakti (name of the Goddess), 36, 43, Raseśvara Darśana. See Sarvadarśana- rice, 196, 289, 299, 351 sadbbāva. See clan nectar 92, 142, 202, 247, 251, 253, 278, samgraha Roberts, John, 468 sādbana (realization, practice), 2, 185, 289, 317, 368, 421, 427, 432, 439, Rasopanisat, 102, 146, 155, 157, 381, rodbana (coagulation of mercury, medi- 236, 244, 264, 265, 288, 335, 345; 442, 452, 473 410, 416, 431; authorship of, 127. tative mind), 267, 290, 291; Rod- bila-sādhana, 59; nidhi-sādhana, 59. Šaktisangama Tantra, 394, 421 See also Mabodadhi hinī, 291 See also reversal, yogic Śaktivijñāna of Somānanda, 501

586 587 Index Index

Śālivāhana; Salwān. See Šātavāhana, Śankara: Advaita Vedānta philosopher, king 30-31, 87; name of Siva, 163 Satyanāth, 87, 93, 98, 112, 392, 473 464, 465, 483, 487, 490. See also

śalyatantra (surgery), 55 Śankaravijaya of Ānandagiri, 112, 392, saumya. See year, solar erotico-mystical techniques

samādbi, tumulus of Nāth Siddha, 94, 408, 418 Schoterman, J. A., 481, 483 Shah Abdul Latif Bhitai, 311-12

288, 333, 338, 350, 396, 407, 450, śankinī (conch duct), 254-57, 322, 484, seeds. See bījas shamanism, 295-99 Sharma, Dharmananda, 390, 481 519; total yogic integration, 45, 485, 487, 489 semen. See sexual fluids

61, 215, 218, 221, 253, 265, 290, Sankrtyayana, Rahul, 404 Seneca, 189 Sharma, Hari Shankar, xiii, 117

369, 472, 483 Serpents, 58-60, 67, 210-17, 218, 240, Sharma, P. V., 105-6, 126-28, 414, 419 sannyāsa, 281, 497 samarasa (fluid equilibrium), 185, 201, Śantideva, 35 324, 327, 333, 350, 462, 463, 496, Shastri, Haraprasad, 105, 402, 416

218, 253, 445, 455, 483, 506 Śāradatilaka of Laksmaņadeśikendra, 501, 504; Nāgalatā ("Serpentina"), Shastri, Krishna Pal, 336-37

sāmbbavī, 481. See also Šambhu; mudrās 308, 322; Caraka, Patañjali, and Sialkot. See Jhelum 312 Sambhu: name of a Rasa Siddha, 82, Sarāha-pā, 68-70, 76, 399 Nāgārjuna as incarnations of, 215. "Siddh Vandanām," list of Siddhas in,

124, 126, 160; name of Siva, 248, sārana (flowing, potentialization of See also Ananta; lead; Śeșa; snake- 83, 85-86, 391

481. mercury), 268, 299, 300, 316 charming; Vāsuki. Siddha: defined by Tirumūlar, 61; ety-

sambāra (universal retraction), 227, Sarasvati River, 225-227, 468, 470, 480 Śeșa, 59, 215-16, 218-19, 284, 289, mology of term, 2

232-33, 240. See also pralaya śarira yoga. See transubstantiation 374, 463 Siddha alchemy, 52, 55-57, 77-78, 100,

Sāmkhya, 17, 20-21, 27, 32-36, 185, Śarngadhara Paddbati, 96, 131, 141, sexual fluids, 11, 13-14, 16, 28, 32, 109-22 passim; 141-44, 146, 162,

206-10, 213, 240, 270, 458, 510; 168-70, 188, 263-302 passim, 396 135-40, 146, 163, 172, 183, 184-

Samkbya Kārikas, 357. See also Śārngadhara Sambitā, 131, 396, 415, 202 passim, 210, 212, 216, 235, 314-18, 327, 372, 439

Prakrti; Purușa; tattvas 420, 437 245, 247, 252, 253, 258, 269, 286, Siddha Cakra, 89, 133, 135, 137, 181,

sampradāya. See Nath sampradāya Śārnganāth, 94, 396 294, 309, 312, 316, 438, 447, 450, 356, 386, 388, 444. See also yuga-

samputa (interlocked, encased), 145, Sartre, Jean-Paul, 511 452, 455, 460, 486, 489, 491; as nātbas

