1. KashmiriShaiviteLogicIsvarapratyabhijnakarikaOfUtpaladevaWithTheAuthorSVrttiAlon_Kashmiri-Shaivite-Logic-Isvarapratyabhijnakarika-of-Utpaladeva-With-the-Author-s-Vrtti-along-with-English-translation

The

Īśvarapratyabhijñākārikā

of

Utpaladeva

with

the

Author's

Vṛtti

Critical

edition

and

annotated

translation

RAFFAELE TORELLA

To Raniero Gnoli

The

Īśvarapratyabhijñākārikā

of

Utpaladeva

with

the

Author's

Vṛtti

Critical

edition

and

annotated

translation

Raffaele

Torella

MOTILAL

BANARSIDASS

PUBLISHERS

PRIVATE

LIMITED

•

DELHI

Corrected Edition: Delhi, 2002

First Edition: Roma, 1994

© 1994 ISTITUTO ITALIANO PER IL MEDIO ED ESTREMO ORIENTE

All Rights Reserved

ISBN: 81-208-1867-9

Also available at:

MOTILAL BANARSIDASS

41 U.A. Bungalow Road, Jawahar Nagar, Delhi 110 007

8 Mahalaxmi Chamber, 22 Bhulabhai Desai Road, Mumbai 400 026

236, 9th Main III Block, Jayanagar, Bangalore 560 011

120 Royapettah High Road, Mylapore, Chennai 600 004

Sanas Plaza, 1302 Baji Rao Road, Pune 411 002

8 Camac Street, Kolkata 700 017

Ashok Rajpath, Patna 800 004

Chowk, Varanasi 221 001

FOR SALE IN SOUTH ASIA ONLY

Printed in India

BY JAINENDRA PRAKASH JAIN AT SHRI JAINENDRA PRESS,

A-45 NARAINA, PHASE-I, NEW DELHI 110 028

AND PUBLISHED BY NARENDRA PRAKASH JAIN FOR

MOTILAL BANARSIDASS PUBLISHERS PRIVATE LIMITED,

BUNGALOW ROAD, DELHI 110 007

CONTENTS

Preface ............................................................................... VII

Acknowledgements ............................................................... VIII

Introduction ........................................................................ IX

About this Edition ................................................................ XLVI

Text ....................................................................................... 1

Translation ............................................................................ 83

Abbreviations ........................................................................ 221

Bibliography .......................................................................... 223

Index of important words in the Text ..................................... 235

Index ...................................................................................... 257

v

PREFACE

This second Edition differs only slightly from the first one (IsMEO, Rome 1994). Basically, I have limited myself to correcting a not negligible number of misprints, kindly pointed out by colleagues and students. Among them, I wish to thank first of all J. Bronkhorst, then M. Hattori, F. Sferra, A. Pelissero and C. Pecchia. Heartfelt thanks I also owe to Malgosia Sachha.

I take the opportunity to express my gratitude to the many reviewers of this book. Some of them (J. Bronkhorst, J. W. de Jong and M. Hattori) have proposed alternative translations of some passages. In order to illustrate the grounds of my translation of I.5.18 and I.5.21 (and vṛtti), I wrote a detailed letter to the now greatly lamented Professor de Jong, who, in his reply, declared himself fully convinced. The remarks of Professor Hattori concerned basically the same points; he also corrected and integrated the last part of fn.20, p.117 (see below).

Several objections were raised by my colleague and friend Professor Johannes Bronkhorst in his thorough review article published in Asiatische Studie / Études Asiatiques (L, 3, 1996, pp. 603-621), which have induced me to re-examine carefully some thorny points of this not easy work. I do agree that his interpretation of alokādi (vṛtti on I.7.10) is better - and plainer - than mine. It is to be said, however, that this does not affect the overall meaning of the passage concerned: "The light etc. [perceived in a certain spot] means the absence of something else [in the same spot]..." (Bronkhorst), versus: "The sight [of a light] etc. ..." (mine). After having attentively checked the other points questioned by my learned friend, I remain convinced of the correctness of my own translation.

In his capacity of Director of the Editorial Committee of IsIAO (ex IsMEO), Professor Maurizio Taddei took a significant part in making the re-edition of this book possible. To our greatest sorrow, this outstanding scholar, and dear friend, came to a sudden and untimely end in February 2000. I wish to dedicate this Indian edition to his memory.

Rocca Priora, October 2001 RAFFAELE TORELLA

VII

ACKNOWLEDGEMENTS

First and foremost, I should like to express my gratitude to IsMEO for accepting this book for publication in the Serie Orientale Roma, founded by Giuseppe Tucci. In paying tribute to the memory of Prof. Tucci - my paramaguru - I must also acknowledge him as the first western scholar to realize the importance of the Kashmiri Śaiva tradition and to promote its scientific study. I ideally place this work at His feet.

I thank IsMEO and the University of Rome "La Sapienza", whose financial support enabled me to make several journeys to India in search of MSS of Utpaladeva's works.

Then, I express my gratefulness to the Libraries that allowed me to photograph and use their MSS for this edition: Research and Publications Department, Srinagar (A.K. Reina, Assistant Director), Oriental Research Institute and MSS Library, Trivandrum (Dr. K. Appukuntan Nair and Dr. T. Bhaskaran, former Directors; P.L. Shaji, MSS Assistant) and Oriental Institute Library, Baroda (Dr. R.T. Vyas, Director). I heartily thank Pandit Dinanath Shastri, Srinagar, who generously put at my disposal his private collection of MSS, then destroyed during Muslim desorders in Srinagar.

Special thanks are due to Felicity B. Lutz for her valuable assistance in the English version of this book, and to my student Francesco Sferra for preparing the Indexes and helping me very much in correcting the proofs.

Finally, I wish to thank Raniero Gnoli with whom I discussed many passages of this work. To this extraordinary scholar - and gentleman - I dedicate this book.

Rocca Priora, October 1993.

Raffaele Torella

VIII

INTRODUCTION

In the complex and varied cultural panorama of ninth-century Kashmir we find all the major components of the religious-philosophical tendencies in India at the time and, together with these, some particular situations that led to new developments, which were also later to spread very far from their place of origin. The tradition of Buddhism (rooted in Kashmir since very ancient times) is present both with the realist schools and the Vijñānavāda, and culminates in the great cultural prestige of the so-called 'logical school', which seems to elude too definite a collocation within one or other of the great Buddhist currents to which, by turns, contemporary followers or opponents and modern scholars assign it. The Brahmanic élites were still engaged in elaborating or perfecting their reply to the doctrines of Dignāga and Dharmakīrti, whose circulation in Kashmir was to be further stimulated by the presence of Dharmottara, summoned by King Jayāpīḍa to his court (Rājatarangiṇī IV.5.498). Thus works of great importance were composed that had a major influence throughout India, such as the Nyāyamañjarī of Jayanta Bhaṭṭa and the Nyāyabhūṣaṇa of Bhāsarvajña. Then, still in the Brahmanic sphere, though pertaining to another side, there was the spread of Vedāntic schools apparently independent of Śaṅkara, each of them being distinguished by a particular configuration of the clements common to them all (Brahman, māyā, avidyā, vivarta, etc.) and indicated by a variety of denominations1. There are various evidences of the activity of Mīmāṃsaka schools. In addition to this, a tradition of studies in literary criticism and acsthetics flourished during this period (Ingalls 1990: 1-10), and was to constitute the major cultural branch of Indian aesthetics. Lastly, there was an equally important tradition of studies in grammar and general linguistics, in which the legacy of Bhartṛhari seems to have a prominent place.

Corresponding to this 'high' cultural tradition on the religious plane, there was - obviously with the exception of the Buddhists - the smārta

1 Cf. citrabrahma vādin (ŚD VI.3b), sphuliṅgātmavādin (ibid. 10c) etc.

IX

dimension of Brahmanic orthodoxy, characterized by the ancient popular

devotion to Śiva and Viṣṇu, and to the pantheon connected with them. A

series of texts, linked to the cult of Śiva and Viṣṇu, were already becom-

ing stratified probably a couple of centuries before the period under con-

sideration.(but some are definitely much older). These express ideals and

thrusts that are absent in the Brahmanic sphere, mainly since they have

been consciously excluded because of their non-conformity to existing

social and religious structures, or even their possible disruptive force.