256-57, 268, 282, 294, 299, 454, Sarvadarsana Samgraba of Mādhavā- fuel of yogic transformations, 4-5, Siddha Kaula, 74, 88, 133-34, 171,

486, 509 cārya, 56, 102, 131, 187, 372, 400 27, 39, 41, 45, 72, 173, 207, 232- 423; Siddhāmrta Kaula, 133

samskāras: alchemical operations, 44, Śarvaja, 151, 157, 167, 430, 432 33, 237, 243, 250, 274, 285, 289, Siddha Khanda of Rasaratnākara. See

145, 147-48, 155, 158, 159, 166, Sarvajña. See Šarvaja 293, 320, 321, 351, 360, 363, 425, Rasaratnākara

169, 181, 198, 259, 260, 263-302 Sarveśvararasāyana of Vyādi, 383 430 (see also adhoretas, ürddhvare- Siddha maidens. See Siddhas

Sastri, Radhakrishna, 437 tas); as end products of digestion, Siddha medicine, 102, 208, 372, 457 passim, 314, 368, 427, 428, 442, 461, 492, 494; term kūrpa used for Šātavāhana, name of several kings, 67, 21, 25-26, 339-42, 360, 363; as Siddha Siddbanta Paddbati, attributed to

in Rasopanisat, 155; life-cycle rites, 69, 104, 114, 298, 381, 409, 504 source of kula continuity, 79, 137 Gorakh and Nityanatha, 129, 141

43-44, 264, 271, 491, 493. See also Satī, 195, 196, 451. See also pīthas (see also clan nectar); ritual con- 201, 260, 418, 460, 485, 488, 496

names of individual samskāras satkarmāņi (six purificatory practices of sumption of, 137, 424, 455, 483, Siddhācāryas. See Mahāsiddhas

Samundranāth, 350 hathayoga) 265, 271 509; used in treating mercury, siddba-deba. See Siddhas (Perfecti), in-

Sanderson, Alexis, xiii, 136, 389, 417, Satruñjaya (Mount), 114-15 153, 158, 433. See also cycles, men- destructible bodies of

424, 439, 444, 445, 465, 509 Satsābasrasambitā, 88, 91, 180, 234, strual; kundagolaka; power sub- siddhadravya. See power substances

Sangītaratnākara of Sārngadeva, 408 260, 385, 392, 488 stances; yonitattva Siddhanātha, 73, 88, 230, 386, 418,

Sangye Sonam, 483 sattva. See gunas sexual union, 4-5, 28, 31, 135-39, 157, 513

sanjivanījalam. See waters, magical sattvas. See minerals 158, 162, 194, 198, 199, 220, 234, siddhāñjana. See añjana 235, 252, 258, 323, 341, 441, 453, Siddhas (Perfecti): abode of, 3, 150,

588 589

Index Index

Siddhas (continued) Sikkim, 109 (see also Ardhanarīśvara); relation- Somnath. See Gujarat 238, 297, 303, 327-34, 373, 463, Silburn, Lillian, 235 ship to the moon (see moon); re- sorcery, tantric. See Kriyā Tantras 514; acceding to the world of, 3, Silk Road, 2, 62-63, 65, 203, 204, 378 vealer of Tantras, 54, 127, 142, soteriology, 72, 385 6, 52, 56, 244, 265, 303-334 pas- Simhala. See Sri Lanka 145, 161, 172, 223, 225, 229, 236, soul, individual. See ātman sim, 514; as mountains or moun- Simbāsanadvatrimśika, 115, 451 447, 468; Siddhas in retinue of, soul, universal. See brabman tain gods, 327-34; cult of, 2-3, Sind, 55, 59, 98, 116, 204, 295, 311, 331, 512; Siddheśvara (temples sound, body of, 72; substrate of, 291, 135, 157, 180-81, 292, 329, 335, 414, 457, 491, 505. See also Indus of), 60, 95-96, 103, 110, 367, 375, 293, 477 388, 391, 441 (see also Siddha River; Pakistan 398; Tripurāntaka, 438. See also south India, 49, 53, 55, 61, 67-68, 77, Cakra); divine and semidivine, 2, sindūra (red mercuric oxide, vermil- Bhairava; Kālāgnirudra; linga; and 80, 105-6, 11O-I1, 124, 126, 131, 57-58, 124, 135-36, 157, 161, lion), 194, 196, 284, 351, 455 names of deities identified with Siva 168, 169, 307, 372, 376, 379, 382, 180-81, 323-34 passim, 355, 421, Singhaņa, Yādava king, 112, 114, 127- Śiva Sambitā, 324, 325 393, 431. See also Andhra Pradesh; 512, 513; formation of canonical 28, 167 Śivahood, 4, 6, 42, 52-53, 143, 157, Kadri; Karnataka; Kerala; Kon- lists of, 74-75; 80-101; four kula- Sircar, D. C., 412 163, 166, 173, 181, 188, 220, 228; kaņa; Malabar coast; Sittars; Srisai- siddhas, 324; indestructible bodies Śitalā (goddess of smallpox), 29 234, 240, 312-25 passim, 331, lam; Tamil Nadu of, 191, 202, 271, 281, 315, 319; Sittars, 2, 58, 61, 65, 76, 91, 99, 105, 322, 432, 439, 465 Southeast Asia, exchanges with India, Siddha maidens, 150, 192, 323, III, 125, 373, 376, 394, 400-2, Śivakalpadruma of Sivanātha, 130, 448 48, 55, 61, 64, 103, 369, 384, 389, 325, 326, 512; Siddhavidyās, 58; 415, 450; lists of, 87 six: metals, 189, 190; months, 170, 401, 457, 487 three primal, 73, 91, 230-31, 329, Siva, 14, 92-93, 101, 115, 134, 152, 294; rasas, 185; schools of Indian Southern Transmission. See Śrīvidyā 330, 331, 514. See also Eighty-four 188, 190, 196, 202, 210, 216, 220, philosophy, 174 Spanda Kārika of Vasugupta, 222 Siddhas; Mahāsiddhas; Māheśvara 221, 232, 261, 262, 264, 284, 286- sixteen, 