These texts give voice to a need for a more direct participation in the

experience of the divine, no longer seen as a transaction managed by

specialized personnel with a view to obtaining clearly defined and cir-

cumscribed benefits - culminating in an impersonal liberation projected

into an indefinite future - but as a transfiguration here and now of the

whole person, whose components, including the purely physical, become

the very protagonists of the path of liberation and not the unwelcome

extras. In this framework, and by carefully observing the nature of these

texts and of their addressees, it becomes immediately clear that we

cannot propose a scheme of interpretation centring on the polarities

Brahmanic orthodoxy - popular thrusts, or ritualism - ecstatic anti-ritualism.

In general, the Tantric adept, as Sanderson (1988: 662) has correctly

described him, is in actual fact a hyper-ritualist, who does not cancel the

inherited ritual but replaces it, often only by modifying, enriching and

making it more complicated2. He sometimes goes as far as to transform

it into a labyrinthine religious play, the highest and most complex ex-

ample of which is perhaps the Śaivasiddhānta ritual, analyzed in such a

masterly manner by Hélène Brunner (see Bibliography). The ritual elem-

ent rests on a basis that is broadly shared both by Vaiṣṇava and Śaiva

Tantrism, even in the currents that are ideologically more far-removed

from each other.

In ninth-century Kashmir the Vaiṣṇava school of Pāñcarātra and the

various currents of Tantric Śaivism had begun to follow different paths.

The Pāñcarātra was already heading towards an inexorable decline which

was gradually to lead to its disappearance from Kashmir and its firm

transplantation in the South of India. However, this was not without

producing as its final fruits some original figures of thinkers poised bet-

ween the Vaiṣṇava and Śaiva faith, such as Vāmanadatta (cf. Torella

1994. and Utpalavaiṣṇava, aware of the affinity of the two tradi-

2 In any case, never totally negating it (cf. Sanderson 1990: 77, 82-83).

X

tions, which appears also, on the scriptural level, in a line of Pāñcarātra

texts marked by non-duality.

The Śaiva tradition, which would continue to characterize social and

religious life in Kashmir, is for its part far from being unitary and if the

skein of texts, subdivisions and orientations is now beginning to be

unravelled, we are particularly indebted to a series of penetrating studies

by Alexis Sanderson3 – to which I refer here. What is evident is the

presence in Kashmir during this period of a sharp distinction between a

dualist tradition and a non-dualist one with extremist tendencies. The

former, the so-called Śaivasiddhānta, which embodies a kind of ‘normal-

ized’, mild, Tantrism – so much so that some scholars tend even to

exclude it from the Tantrism proper – perpetuates many of the social and

religious institutions of the ‘orthodox’ order which it aims to replace: it

basically accepts the caste system, prefers to understand the non-vedic

nature of their Āgamas and tends to identify with the establishment, in

maintaining a marked separation between pure and impure. It creates a

self-sufficient and all-pervasive ritual system, which, integrated by yoga,

leads the individual soul by degrees on a journey towards itself and the

reappropriation of its own true nature of Śiva: the last station is the

achievement of the state of liberated Śiva, for ever distinct from the

other monads of liberated individuals and from the supreme Śiva, who

alone performs the five functions on the cosmic plane.

Opposed to the Śaivasiddhāntins in many respects, the non-dualist

Śaivas follow the teachings embodied in another great division of the

Śaiva Āgama, that of the Bhairava Tantras, in which the terrific form of

the God – at least in the most radical texts – expresses first and foremost

the overthrow of the behavioural norms, conventions and distinctions

through which an attempt is made to give a definite and recognizable

structure to reality – fragile fences that the Power of the God, his Śakti,

sometimes superordinate to him, now derides now disrupts. The adept

enters into harmony with this Power, or rather he is possessed by it,

through practices and behaviours that here and there reveal their more or

less remote rooting in the practices of the Kāpālikas, who challenge what

the Indians see perhaps as the horror par excellence, the macabre fre-

quentation of the cemeteries, where they evoke the terrible Mothers by

eating putrefying human flesh etc. Some divinities of important Bhairava

Āgamas still display the animal features and hideous attributes, as well

3 See Bibliography; cf. also Dyczkowski 19882.

XI

as the names, that link them to this ancient substratum (cf. Sanderson 1988: 670, 674-75). At other times, instead of this (or alongside this), we find that sex is resorted to as the central moment of the rite: the adept worships the Goddess by offering her a cup containing the kundagolaka, or mixture of sperm and female secretions that the ritual union with his partner has just produced. Gradually relieved of its original burden of antinomianism and orgiastic transgression - and later totally eliminating the funerary aspect (Sanderson 1985: 202) -, this mobilization of energies through practices that challenge the current concept of purity and tend towards the overflow of the powers of the senses becomes the characteristic of the so-called kaula way (kaulaprakriyā) to liberation.

Several of the Bhairava Tantras are kaula in nature, especially the Yāmalas, but also, though to a softened and subtler degree, the so-called Trika Tantras. Among them an older stratum may be distinguished, consisting of the Siddhayogeśvarīmata, the MV and the Tantrasadbhāva, in which the term trika does not yet appear and which feature the essential doctrines but not the awareness of forming a group or school4, as is the case for the later Devyāyāmala, Trikasadbhāva, Trikahrdaya (or °sāra), Niśisaṃcāra etc.5.

SOMĀNANDA (875/900-925/950)

This 'school' (with the scriptures that are linked to it in various ways, such as those of the Krama) would probably have been destined to remain one of the many Kashmiri Tantric schools whose names we hardly know and whose outlines are blurred, had it not been for the development, mainly within it, of the first seeds of what was to become the

4 Cf. Dyczkowski 19882: 187 n. 189. On the three phases in the development of the Trika see Sanderson 1986: 170 n. 6; etc.

5 In the light of current knowledge, it is not easy to define what is really meant by one Tantra belonging to a particular 'school'. Every Tantra contains a mass of elements - at times even contradictory to one another - a portion of which may be shared also by Tantras belonging to different currents that are sometimes even far-removed from each other ideologically. For instance, cf. the long, important passage on the 'voids' etc. in chapter I (praśnayogādhikāra) of the Tantrasadbhāva (MS, National Archives, Kathmandu, No.A363, f.8a l.6 - f.11b l.1) - an old and 'radical' Trikatantra of the Vidyāpīṭha -, that is present almost verbatim in a 'mild' Bhairavatantra, in many respects close to the Śaivasiddhānta, like the SvT (IV.234-296), belonging to the lower Mantrapīṭha.

XII

flowering of an extraordinary series of works and masters that between the ninth and twelfth centuries constituted one of the highest achievements of Indian speculation and spirituality of all time. The complex work of exegesis of the scriptures6, the reformulation of their teaching and the organizing and hierarchizing of their contents indicate first and foremost its decision to emerge into the open, to escape from the dimension of a restricted circle of adepts - which is what must have been the original nature of these schools - and to offer itself implicitly as an alternative to the dominant Śaivasiddhānta, or at least to establish itself within it as a non-extraaneous element (or rather as an element which, though not intended to constitute the whole, was able to give it its ultimate significance). In order to do this it was necessary to extract a homogeneous though varied teaching from the diverse texts; to purge it, without changing its essential nature, of all that it was felt could not be proposed to a wider circle - in other words, of all that was bound to create an instinctive and insurmountable resistance - by attenuating the sharper points or removing every actually concrete aspect, and finally translating it into a discourse whose categories were shared by its addressees and engaging in a dialogue that would not be afraid to confront rival doctrines.