17, 36-44, 365, 368, 392; ād- sparśamani. See transmutation Siddhas; Nāth Siddhas; oghas; 89, 296, 297, 298, 307, 312, 318, bāras, 366; expansion to eighteen, spitting. See initiation Rasa Siddhas; Sittars; sixteen 329, 332, 393, 415, 430, 438, 439, 43-44, 265, 491; Mondays vow, Śrī Cakra or Śrī Yantra. See mandalas Siddhesvara. See Siva 441, 460, 463, 471, 477, 487, 495, 37, 366; number of women in Sri Lanka, 55, 60-61, 119, 126, 236, siddbis (supernatural powers), xii, 2, 5, 499, 515; becoming a second (see Plantain Kingdom, 475; samsk- 260, 262, 299, 415, 489 58, 60, 97, 102, III, 114, 132, Sivahood); cult of, 4, 9, 98, 143, āras, 43-44, 265-69; Siddhas, 85, Śrī Nath Tīrthāvalī of Mān Singh, 496, 137, 145-46, 153, 157, 161, 163, 163, 173-75, 197, 454; earliest im- 91, 169 500 170, 174, 176, 180, 186, 202, 222, perial devotees the Kushanas, 398; Skanda, 131, 159, 189-91, 224, 329, Śrīkantha, 73, 97-98, 126, 128-29, 242, 262, 263, 265, 274, 284, 294, identified with guru in initiation, 359, 376, 440, 441, 447, 448, 472 131, 133, 158, 230, 250, 416, 418, 301, 303, 306, 307, 313, 315, 316, 271; location of in subtle body, 4, snake charming, 99, 222, 350, 399, 519 442, 465; revealer of Bhūtiprakar- 317, 319, 321, 325, 326, 329, 330, 38, 221, 241, 243, 245, 246, 293, Sodhala, 106, 114. See also Gadani- ana, 140, 142, 156, 157, 158 331, 401, 424, 425, 429, 439, 461, 320, 462; mantra-bīja identified grab Śrīmatottara Tantra, 156, 157, 386, 496, 510, 513; called vibbūti, 284, with semen of, 271; Pine Forest sodbana. See purification 425, 433 286, 432, 441, 499; early Buddhist myth of, 285; relationship to soma: as herbal medicine, 365, 366; as Śrīnātha, 73-74, 91, 129, 135, 230, lists of, 371, 507; subject of Vama Brahmā, Visnu, 46, 141, 146, 159, moon god, 11, 16, 184, 359 (see 255, 330, 418 Tantras, 419. See also diamond 162, 294, 315, 317; relationship to also moon); as sacrificial oblation, Śrīparvata. See Srisailam body; flight; kbānepāne; parakāya- the Goddess, 5, 35-36, 43, 54, 24, 36, 365, 455, 478; as Vedic Srisailam, 49, 53, 60, 68-69, 76-77, praveśa; rasa-rasāyana 128, 143, 145, 158, 173, 176, 185, draft of immortality, 10-11, 25, 92, 103-4, 110-12, 136, 152, 161- Siddhiyoga of Vrnda, 430 195, 211, 214, 224, 235, 247, 251, 28, 189, 232, 242, 243, 259, 285; 62, 164-65, 168, 196, 238, 245, Siegel, Lee, 496 252, 253, 256, 260, 368, 421, 427 theft of, 190, 462. 260, 331, 375, 398, 405, 408, 409,

590 591 Index Index

Srisailam (continued) flowers of (puspita gandhaka), 199, Swami Sevananda, 517 329, 331, 395, 424. See also Abhina- 422, 434; site of Mallikārjuna 440; origin myth of, 148, 168, swooned, killed, and bound mercury, 6, vagupta jyotirlinga, 111, 162, 398, 405, 192-93, 221, 234, 325; springs (see 144, 146, 149, 158, 159, 162, 274, Tantrarāja Tantra, 91, 394, 514 407, 480, 488 geothermal phenomena; waters, 384. See also bandbana; mārana; tantric alchemy, 51-55, 72, 78, 104, Śrīvidyā Kaula, 79, 139, 178, 181, 357, magical). See also yoni mercury; mūrccbana 109-22 passim, 145, 171-83 pas- 391. See also Bālātripurasundarī sun, 21, 23-25, 28-29, 38-41, 185, syncretism, Hindu-Buddhist, 53, 55, sim, 187, 372; "tantric element" srsti (cosmic emission), 47, 251, 263, 202, 205, 226, 232, 243, 244, 246, 57-59, 61, 66, 70-71, 74, 78, 94, in, 139, 141-44 463. See also emanation; with- 248, 250-54, 280, 282, 296, 301, 104-I1, 118, 136, 148, 152, 187, Tao te ching, 62, 377 drawal and return, yogic stabilization and stability: of breaths, 315, 318, 323, 373, 445, 455, 456, 196, 206, 361, 373, 375, 383, 390, Taoism and Hinduism. See syncretism, 457, 468, 481, 482, 504, 514; god 513 (see also Buddhist alchemy; Vaj- Hindu-Taoist 45, 232, 278; of body, 137, 174, of generates copper from seed, rayāna Buddhism; Highest Yoga Taoist alchemy. See Chinese alchemy 264, 265, 273, 274, 294, 295, 439; 190. See also moon Tantra); Hindu-Jain, 57, 80, 93, tapas (ascetic ardor), 18, 27, 39, 184, of mercury, 202, 258, 266, 277, Sundarnāth Siddha, 339 243, 270, 281, 284, 295. See also šūnya (void), 43, 106, 209, 211, 241-42, 99, 114-19, 331-34, 399, 410, 498. See also bandbana 475, 479, 498, 515 (see also Jain- thermal energy stars, 24, 315, 323. See also Rohiņī 253, 295, 403, 462, 477,482, 511; ism, Siddha and tantric traditions Tārā, 64-65, 69-70, 110, 378, 380, Stein, Sir Aurel, 309 mabāšūnya, 456; trišūnya, 178, of; Neminäth/Nimnāth; Pāras- 387, 465, 516; Tarā Tantra, 394, sthairya. See stabilization and stability 302, 506 nāth/Parśvanāth); Hindu-Muslim, 412 sthula sarira. See gross body Sūnyasampādane of Gūļūra Siddhavīra- 80, 99, 106, 121, 171, 205, 228, Tāranātha, 70, 107-8, 383, 403 subtle body, 4-5, 19, 138, 185, 207, ņārayu, 406 238, 311, 399, 403-5 (see also tattvas, 33-36, 43, 210, 214, 317, 365, 218-62 passim, 280, 288, 291, Sūrasena, 82, 104, 125, 433 Bāuls; Jogis, Muslim; Rāwals); 477, 510. See also five, elements; 301, 321, 327, 328, 333, 385, 414, Śūrpanāth, 86 Hindu-Sikh, 126; Hindu-Taoist, Prakrti; Purușa 422, 440, 450, 470, 480, 482, 485, Suśruta Sambitā, 52, 76, 210, 356, 62, 71, 196 Tāyumānavar, 340 506; lunar nature of, 18, 30-31; sa- 370, 373, 375, 387, 415; dating of, Telugu literature, 92-93, 111-12, 369, cred geography of, 112, 218-40. 361 Tabqīqāt-i-Cisti, 395, 501 429. See also Navanātha Caritra of See also cakras; cranial vault; left susumņā. See medial channel Taishang suling dayou miaojing, 250 Gauraņa; Śūnyasampādane side of body; linga, western; lower sūta (name for mercury), 188, 197, Taittirīya Sambitā, 24, 359 tenth door, 252, 254, 256, 260, 261, abdomen; matsyodara; Meru; nādīs; 208, 265, 446, 490 Tamil Nadu, 2, 60-61, 63-64, 77, 87, 483-85, 489, 504. See also nine nine doors; pīthas; tenth door; Svacchanda Bhairava, 83, 126, 128-29, 97, 106, 124, 178, 340, 357, 376, doors trikuta 142-43, 156-57, 175, 176, 178- 378, 392, 398, 407, 415, 447, 450, thermal energy, 13, 39, 284. See also Sufism, 109, 405 179, 417, 424, 432, 434, 442, 514; 504 geothermal phenomena; tapas sūksma śarīra. See subtle body svacchandabbairava rasa, 160, 417, Tanjur, 70-71, 81, 105, 108, 125, 224, Tibet, exchanges with India, 2, 55, 64- sulfur, 6, 11, 16, 50, 143, 146, 154, 436. See also Bhairava 380, 383, 386, 387, 390, 403, 416, 66, 69, 80, 105, 196, 247, 251, 157, 170, 174, 187, 189, 190, 192, Svaccbanda Tantra, 129, 156, 175-176, 425 379, 380, 384 193, 197, 198, 199, 202, 207, 248, 178, 329-31, 417, 439, 442, 514 Tanti-pā, in Siddha lists, 81, 83-84, 86, Tibetan literature, 64, 67-71, 80, 218, 252, 258, 268, 291, 292, 293, 352, Svacchandasaktyavatāra. See Bhūtipra- 91-92 447. See also bCud len gyi man 370, 377, 381, 409, 441, 449, 452, karaņa tantra, etymology of, I-2 ngag bsbad pa of Bo dong; Chos 453, 498, 501; as Goddess's uter- svedana (sweating, trituration of mer- Tantra Mabārnava, 91, 113, 394, 408, bbyung of Bu-ston; Dragpa jets'en; ine blood, 5, 191, 193, 194; ety- cury), 266, 273 425 Grub thob; Kanjur, Tanjur; Tāra- mology of term gandbaka, 193; swallowing. See grāsa Tantrāloka, 73, 88, 137, 225, 312-14, nātha