Vasugupta and Kallaṭa with the doctrine of the Spanda, on the one hand, and Somānanda, on the other, were the first to undertake this task. The former chose a plan that was more closely connected with spiritual experience, the latter one that was more in terms of conceptual elaboration (though his work is clearly based on direct experience7, which is sometimes visionary and ecstatic). But both the teachings and the tone of the ŚS and SK and of the ŚD are perfectly compatible: Somānanda is certainly aware of the doctrines of the Spanda which echo here and there in the ŚD (even the term spanda is to be found8), they all implicitly refer

6 The first to take this path must have been the philosophers of the Śaivasiddhānta. Sadyojyotis and Vyākhyāni (cit. ŚD III.13cd-14) were certainly Somānanda's predecessors (though their date cannot be fixed more precisely). Earlier than Somānanda is probably also Brhaspati, the author of the Śivatanuśāstra repeatedly quoted by Abh. in the TĀ, who is coupled with Sadyojyotis in the introductory stanza of Rāmakāntha's commentary on the Moksakārikā.

7 Cf. the vrtti on ĪPK IV.16, sākṣātkrtaparameśvarabhattārakākārair bhatta-śrīsomānandapādaih.

8 ŚD VII.19a, 24b.

XIII

to the authority of the VBh9. A late tradition (Chatterji 19142: 25 n.3) even includes Somānanda (along with the other masters of the Pratya- bhijñā) in the spiritual line descending from Vasugupta. Somānanda knows and criticizes (for his Śākta tendencies) Pradyumna Bhaṭṭa, who comes immediately after Kallaṭa in the Spanda lineage. I will not dwell here on the Spanda school, to which an exhaustive monography has recently been devoted (Dyczkowski 1987), except to say that its teach- ings, which had already emerged as a post-scriptural elaboration, were

subsequently re-elaborated, especially by Kṣemarāja, and incorporated in the synthesis of Abh.'s Trika, not without being to some extent strained and depersonalized10.

All the scant information we have on Somānanda indicates his close relationship to the Trika and the Krama. At the end of the ŚD he himself informs us that he belongs to the Tryambaka (or Teramba) maṭhikā founded by the son of Durvāsas, who was asked by Śiva himself to save the secret doctrine from oblivion. This is the same Tryambhaka who in the TĀ is considered the depositary of the non-dualist Śaiva tradition, in other words especially the Trika11. We learn from Abh. that he wrote a commentary - now lost - on the Parātrimśikā, which belongs to the ekavīra form of the Trika. A surviving passage of the Kramakeli of Abh., quoted and commented on by Jayaratha (TĀV III p. 192), men- tions Somānanda as a disciple of the Krama teacher Govindarāja, who, before dying, passed on to him the doctrine of the Kālīs taught in the Devīpañcaśatika; this doctrine was transmitted by Somānanda, through his spiritual line, down to Abh. (ibid. p. 194).

The only work by Somānanda that has come down to us, the ŚD (cf. Gnoli 1957, 1959), is considered to be the first of the Pratyabhijñā school. In this difficult, discordant but fascinating work, he alternates expositions of a non-dualist doctrine that is felt to be still very near to its

9 Kṣemarāja connects the Spanda, first of all, with the Krama (SN p. 74 evam copakramopasamhārayor mahārthasamputīkāram darśayan ...) and, secondly, with the Trika (Sadardha) and the Mata (p. 49, 47). The concept of spanda (Sanderson 1988:695) is present in the Jayadrathayāmala and in other texts linked to the Kālī cult.

10 See, for instance, in the ŚSV the superimposition on the ŚS of the doctrine of upāyas, which Abh. had derived from the MV.

11 TĀV vol. I p. 28 nikhilaśāstroponiṣadbhūtasya śadardhakramavijñānāsya traiyambhakasantānadvārena avatārakatvāt; TĀ XXXVI.12 tryambhakāmardakābhi- khyaśrināthā advaye dvaye / dvayādvaye ca nipunā kramaṇa śivaśāsane (Jayaratha: advaye iti trikakulādau).

XIV

scriptural sources - though these cannot be clearly distinguished -, logical justifications, visionary tirades and summary dismissals of opposing doctrines; he presents an exceedingly long and malevolent demolition of the theories of the Grammarians. The central nucleus is the powerful and effective affirmation of Śiva as the only reality: he does not render the universe so pale that it is transformed, as in the contemporary Vedāntic schools, into an indefinable play of shadows, but illuminates it from within and constitutes its life. In this universe where everything is penetrated by Śiva nothing remains in the margins, all is mirrored even in the humblest thing and the whole nature of Śiva is present in it (ŚD III.18ab svaniṣṭhe śivatā deve prthivyādāv apīdrśam) - directly, just as gold is directly gold in the jewel, though it assumes a particular form.

The doctrine of the sarvasarvātmavāda is the other face of the ekaśivatā. Though this was already a long-standing concept12, here it acquires very special importance and implications because it serves as the theoretical context for the experience of the Tantric adept who has set out on the path of the expansion of consciousness and energy: the object, though maintaining its own identity, must not act as a dam, nor oppose itself rigidly, but allow itself to be bent and, finally, uprooted, become an integral part of this free flowing.

The principal argument in favour of this doctrine clearly derives from the Vijñānavāda. If there were a real diffence in nature between consciousness and things, the phenomenon of knowledge would be impossible, because things, whether they have a concrete form or not (atoms), cannot become the object of consciousness (IV.30). No relation between things themselves would be possible, if they did not share one and the same principle (V.1); this principle is precisely consciousness, which exists equally in all things (V.12). 'Being' is, actually, being united with the manifestation of consciousness (cidvyaktiyogitā; cf. IV.29; IV.7ab sarvabhāveṣu cidyaktekh sthitaiva paramārthatā). Thus everything is pervasive, incorporeal and endowed with will, like consciousness (V.4). If things can be efficient, it is because they 'want' one particular action that is peculiar to them (V.16,37). And if they want it, they must also know it, in.other words be conscious - first and foremost,

12 This conception is already referred to in the Mahābhāṣya (Wezler 1982). See also Wezler 1992.

XV

of themselves. All things are in all conditions13 knowing their own self (V.105ab sarve bhāvāḥ svam ātmānam jānantaḥ sarvataḥ sthitāḥ). And that self, continues Somānanda in a visionary crescendo, is the same as mine, as that of other subjects: “The jar knows by my own self. I know by its self; I know by the self of Sadāśiva, Sadāśiva knows by my self ...” (V.106 ab). All things have the essence of all others, because every thing has the form-nature of all things. Everything is in everything, with the various configuration of forms. I have the nature of the jar, the jar has mine. In knowing himself through the different things, Śiva resides autonomously, being constituted by the manifestation of consciousness and differentiated in the manifold differentiations (V.107cd-109). Once it has been established that everything has the nature of Śiva, the question of the unity and multiplicity of the universe becomes pointless, and so does the question of its reality-truth; even as regards the so-called ‘erroneous’ cognition there is no real falsity (mithyātva), because even in the eyes of the opponent it has, anyhow, some kind of efficiency (IV.18-20). Though the world of ordinary experience (vyavahāra) is said to be unreal, the reply is that the Lord is also manifest in unreality14 (III.77). But what is then the nature of manifested things, which on the one hand are limited and, on the other, enjoy a state of expansion, being united with the supreme pacification of Śiva (V.6)? Like everything that exists, they are states in which Śiva freely manifests himself (IV.47). It is true that they have power, they are śakta, but, as everyday experience also shows, having power, or being able to do something, does not mean being independent15. If they are śaktis, Śiva is the possessor of these śaktis, though being inseparable from them; he is the only real śakta in ways as infinite as his powers (IV.4-5). If they are states (avasthā, bhāva), he is their substratum16.

We are, therefore, in a markedly śaiva or sāmbhava context, in the sense that the Power - though being inseparable from Śiva (III.2cd) - is subordinate to him. Once this premise is established, it is possible to let it take the stage and see it rise in many points of the work to the role of

13 The subject cannot be insentient even with regard to the mere 'being' or 'standing' (V.14).

14 And also in plain (V.9).

15 If things were in themselves totally independent, the unification (anu-samdhāna) of their cognitions would be impossible (ŚDVr p. 147).