592 593 Index Index

time, 23, 28, 42; units of, 46, 317. See twelve: Näth panths, 90, 93-100, 110, uttarāyana. See year, solar Valabhī[-śaila], 115-16, 119, 300, 409- also death; kalpa; mātra; pala; ma- 113, 119, 121, 129, 398-99; years, utthapana (resurrection of swooned 10, 505 bāyuga xii, 239, 244, 288, 294, 295, 299, mercury), 267, 275, 290 Vāmakeśvara Tantra, 163 Tintini. See Taņți-pā 300, 345 Vāmamārga, 256 Tirumūlar, 61, 76, 87, 375, 412. See Vacaspati Miśra, 58 vampires (siddbi of control over), 305, also Mūlanātha Uddiyāna, Buddhist pītba, 260, 495. Vādava, 472; birthplace of Gorakh, 306, 307, 309, 431, 442, 476; vam- Tod, Colonel James, 122, 344, 345, See also bandbana and bandhas; 422, 472; name of a Rasa Siddha, piresses (women and demonesses 412-14 pīthas 82; vadavāmukba-bīda, 449; vadavā- characterized as), 341-42, 441, Țodala Tantra, 164 Ujjain, 49, 62, 64, 311, 335 mukba-rasa, 436. See also bādava 473 transmigration, 18, 29-30, 38, 185, ulatā sādbana. See reversal, yogic Vägbhatta, the Elder, 52, 65, 379, 390, Van der Veer, Peter, 345 201, 239, 240, 326, 367 Umā, 126, 128, 140, 157, 177, 197, 415 (see also Astānga Samgraba); Van Kooij, Karel R., 295 transmutation, 7, 11, 14, 53-55, 68, 44I Vāgbhatta the Younger, 390 (see Varaņā River. See Benares III, 122, 137, 147, 149-50, 160, unicorn, 203, 205 also Astānga Hrdaya); Văgbhațta Varņaratnākara of Kavišekarācārya Jy- 165, 166, 177, 199, 202, 244, 245, unmanā (beyond mind), 45, 178, 302 II, 82, 127, 167, 411 (see also otīśvara, 80, 391, 393, 395; list of 246, 258, 259, 265-68, 292, 303, Upadhyaya, Nagendranath, 404 Rasaratnasamucchaya) Siddhas in, 84 311, 314, 320, 322, 330, 336-39, Upadhyaya, Yadunandan, 336 vaidyas (Ayurvedic physicians), 13, 22, varnas (castes or colors of minerals), 352, 371, 375, 407, 452, 506, 5I1. Upanişads: Aitareya, 511; Brhadāran- 113, 336, 359, 421 100, 154, 192, 198, 208-9, 214, See also dhātuvāda; paras patthar; yaka, 17, 213; Chāndogya, 27, 30, Vaikbānasa Smārta Sūtra, 491 431 vedba 207, 367, 459; Kathaka, 206, 212, Vairāg panth, 93, 97, 396, 399 Vāsavadattā of Subandhu, 60, 64, 110, transmutational alchemy. See dhātu- 241; Maitrī, 20, 28, 185, 253; One Hundred Eight, 131, 420; Prāņā- Vaisnava sects and sectarianism, 99, 405 vāda 159, 357 Vāsuki, 114, 190, 214, 221, 289, 447, transubstantiation, 7, 11, 14, 53-55, gnibotra, 497; Praśna, 20; Tait- vājīkarana (sexual rehabilitation ther- 464, 505 159, 166, 258, 265, 267, 268, 282, tirīya, 32; Yogadarśana, 469; Yoga- apy), 25-26, 266, 518. See also aph- vayu (wind). See sacrifice, Vedic 314, 315, 324, 375, 491 kundalī, 169, 313; Yogašīkha, 131, rodisiacs Vāyutattvam Bhāvanopadeśa of Ghora- Trika Kaula, 4, 36, 73, 79, 89, 135-38, 420; Yogatattva, 371 Vajrasattva, 70, 365, 373, 382, 394, 449 kșa, 404 172, 176, 178, 214, 255, 314, 389, uparasas (secondary alchemical re- Vajrayāna Buddhism, 35, 65, 66, 70- Vedānta, 18, 20, 32-36, 189, 206, 222, 424, 427, 441, 464, 465, 477. See agents), 145, 147-48, 154, 155, 72, 80, IO1, 126, 171, 196, 201, 357,426 also Abhinavagupta; Parā-Aparā- 159, 160, 164, 166, 168, 187, 384, 221, 234, 247, 251-52, 260, 379, vedba[na] ("piercing," transmutation of Parāparā; Siddha Cakra 431, 446, 449, 461 381, 382, 384, 385, 389, 397, 399, metals by mercury), 265, 268, trikuta (triple-peak configuration lo- ūrddhvapātana yantra. See pātana 454, 455, 483, 509; Highest Yoga cated in cranial vault), 471, 506 ürddbvaretas, 233, 240, 248, 251, 284, Tantra, 136, 386, 390, 404; Ti- 303, 313-15, 321, 490, 507; bindu- vedba, 312; five types, 159, 209, Tripathi, Bhagiratha Prasada, 371, 433 341, 483. See also adhoretas betan Nyingma school, 65, 71, 459; kotavedbi-rasa, 409; linguistic Trīpurasundarī. See Bālātripurasundarī urethral suction. See vajroli mudrā 105, 221, 379, 385. See also Bud- alternation with bedhana/bhedana, Tristhalīsetu of Nārāyana Bhatta, 469 uterine blood. See sexual fluids dhist divinities; Mahasiddhas 313; mabāvedba (yogic practice), Trita, 465 Uttar Pradesh, 98, 112, 121, 245, 310, vajrolī mudrā, 198-201, 253, 276, 282, 282, 303, 319, 511; rasavedha, 259; Trivikrama. See Vikramāditya 346-47, 392, 406, 414, 435. See Tryambaka Pant, 396 322, 341, 454, 455, 473, 482, 483, rasaviddbi, 507; śabda-vedha, 209, also Ayodhya; Benares; Gorak- 498; in Mahāyāna sources, 63, 212, 320, 511; vedba[mayī dīksā], Tucci, Giuseppe, 80, 312, 470 hpur; Hardwar 201; parallel Chinese technique, 182, 312-14, 321, 368, 508, 509. Tvastr, 189 Uttarakurus, Land of, 355 63, 72 See also piercing

595 594 Index Index

Vetālapańcavimśati, 306, 309. 162, 224, 227, 289, 294, 296, 315, Weber, Max, 385, 387 worship, of chemicals. See rasacikitsā

vibhūti. See siddbis 317, 318, 327, 347, 415, 451, 495; weights and measures, 510. See also worship, tantric, 4, 54, 79, 89, 131, Vidyā Pītha, 79, 135-37, 156, 173, identified with Dattätreya in Ma- pala 135-40, 143-45, 150, 152, 157,

176-77, 386, 389, 404, 416, 424, harashtra, 92, 411; Nārāyana, the Weikzas. See Zawgyis and Weikzas 163, 167, 168, 169, 172, 173, 176,

441, 444 cosmic yogin, 215, 221 (see also Weinberger, Steven, 404 179-81, 241, 258, 284, 306, 324, Vidyādharas (Wizards), 3, 52, 56-58, pralaya) Weinberger-Thomas, Catherine, 517 355, 391, 424, 430, 440, 441, 442,

74-75, 99, 118-19, 124, 247, 259, Viśva Hindu Parisad (Hindu World Wellcome Institute for the History of 443, 450, 454, 473, 488. See also

292, 307, 323-34 passim, 355, Council), 347, 519 Medicine, 435 erotico-mystical practice; pañca-

373, 415, 512, 514, 516; Jain tradi- Viśvāmitra, 304-7 wells, 237, 239, 240-53 passim, z80, makāra; yoni-pūjā tions of, 387, 516; rig 'dzin in Ti- Viśvanāth, 226, 398 288, 289, 294, 297, 344, 449, 478- written transmission, xii, 145, 161. See

betan traditions, 387; three levels Viśvarūpa, 189, 315 79, 505; dedicated to Goddess, also oral transmission

of, 329-30; vidyādhara yantra, Vivekamārtanda of Gorakhnāth, 141, 245; inverted, 242-44; jīrnāndha- Wujastyk, Dominik, xiii, 104, 115,

248-51, 480, 482, 490, 494; Vidy- 169 kūpa, 244; of mercury, 115, 117, 410, 435

ādharīs, 192, 325; Vidyārājas in Vividbatīrtbakalpa of Jīnaprabhāsuri, 159, 162, 168, 191, 202, 205, 208, Buddhist traditions, 373, 387, 516. 117, 479 212; Pūran's Well, 245, 299-300, Yadavas of Devagiri, 94, 104-6, 112- Vidyeśvaras, 168, 177-78, 213, 416, volcanos. See geothermal phenomena 478; romakūpa (pores), 273, 495; 14, 127, 147, 167, 393, 408-9. See

442, 462 Vrnda, 76, 382, 386. See also Sid- water pots represented as, 242-43, also Devagiri; Singhaņa

Vijayanagara (Hampi), 102, 383, 396, dbiyoga 478; Well of Truth, 457 Yadavji Trikamji Acarya, 117

406, 407, 419 Vyādi, 71, 74, 81-82, 86, 91, 125-26, West, Indian exchanges with, 2, 87, Yajñavalkya Smrti, 37, 494 Vikramāditya, name of several kings, 199, 311, 383, 390, 416 203-6, 376, 456 Yajur Veda, 189, 499

115, 236, 306-7, 311, 313 vyāpana. See emanation western India, 110-22, 422, 476 Yakşas (Dryads), 58-59; Yaksiņīs, 104, Vimalananda, 201, 341, 454 Vyāsa, 57-58, 422 Western Transmission, 2, 4-5, 73, 79, 153, 475 Vimalaprabbā. See Kālacakra Tantra vyoma (term for both mica and ether), 87-90, 94, 112-13, 126, 129, 133- Yama, 11, 23, 46, 171, 259, 280, 295, vimarśa. See light 210-12, 241, 459-60, 477 35, 139-40, 143, 151, 180-81, 340, 409, 429, 475, 497; iron Vīnāšikba Tantra, 432, 472 228, 230, 255, 389, 423, 481. See arises from seed of, 190 Vināyaka. See Gaņeśa Wakan [Ho ban] sans ts'ai t'ou boei, 203 also Goraksa Sambitā; Kubjikā; Yamāntaka, 220-21. See also Bhairava Vindhyas, 157, 381 Walleser, Max, 67 Kubjikāmata; Manthānabbairava Yamuna River, 225-26, 277, 480

viparīta, reverse sexual position, 308, waters, magical, 113, 205, 238, 245, Tantra; Satsābasra Sambitā yantras (alchemical apparatus), 78, 145, 313; viparītakaraņī (yogic posture), 261, 409, 446; Amrt Kuņd, 246; Wilkinson, Chris, 386 160, 169, 181, 183, 241, 248, 250-