16 Cf. (punning on the root bhū-) IV.51ab tasmāt svayam svabhāvena bhāvair bhāvir bhaved bhavah.

XVI

protagonist17. Somānanda, following the teaching of the Trika, distinguishes a triad of powers - icchā, jñāna and kriyā - (connected with the level of Śakti, Sadāśiva and Īśvara respectively), which in fact are never really separated from each other. Prior to them, at the top of the Śaktitattva plane, there is a moment in which they are completely mingled and about to unfold: this is the state called unmukhitā, aunmukhya ‘protention’ or prathama tuṭiḥ ‘first moment (of the will)’ (I.18cd), ‘wave’ (SDVṛ p. 16 taranga, ūrmi). When these powers are totally dissolved in unity with Śiva, we have the supreme state, characterized by the supreme lysis (nirvṛti) and the conscious beatitude (cidrūpāhlāda). But in no case can one speak of Śiva as quiescent or isolated (śānta, kevala) (III.87b-88), insofar as devoid of power (III.90ab)18, nor can one speak of cessation of the perception of the self, since Śiva would then be insentient (jada; III.90cd-91ab). His turning to creation - and more generally to his five functions - springs out of his own joy (amoda); he plays at concealing himself and assumes the nature of māyā down to the Earth (I.32). Thus it cannot be said that the universe is ‘imagined’ as Śiva, or vice versa, because the one is directly the other (III.83). Just as gold is not ‘imagined’ as such neither in the simple jewel of solid gold nor in the earring in which the work is so refined as to set aside, as it were, its nature of pure gold, so Śiva is ‘formed, arranged’ (kṛpta) as universe - in the sense that he has become such, or freely presents himself in this form (ŚDVṛ p. 133, kṛptah sampannah svayam sthitah); kalpana is, on the contrary, imagining something where it is not (III.87c-91ab). Therefore, the nature of Śiva is present in everything, whether we know it or not (VII.1cd). Even differentiation has Śiva as its essence19, hence there is not really a bond nor, consequently, liberation (III.72). Furthermore, it is the very belief in the actual existence of bond and liberation that constitutes the basic impurity (VII.87cd). All the ritual prescriptions, śāstras and teachers only serve the purposes of everyday life and are in any case a way in which Śiva manifests himself (III.48-49ab). Ritual bathing,

17 This is anything but an isolated case in the Śaiva scriptures and treatises (see also, in the Pāñcarātra tradition, the Lakṣmītantra etc.).

18 If we admit the reality of Śiva's state of inactivity, we are obliged to admit that, in spite of everything, the three powers are present in it; it is precisely this meeting of the three powers that constitutes the requisite of reality (III.56).

19 Even if in Śaiva scriptures themselves we sometimes find the unreality of the world affirmed, there is no real contradiction here, since this is merely a device to promote detachment (III.95cd).

XVII

offering to the fire, pūjā etc. (VII.87 ff.) are external forms whose core is the awareness that everything is resolved in Śiva. He who worships, he who is worshipped and the act of worship are in reality all one – Śiva (VII.94cd). The Śiva principle, whether it is known or not, does not suffer any real obstruction (VII.1 ff.). Fire, whether externally perceived or not, still generates light, and gold, if it is not known, does not therefore become a stone. All this is doubtless true; it is true that the gem of desires remains such even if it is not known, but only if man knows it as such can he enjoy its effects (VII.4). The same may be affirmed regarding individuals' identification with Śiva. It has been said that even a fire that is not known generates light, but so much more will be generated by a fire aptly arranged for this purpose, such as a lamp in a house (VII.10cd ff.). Thus a means must be taught whereby the attained awareness of the nature of Śiva produces visible effects in the souls (VII.12ab). This means is represented by logical argumentation, the scriptures and the teacher20 (VII.5cd). The ŚD condenses all this.

It is therefore easy to imagine on which arguments in particular Somānanda challenges the opposing doctrines. First of all he attacks all those who in various ways claim that the external world is unreal, especially the various types of Vedāntins (VI.3 ff.), who consider it as an illusory manifestation (vivarta) of Brahman, caused by nescience, or the Vijñānavādins (VI.33-34) who affirm the reality of knowledge but make unreal objects arise from it and, moreover, do not admit a subject of this knowledge – whereas every action, and therefore also the action of knowing, is necessarily dependent on an agent subject. The thesis of the Śaktivādins is also rejected for a similar reason: they consider Power as the sole reality, without taking into account that a power, even though it is supreme, is implicitly dependent on a subject that possesses it. Then there are those who admit agent subjects but make each of them an entity that is separate and distinct from the Lord; those who sharply divide bond and liberation, like the followers of the Nyāya-Vaiśeṣika (VI.28cd); and those who claim a Brahman devoid of consciousness, and hence jada (VI.29). Even those who claim the reality of the external object, like the bāhyavādin Buddhists, are wrong because they do not admit a unifying principle of reality, and thus make the passage from the moment of sensation to the moment of mental elaboration impossible (IV.80cd

20 Cf. p. XXXVII.

XVIII

ff.), and likewise the operation of apoha, through which they seek to elude the universal.

It is essentially on the basis of these arguments, used either in isolation or together, that the opposing doctrines (apart from those already mentioned, Jaina, Pāñcarātra, Śaivasiddhānta, Sāṃkhya etc.) are gradually dismissed. Several of these objections are concentrated in the criticism of the doctrines of the Grammarians - namely of Bhartrhari - which differs from the others for its disproportionate length (the whole of āhnika II) and its aggressive, sarcastic tone.

These Grammarians who think themselves so clever - says Somānanda quoting passages from the VP and the vrtti - claim that the supreme Brahman is to be identified with Paśyantī-vāc, but the latter at the most is identifiable with the power of Knowledge, that corresponds to the plane of Sadāśiva - definitely not to the highest plane. In fact, as the word itself says, Paśyantī 'sees' something. But what is the nature of its objects? If we say that they are externalised images that it itself has produced, we must ask ourselves whether they are real or unreal. In the first case the distinctive feature of this philosophy is lacking, namely seeing the manifestation as illusory (vivarta); the unreality of what it sees would have repercussions on it, rendering it asatyā, which is inadmissible. If the cause of this perceiving unreal things is nescicence, it is the latter that must be established as being real or not: if it is real, then Paśyantī is contaminated by it; if it is not real, it is not understood how there can be a relation between a real thing and an unreal one. The same is also true of its creation. Moreover, this nescicence can be seen neither as an attribute of Paśyantī (for the above-mentioned reasons); nor of something else, since nothing real exists apart from it; nor is it to be considered as independent, because then it would be impossible to suppress it. Neither can avidyā be imputed only to the middle level, because in any case it is Paśyantī that is the cause of it. If Paśyantī is to be identified with the sphoṭa, we must ask ourselves how words, unreal as they are, can manifest it. No means of knowledge can prove its existence, since for you nothing is real except it. Even the belief that pronouncing a correct word leads to heaven results in attributing to the only reality that can be its subject characteristics which are contrary to its nature, such as the desire for particular fruitions, etc. It is not possible to posit as the highest reality that which, because of its very nature, always remains - however you put it - an instrument of action (vāc). Not even identifying the śabdatattva without beginning and end with Parā Vāc makes sense, since this means identifying the object (sound) with the in-

XIX

strument (voice). If, then, Paśyantī is said to be only a proper noun and hence to escape all the criticism concerning the fact of seeing, etc., and that it is only the feminine ending that counts - which is intended to express its being the power of Knowledge -, the reply is that, if it has nothing to do with seeing, it is insentient and therefore cannot be the power of Knowledge either. Then, the attributes that a verse21 assigns to Paśyantī are criticized one by one. And so on in this vein, until his final outburst: But why on earth have you left the sphere of grammar and taken it into your heads to deal with a field which is not yours, like philosophy?

Utpaladeva (900/925-950/975)

Though Somānanda is unanimously recognized as the founder of the Pratyabhijñā, the school was not to be named after him or his work, but after the work of his disciple Utp. The term pratyabhijñāna occurs only once in the ŚD (IV.120a) and has no pregnant meaning, since it is used as an argument to demonstrate the essential unity, or possibility of unification, of cognitions (against Buddhists etc.). If, on the one hand, it is true that the ĪPK are a 'reflection' of the ŚD, as Abh. says22, on the other hand it is undeniable that they contain important novelties, not so much in the basic doctrine as in the far more aware and acute determination of the aims proposed, the ways of attaining them and the ambit addressed.

First and foremost Utp. , being perhaps actually aided by the negative example of the ŚD, decides not to mix registers. The ĪPK is to be a purely theoretical and rigorously argumented work, though based on a scriptural background. The authority of the revelation must not be superimposed on the rational demonstration, and for this reason it is allotted a specific space, the third Adhikāra23. Unlike what occurs here and there

21 avibhägaśā tu paśyantī sarvataḥ samhrtakramā / svarūpajyotir evāntaḥ sūkṣmā vāg anapāyinī // This verse, frequently quoted, is included (with reserve) by Rau - along with the entire passage to which it belongs - in the kārikā text (1.167); Iyer, instead, takes it as a quotation given in vrtti.

22 ĪPV I p. 7 śrisomānandanāthasya vijñānapratibimbakam.

23 Cf. ĪPVV III p. 400 jñānakriyādhikārayugalenā tattvārthasamgrahasahitena svaprakāśe 'pi yuktyupabrmhite nīrūpite 'svalah' ity amśah pūritah, āgamādhikāreṇa

XX

in the ŚD, any emotional colouring is banished. This does not mean that Utp. was solely a logician: the extraordinary intensity of his hymns24, which were to be collected in the ŚSĀ and are still recited daily by the brahmins of Kashmir, proves it. In a sense Utp. inaugurates what was to become a salient feature of the whole Trika in Abh.'s synthesis: namely, the tendency not to constitute a monolithic doctrine and a world of religious experience to oppose en bloc everything that does not coincide with it (as in the ekāntin trends) but to distinguish planes, that are hierarchically ordered but in which the 'higher' does not automatically cancel the 'lower' (as Somānanda had already said, Śiva is everywhere, even in differentiation, pain and hell). This is the perspective of the paramādvaita, such an elevated viewpoint that it does not fear what is different from itself, is not put in a critical position by it, is not forced to make a choice. If we take the ĪS, then we find a still different face. Were it not for its certain attribution, we would never suspect that the refined Naiyāyika speaking here is the same person as the Śaiva theologian of the ĪPK or the bhākta of the ŚSĀ25.

But let us return to the ĪPK and, firstly, let us see how Utp. determines the choice of his opponents. The most important feature is that the Śaivasiddhāntins virtually disappear, at least as direct targets, and the stage is dominated by his critique of Buddhism. This is a message Utp. addresses to his dualist cousins, whom - as will be seen more clearly in Abh. - the Trika does not intend to defeat, but to use as a basis for its emergence from the dimension of a restricted circle and for its establishment in the stratum of social normality, by internalizing, or in any case circumscribing, its own specific difference26. The fact that the MV - that is, a text containing distinct dualist lines (Chatterji 19142: 10, Sanderson 1992) and as such is venerated also by the Siddhānta - was chosen by Abh. (following his master Śambhunātha ?) as the basic text for the

tam eva artham upodbalayatā "śāstrataḥ" ity ayam. Here Abh. is referring to the well-known passage of the Kiraṇāgama, Vidyāpāda IX.14b.

24 We should recall, on the other hand, that the very starting point of Pratya- bhijñā theology is the state of devotion, the maheśvarasya dāsyam mentioned in the first verse of the ĪPK. Cf. ŚSĀ XV.4 etc.

25 It is only the last verse that enables us to see it in the right perspective.

26 This attitude is aptly outlined by an oft quoted verse: antah kaulo bahih śaivo lokācāre tu vaidikaḥ / sāram ādāya tiṣṭheta narīkela-phalam yathā // (cf. Sanderson 1985: 203-205; id. 1988: 699)

XXI

Trika is a clear indication of this attitude27. Choosing Buddhism, and particularly its logical school, as number one enemy means reinforcing the undeclared alliance with the Śaivasiddhāntins, who see Utp. siding with them against those who had already been the opponents of Sadyojyotis (see for instance the first chapter of his Nareśvaraparīkṣā) and were to be the opponents criticized by Nārāyanakantha and especially by Rāmakanṭha, who displays a particularly vast knowledge of their works and doctrines. This lengthy examination and criticism of the teaching of the Buddhist logicians resulted in, or at least was accompanied by, the peculiar phenomenon of a more or less conscious absorption of their doctrines and their terminology, that was to leave substantial traces in the structure of the Pratyabhijñā28. This may have been a deliberate choice by Utp.: to increase his own prestige by assuming the ways and forms of a philosophical school which was perhaps the most respected and feared, even by the many who did not agree with it. But certainly one must also view this in the light of the complex relationship with the Nyāya.

The Nyāya was also given a wide berth as an opponent. This, too, is probably part of the manoeuvre to place the Trika in a broader Śaiva context, seeing that the Śaiva faith was known to be pre-eminent among the Naiyāyikas. It is probably for this reason that we see Utp. assuming a Naiyāyika guise in the ĪS (Abh. also dealt with the philosophy of this school in the Kathāmukhatilaka, now lost29). Against the realism of the Nyāya, that claimed the reality of the external object and the actual existence of concepts such as relation etc., Utp. sets the criticism of the Bud-

27 According to Abh., the Trika - and consequently the MV, which is its essence - represents the essence of the entire Śaiva tradition and not of the sole non-dualism (TĀ 1.18 daśāṣṭadaśavasvasṭabhinnaṃ yaḥ chāsanaṃ vibhoh ; tatsāraṃ trikaśāstram hi tatsāraṃ mālinīmatam !!).

28 I will not dwell on this point, which I have already dealt with elsewhere (Torella 1992). Also Somānanda presents and criticizes themes and doctrines of Buddhist pramāṇa tradition (inference, the universal as exclusion of the other, instantaneity; cf. in particular ŚD IV.68 ff., V.44 ff. - within a general criticism of the inference placed outside the context of ekaśivatā -, VI.77 ff.) and refers, more or less explicitly, to two verses by Dharmakīrti (PV III.282 in ŚD I.45 and PV III.354 in ŚD VI.39). In the ŚD we find peculiar terms of Buddhist logic, such as svalakṣaṇa (IV.71a), apoha (VI.76c), svārthānumāna (V.55a), parānumāna (VI.61c) etc., but they are always referred to the opposing doctrines, without showing any trace of conditioning or appropriation (as it is in Utp. and Abh.)

29 Cf. IPVV I p. 20.

XXII

dhist logicians, but only to show that they would easily get the better of its relatively ingenuous realism if the Pratyabhijñā did not intervene to support it. After letting the Buddhist logicians demolish the Nyāya categories, he shows how the Buddhist alternative is in fact equally inadmissible. It does overcome the Nyāya, but remains as though suspended in mid air, since it is proved - in its fragmented and isolated universe - to be incapable of accounting for the network of relations and the circularity of the world of human experience. So - Utp. seems to say, and later Abh. was to put this more explicitly - one might just as well accept the view of the Nyāya in the sphere of the vyavahāra, on condition that one sees through it the pervasive presence of Śiva as constituting its dynamism and internal coherence. It could equally well resolve the inconsistencies of the Buddhist view and render them acceptable. In this way, Utp. achieves the result of both showing the superiority of the Pratyabhijñā to Buddhism and warning the Nyāya not to count too much on its forces alone, detached from those of the new Śaiva theologians.