249, 482, 490 Cleaver Water (kartarī jalam), wine, ritual use of, 137, 220, 231, 234, 51, 257-58, 327, 333, 426, 480; bi- vīra, 239, 323. See also pasu 384;resuscitating water (sanjivanīja- 252, 455. See also pañcamakāra cameral configuration, 247-52, Vīrabhadra, 286, 417 lam), 113; Ret Kund, 246; Rock or Wink, Andre, 412 256, 328, 479. See also damaru; pā- Vīramābeśvarāgama, 11I Peak Water (śailodaka), 153, 316, withdrawal and return, yogic, 33, 38- tana; Vidyādharas Vīraśaivas, 46, 97, 102-3, III-12, 129, 355, 409, 510; ntu Kund, 245. See 39, 42, 46-47, 185, 221, 251, 263, Yasodhara [Bhatta], 82, 159. See also

237-38, 314, 375, 391, 393, 406, also Candraküpa; geothermal phe- 290, 293, 318, 463 Rasaprakāśa Sudbākara

477 nomena. Work in Two Parts, 14, 57, 173, 263- year: solar, 23-25, 28, 362, 469; semes- Virūva-pā (Virūpā), 252 Wattelier, Alain, 478 65, 266, 274, 275, 277, 281, 294, ters of, 22-26, 30, 41, 46, 318, Visnu, 46, 59, 111, 124, 141, 146, 159, Wayman, Alex, 382 315, 372 363; seasons of, 23-24, 29, 362

596 Index

yoga: ati-yoga in Åyurveda, 23; a-yoga in Yoginīs, 4, 134-35, 138, 152-53, 156, Āyurveda, 23; sama-yoga in Åy- 157, 159, 168, 169, 173, 176-77, urveda, 20, 23; yogāgni (see fire); 180, 182, 199-200, 223, 229, 235, yoga-ksema in Rg Veda, 19-20. See 261, 323, 324, 334, 442, 466, 512, also Nath Siddhas; Pāsupatas; Yoga 517; Buddhist Kāpālika-Yoginī Sūtras cult, 136, 386, 390, 404; sixty- Yoga Sūtras of Patañjali, 57-58, 373. four, 421, 424, 483; yoginī-bbu (yo- 422, 494; classic definition of yoga gic child), 200. See also Dombi; in, 273 Dutī; mouth Yoga Vasistha, 239, 439 Yogis. See Jogis Yogabija of Gorakhnāth, 100, 141, 255, Yogisampradayāviskrti, attributed to 425, 512, 513 Jñāneśvara, 113, 134, 236, 391, Yogadīpikā of Gorakhnāth, 141, 407 408, 411, 466, 474, 500, 504 yogāgni. See fire yoni, 194-96, 234, 247, 269, 296, 485; Yogamartanda of Gorakhnāth, 141, yonicakra (fire-pit), 258; sulfur 169, 482 called umā-yoni in Bhūtiprakarana, Yogananda, 338 157, 197, 432; upper and lower in Yogaratnamālā of Nāgārjuna, 76, 112, subtle body, 256, 443, 467, 482 114, 125, 131, 165, 408, 419, 430 (see also mouth, Yogini's upper and Yogaratnāvalī of Šrīkantha Šiva, 131, lower); yoni-puja, 195, 473; yoni- 465 tatrva, 454, 455 Yogašataka of Nāgārjuna, 75-76, 387, yuganāthas, 135, 137, 181, 416, 513. 388 See also Siddha Cakra Yogavisaya of Matsyendra, 484, 485 Yogi, Surya Kumar, xii, 401, 479, 517 Yogīndras, 440 Zawgyis and Weikzas (Burmese alche- Yoginī Kaula, 5, 73, 88-89, 134-36, mists), 48, 124, 354, 384 172, 200, 223, 229, 235, 261, 423, Zimmerman, Francis, 22 466, 471 Zulmat, 238