Another important aspect of the strategy of Utp.'s Pratyabhijñā is the glaring reversal of Somānanda's attitude towards Bhartrhari, who, from being a punctiliously criticized and even derided opponent, becomes one of the major inspirers30 throughout the ĪPK (as I have sought to point out in the notes to the translation), and in particular in one of its essential doctrines. I am referring to the identification in Śiva of the dual prakāśa-vimarśa pole - the first understood as the motionless cognitive light that constitutes the basic fabric, the founding structure of reality, of the 'given'; the second as the spark that causes this luminous structure to pulsate by introducing self-awareness, dynamism, freedom of intervention, of self-assertion, thus expressing in theoretical terms what is the nature of an unpredictable divine personality, like that of the violent and loving Śiva handed down in the Scriptures and with whom Utp. dialogues in his mystical hymns. Prakāśa forms, together with a large group of synonyms or quasi-synonyms (from the roots bhā-, pratibhā-, bhās-, avabhās-, ābhās-, pratibhās-, prath-)31, a close-knit constellation

30 The fact that the Pratyabhijñā shows at one time the influence of the Buddhists and of Bhartrhari does not appear as a contradiction, if we take into account the undoubted relations and mutual exchanges between the doctrines of the Buddhist schools (Mādhyamika, Vijñānavādin, logicians) and the doctrines of Bhartrhari (cf. e.g. Lindtner 1993).

31 To these must be added another key term, sphur-, where the light is enriched by a 'vibrant', dynamic, connotation - which makes it into the intersection point with the doctrine of the Spanda.

XXIII

of 'luminous' terms indicating the notions of being manifested, emerging from the dark, coming to consciousness or, more in general, of being the object of knowledge and finally simply 'being', whose use was already firmly established especially in Vedāntic and Buddhist (particularly logical) contexts; prakāśa and synonyms frequently occur in the VP.

Apart from isolated and uncertain cases in the Śaiva scriptural tradition (cf. Dyczkowski 1987: 233 n. 10; cf. also ŚD II.83d, 84c), vimarśa32, in the pregnant sense Utp. attributes to it, cannot but derive from Bhartrhari's teaching, especially if we consider its link with light, on the one hand, and the word on the other. Of course, I am referring here to the two very famous and most quoted stanzas I.131-32 of the VP.33, whose influence, though extending over the whole structure of the Pratyabhijñā, we find concentrated in particular on two closely connected aspects. One (ĪPK 1.5.19) concerns the only way deemed possible to account for a common fact in everyday experience, such as the immediate and seemingly thoughtless action that still achieves its purpose - namely, that of affirming the presence of a subtle reflective awareness even within the sensation or movement captured at its most direct and undif-

32 Vimarśa has been translated in various ways and there are good reasons for each of them: 'cogitazione, pensiero' (Gnoli), 'prise de conscience (Silburn), 'self-consciousness, freedom, determinate consciousness' (Pandey), 'raissaisissement infini' (Hulin), 'Betrachtung, Urteil' (Frauwallner), 'self-representation' (Sanderson), 'prise de conscience active, libre activité de la Conscience' (Padoux) etc. The translation 'reflective awareness' that I have generally adopted - drawn from Matilal 1968a, who however uses it to translate anuvyavasāya (also Dyczkowski uses it frequently) - seemed to me broad enough to be adapted to the different contexts and meanings in which the term is used. Being neither too precise nor too vague, it permits me not to disseminate the text of different expressions for the same term. Vimarśa is accompanied by a series of terms deriving from the same root with different preverbs (parā°, pratyava°, ava°, ā°). A differentiation between them might be attempted by identifying a more intimate and analytic element in vimarśa, a more instantaneous and indefinite element in parāmarśa, one more characterized by injection and return to the subject in pratyavamarśa. However, the close reading of the texts of this school indicates that, even if these different nuances are in principle not unfounded, they are largely used as interchangeable terms. Another important notion, closely connected with vimarśa - and probably first introduced by Utp. -, is that of camatkāra 'astonished, wondrous, savouring' (see below p. 118 n. 23). It could be defined as a vimarśa enriched with a strong aesthetic connotation.

33 See below p. 125 n. 41.

XXIV

ferentiated moment34. The other aspect regards the two solemn general formulations (I.5.11 and 13) that define vimarśa (pratyavamarśa in the vrtti) as the essential nature of light (avabhāsa; prakāśa in the vrtti) and indissolubly link consciousness (citi etc.), reflective awareness (pratya- vamarśa) and the supreme word (parā vāc)35.

The presence of Bhartrhari, not only in the substance of these doctrines but, often also in their verbal formulation, could not be more direct - cven if the vrtti of VP36 does nothing to emphasize the apparent- tly very general meaning of the verse but, on the contrary, seems to do its utmost to keep it down to earth. In making this doctrine one of the crucial points of the Pratyabhijñā, Utp. appropriates another prestigious tradition, that of the Grammarians, and Bhartrhari in particular, whose teaching on the all-pervasive power of the word is, moreover, connected with a very ancient speculative trend, which also continues in the Āgamic tradition close to Utp. This is incorporated with no effort in the philosophy of the Pratyabhijñā, enriching it with implications, and - something that must not be underestimated - it furnishes a further, penetrating argument in the dispute against the Buddhist logicians, one of whose basic theses was precisely the absolute otherness of direct sensation from determinate knowledge. However, it is true that the criticism of the tripartition of the word held by Bhartrhari (cf. below p. 120 n. 26) remains - according to what appears from the passages of the țīkā sketchily quoted by Abh. in the ĪPVV. But it seems to slip into the background, compared to the general acceptance of Bhartrhari's teaching as a whole. One might even say that Utp. criticizes the tripartition also because this, though unanimously recognized by the contemporary exegesis of the VP and explicitly mentioned by Bhartrhari himself, appears

34 This conception is not to be referred solely to Bhartrhari - as will be seen later on - but may also be considered as a development in philosophical terms of the experience of the 'effervescence of all Powers' (sarvaśaktivikāśa), described in Chapter I of the ŚD and, in turn, derived by Somānanda from the scriptures (first of all, the VBh).

35 See below p. 120 n. 26.

36 The question of the identification of the author of the vrtti on the first two Kāndas of the VP - indicated in the colophons as Harivrṣabha - with Bhartrhari himself is still to be considered an open one (see the recent contributions by Akluj- kar 1972,1993 and Bronkhorst 1988,1991, and for against this identification respec- tively). In the course of this work I shall provisionally accept the thesis of identity, which seems to me the more likely and which, above all, has always been accepted by the Indian tradition.

XXV

to him not in line with the general plan of the VP, to which a quadripartition like the Śaiva one would be more suited. The mention of a supreme form of Paśyantī made in the vrtti on the ŚD (p. 38) - which Somānanda had not taken into consideration -, even if later Utp. considers it included in the criticism of Somānanda, might be understood as Utp.'s airing of a hypothesis of fourth level in Bhartrhari himself. An indirect confirmation might be represented by the explicit equivalence that his disciple Rāmakantḥha (Spandakārikāvivṛti p. 153) establishes between śabdatattva and Parāvāc, which, therefore, does not coincide with Paśyantī but of which the latter is an emanation. Moreover, Rāmakantḥha seems inclined not to underline the illusory nature of the Śabdādvayavādins' vivarta, but he considers it simply as a different way of denominating that process of the Lord's articulation in the multiplicity of the universe which the Śaivas call śaktiprasara 'flowing of the Power' (ibid.).

At any rate, it is clear that we are far removed from the attitude of Somānanda37, who minimizes even that aspect which by contrast was to strike Utp. so forcefully - namely, the inevitable presence of language at the heart of every cognitive activity (ŚD II. 19-20)38. And I like to think that the vidvad° Utp. places in front of Bhartrhari's name when, in ŚDṛ p. 84, he quotes in full a verse of the Saddhātusamīkṣā39, which Somānanda is fiercely attacking at the time, is a kind of veiled apology,

37 Furthermore, the conception of the Absolute that results from an overall view of the VP appears close to that of the Pratyabhijñā. Positions such as that expressed in the Sādhanasamuddeśa (vv. 39-41) could casily have been subscribed to by Somānanda himself (who refers only to the Kānda I), even if we do not read it in the light of Helārāja explanation (Prakīrṇakaprakāśa I p. 262 ekam eva samvinmayam param śabdabrahmā tathā tathāvasthitam iti kārikārthaḥ). The very notions of avidyā and vivarta, examined in the whole of the VP, seem to escape Somānanda's criticism. Avidyā, in fact, is not a principle extrancous to Brahman but a śakti (VP I Vṛtti. p. 9) and, as such, does not weaken but rather accentuates Brahman's centrality. Similarly, the term vivarta (and related forms) seems to be used by Bhartrhari to underline the continuity of Brahman in the manifold world of manifestation rather than to signify a de-realization of the latter. Somānanda, however, is certainly not the only one to interpret Bhartrhari in this way; while some of his early critics, like Jayanta or Śāntarakṣita-Kamalaśīla, show they are at least aware of the problems in interpreting these key-terms, the later Vedāntins will not seem to have any doubts about reading them in the light of the subsequent Śaṅkarian doctrines.

38 The verse figures as an objection that Somānanda essentially pass over.

39 Quotations of this work attributed to Bhartrhari (Iyer 1969: 9-10), now lost, are also to be found in the Spandapradīpikā (pp. 4, 22). Of the two different titles

XXVI

a timid dissociation from what he finds himself commenting on with the devout diligence of a disciple.

A further original contribution by Utp. to the philosophy of the Pratyabhijñā is the doctrine of the ābhāsas. Somānanda had not developed his own idea of the ontological status of the manifested world, but had only asserted strongly and repeatedly its reality (satyatā) and its having the nature of Śiva (śivarūpatā). Things are ‘states’ (avasthā bhāva) of Śiva, and their emergence is due solely to his will, brought about by nothing other than a natural overflowing of energies, whose characteristic feature is ‘joy’ (ānanda) as well as ‘play’ (krīḍā). Utp. includes the world of manifestation in his great luminous metaphor. Every aspect of reality is a light; it is a reflection in the mirror of consciousness and has its ultimate reality in consciousness.

By combining elements drawn from the Vaiyākaraṇa and the Buddhist pramāṇa tradition, Utp. presents these ‘manifestations’ (ābhāsa) as having the nature of universal (in fact, each one of them is connected with a word). They can appear in isolation or aggregated around a dominant manifestation and are provided with a single or complex causal efficiency, on the basis of a compatibility that has its ultimate foundation in the law of Necessity established by the Lord. United among themselves, thus becoming more and more particularized and, finally, combined with three manifestations endowed with a special individualizing force – space, time and form – they constitute the world of everyday experience. It is easy to glimpse in this conception elements drawn from the Vijñānavādins (consciousness as the receptacle of everything that is gradually manifested), from the Vaiyākaraṇas (the idea of a hierarchy of universals) and from the logical-epistemological school of Buddhism (cf. Torella 1992: 332 ff.). In this case, too, the word ābhāsa as a philosophical term, was not invented by Utp., but was commonly used in the Vedāntic and Buddhist schools. It also occurs in the ŚD40, but, even

that are attributed to it – Śabdadhātusamīkṣā in the ŚDVr and Saddhātusamīkṣā in the Spandapradīpikā (Dhātu° is only the abbreviated form; cf. Gaurinath Shastri 1959:61, who traced out one more quotation from this work), the right one is most probably the latter (which, besides, better lends itself to be abbreviated), as can be deduced from Utp.’s commentary: dhātusaṭkopagamāt bhedavāde ‘pi ... This work may possibly have been an examination of the six Vaiśeṣika categories (even though I am not aware of any other occurrence of the term dhātu in the sense of padārtha), or (J. Bronkhorst’s personal communication) of the six dhātus mentioned in Buddhist works, such as the Ṣikṣāsamuccaya.

40 Cf. I.33c, II.21c, III.78b, IV.98c, VI.58d etc.

XXVII

when it does not appear in a context where explicit reference is made to opposing doctrines (various kinds of vivarta), its use seems to be merely sporadic and casual, and in any case devoid of a precise technical con-notation. In one respect, ābhāsa is not distinguished from prakāśa (and related terms), and the two terms often seem interchangeable (svābhāsa, svaprakāśa etc.; cf. ŚSĀ I.1c śivābhāsah). In another (namely. in its more technical use, as in the so-called ābhāsa theory), though the essen-tial unity of nature remains, ābhāsa is seen as a particle, an individualized and introverted form of the 'great light'41, 'cut out' in it. This frag-mentation of the light is accompanied by, and also presupposes, an analogous descent of consciousness to the state of fragmented subjec-tivity of the manifold individual subjects (the true subject is avic-chinnābhāsa; cf. ĪPV II p. 138); see in particular Abh.'s commentaries on ĪPK II.3.1-2. The relation between consciousness and ābhāsa is that between the mirror and the reflected image, subtly analyzed in Chapter III of the TĀ (cf. also MMP pp. 147-53). The ontological status of the ābhāsa is therefore a mixture of autonomy and heteronomy, without its basic reality ever being called into question: reality is precisely being manifest (ĪPK I.4.14 and vrtti; cf. APS 5ab evam apy avabhāsānusaṭā sattocitā bhavet). Though he never explicitly repeats Somānanda's Vijñānavāda argument, Utp. underlines the necessarily common nature of consciousness and its object: only that which is itself light, i.e. sen-tience42, can shine in knowledge; to say that something is insentient is like saying it is inexistent (APS 13ac). Even when an ābhāsa is viewed as external - as 'this' - the 'this' continues to have its foundation in the I; it may also be said that the reflective awareness of something in terms

41 In this case the Śaiva tradition interprets ābhāsa as 'limited light', by attributing to ā° the meaning of īsat; in this sense, the supreme Śiva is nirābhāsa (ĪPVV III p. 271).

42 Among the rare passages in the Śaiva scriptures that explicitly touch on and philosophically elaborate this point, see that of the Ucchuṣmabhairava (cit. ŚSV p. 4; Torella 1979: 39): "How is it possible, O Dear One, that these be objects of knowledge, without being also subjects of knowledge. It is for this reason that impurity does not exist.", and of the Kālikākrama (cit. ŚSV pp. 57-58; Torella 1979: 112): "Knowledge shines within and without, in various forms, and there is no existence of object that is outside knowledge. The whole world is, therefore, constituted by knowledge, and no entity is perceived without knowledge. Knowledge has the same nature as the object. [...] The single nature of knowing and the knowable comes out of their mutual implication in the cognitive act."

XXVIII

of separation, of 'this', has been fully achieved only when it rests in its innermost being, thus becoming the reflective awareness 'I' (APS 15). It is the same light of the self that is manifested as self and as other (APS 13cd).

Conceiving of the supreme reality as an absolute I is another innovation made by Utp. (cf. Dyczkowski 1990), though in my opinion there are no substantial differences between this and Somānanda's dynamic Self-Śiva which underlies the whole universe and express himself in it. Utp. is the one who chose to use this word regardless of the associations generally attached to it in Indian thought, being aware of the fact that the risk of a reification that has always weighed heavily on the word ātman was even more negative, and that this makes it less suitable for expressing the unpredictable overflowing of the divine personality. The term 'I' is implicitly aimed against the two conceptions that are, after all, closest to the Pratyabhijñā and which it most aspires to differentiate itself from: the consciousness devoid of a subject of the Vijñānavāda and the static ātman-brahman of the Vedānta. "The resting of the light in itself, or the supreme condition of the I, is absolute independence from every other, autonomy (svātantrya), agency, Lordship " (APS 22cd-23).

I will not present in detail all the argumentations Utp. develops in the Jñāna° and in the Kriyādhikāra (regarding which I refer directly to the translation and notes). They mainly take the form of a debate with the Buddhist logicians and also betray, as has been pointed out, their strong influence. Utp.'s constant preoccupation is to show, in every context he deals with, the need for a single, dynamic subject that unifies and animates the discontinuity of reality and constitutes the substratum of every limited subject, as well as of every form and activity of everyday life.

This I or Consciousness is, on the religious plane, Śiva. In his highest form, the supreme divine personality is solely 'I' - consisting of consciousness and beatitude - in whom all the principles are contained though in a state of complete dissolution. He is present throughout the ĪPK as the ultimate essence of every reality and is also directly mentioned here and there, even if the stage is generally occupied by a less extreme form of him, which balances between transcendence and immanence. In fact, being an expository work, which requires an object to teach and a recipient of this teaching, it cannot but deal with that form of the God which is open to the world of manifestation, whilst firmly remaining its sovereign. On the supreme plane there is only the I resting in his fullness and no trace of the knowable remains; even the Pratya- bhijñā has to withdraw from this plane, leaving it in the background. In

XXIX

the various models of tripartition that the ĪPK presents, following in the wake of the Trika (aham, aham-idam, idam; śuddha, śuddhāśuddha, śuddha; para, parāpara, apara), this more accessible form of the God is connected with the second level. It is mainly indicated as Maheśvara, Īśvara, Īśa, Prabhu, whereas the supreme form is often given the name of Śiva or Parameśvara, but there certainly are exchanges between these two series – which indirectly points out the fact that it is a question of a sole reality and that every distinction of degree and figure is purely instrumental to the expository requirements.

After leaving the sole force of independent argumentation to work in the first two Adhikāras, Utp. – though in a subordinate way, as may be seen from the disproportion between these sections – allows revelation to take the floor43, being aware that there is a marked limit beyond which experience and reasoning no longer suffice (cf.ĪPV II p. 213). And thus in the first āhnika of the Āgamādhikāra he goes on to describe the principles that constitute reality – from Sadāśiva to the Earth – without expressly mentioning Śiva or Śakti but considering them, or so it appears, as fused in the figure of the supreme knowing subject (III.1.1c), of

43 Utp. does not name any of the scriptural sources of the ĪPK. Abh. in ĪPV II p. 231 expressly mentions the Sadardhasāra (see below p. 193 n. 13, probably the same as the Trikahralaya (or Trikasāra) quoted for the first time by Bhāskara in the Śivasūtravārttika (p. 31) and then, repeatedly, by Abh., Kṣemarāja and Jayaratha (Dvivedi 1983: Upodghāta pp. 39-40). The nature of this text – belonging to the so-called second stratum of the Trika scriptures, but also quoted by Abh. as an authority for the Krama cult of the twelve Kālīs – indirectly confirms Utp.'s position in the Śaiva tradition. According to Jayaratha's explanations (TĀV III p. 194), the Krama doctrine taught in the Devīpañcaśatika, as well as the Trikadarśana, had been handed down to Abh. through the lineage deriving from Somānanda – that is, the Pratyabhijñā). In a definitely Krama context Kṣemarāja (PH p. 20) quotes as an authority a short passage from the lost ĪPr̥tikā, and a Krama flavour can be noticed here and there in the SSĀ (see e.g. chapter VIII). Regarding a passage from the vrtti on ĪPK IV.5, vastutaḥ śaktivikāso viśvam, Abh. names the Śrīmaṅgalāśāstra as being the source; and, in fact, elsewhere (ŚDVr p. 107) Utp. textually quotes the passage of this Tantra which Abh. alluded to. The Maṅgalā (or Sarvamaṅgalā) is included in the list of Bhairavatantra given by the Śrikanthīsamhitā. Abh. (ĪPV II p. 228) also mentions the Raurava (but with reference to a doctrine referred to but not accepted by Utp. ; see also the quotations from the Kālottara in the ŚDVr) and, in particular, the MPĀ, a text belonging to the Śaivasiddhānta but whose peculiar emphasis on the role of knowledge (presented as a more prestigious alternative with respect to ritual) is well-known.

XXX

the Īśitṛ (III.1.2 vrtti). Abh. (IPVV III p. 258) identifies this supreme principle, which is apparently placed at the top of the scale of tattvas and not beyond it, with the śivatattva . By integrating, as seems possible, the scant indications of the ĪPK with what Utp. says in his commentary on the ŚD, one can state it is the śivatattva that presents itself as constituted by all the tattvas (p. 22 sarvattvamayatayā śivatattvāvasthānam). Still according to the ŚDVṛ, it is the supreme Lord Śiva, endowed with the two powers of consciousness and lysis, or light and beatitude (p. 23 svarūpānupravisṭacinnirvṛtirūpaśaktidvayah prakāśānandamayāḥ pūrṇaḥ parameśvaraḥ śivaḥ prasṛtaḥ); he assumes the form of the various principles, beginning with Śakti (ŚD I.29). Cf. ĪPK IV.14ac sarvathā tv antarālīnānantatatvaughanirbharaḥ śivaś cidānandaghanāḥ; ŚDVṛ p. 106 evam suvarṇavat parameśvara eva pūrvoktecchādiśaktimān tathā tattvabhuvanakāryakaranādiprakāravicitryeṇa [...] sthitaḥ; in ŚD III.21d and in the vrtti he is called śivatattva. In passages like these, we seem to witness the overlapping and merging of Śiva seen as the first or last tattva and of the highest divine personality who is beyond the very dimension of tattva – even if supreme – containing and transcending all the tattvas without exception. This impression is reinforced when Utp. passes from the noetic register of the ĪPK to the emotional register of the bhakti in the ŚSĀ.

A great deal of space is devoted to the structure of the limited subject in the second āhnika of the Āgamādhikāra. The divine subject (pati) – whose body is constituted, as it were, by the universe – is counterposed with the 'beast' (paśu), the fettered soul, in its various forms depending on the maculations that characterize it. In the conception outlined by Utp. there are two components from different sources: dhānta44 – and a version of the three maculations which, though deriving from the analogous doctrine of the Śaivasiddhānta, has an utterly monistic qualification. The āṇava, māyīya and kārma maculations45 completely lose their original nature of 'substances' that physically obstruct the self of the paśu from without and resolve themselves into erroneous attitudes of the individual consciousness46. The āṇava maculation, with its

44 Which Abh. was to take up through the MV.

45 On the different denomination of the three maculations in dualist and non-dualist schools cf. Sanderson 1992:286 n.23.

46 Cf. MV I.23cd malam ajñānam icchanti; ŚS I.2 (a)jjñānam bandhah . Cf. TĀ 1.22 ff.

XXXI

obliterating the one or the other of the components of subjectivity (consciousness and freedom, bodha and svātantrya) determines that identity crisis onto which the other two are grafted: the māyic one - which causes the I to see the world of objects as separate from himself - and the karmic one - which makes him consider his own actions as the cause of the series of rebirths, miring him in the saṃsāra. The cause of all three is the power of māyā, which has its roots in the very will of the Lord (III.2.5 vrtti). The individual knowing subject, variously contaminated by these three maculations, is then distinguished according to the four levels which he tends to identify with (void, vital breath, mind and body) and the conditions (avasthā) in which his experience of reality may take place (waking, sleep and deep sleep, corresponding to direct perception, to mental construct and to partial or total suspension of all activity and knowledge). The fourth state, in which duality is overcome, corresponds gradually to the conditions of Vijñānākala, Mantra, Mantreśvara and Īśa; the state beyond the fourth is the one in which every trace of the knowable is dissolved in the absolute I of Śiva.

As we have seen above, the limited subject tends to identify with four aspects progressively higher in rank, which are however still included in the sphere of the objects: the body, the mind, the vital breath and the void. Their predominance causes the eclipse of the two components of his authentic subjectivity, consciousness and freedom (bodha and svātantrya-kartrtā). Liberation, therefore, consists in inverting this relation - namely, in the consciousness-freedom establishing themselves at the centre of the individual, while pushing the components of the fictitious I in the margins47 - and takes on the form of a knowledge that is 'characterized by penetration in it' tatsamāveśalakṣaṇam (ĪPK III.2.12d). Here we have one of the key words of the Kashmiri Tantric tradition, which, furthermore, possesses the particular feature of allowing one to glimpse the succession, or rather, the coexistence of two different conceptions of the individual I's escape from his limits. In fact, in samāveśa 'to enter, to penetrate into' and 'to be penetrated, possessed by' coexist, thanks to the dual transitive and intransitive sense of the root viś-48. In the earlier stages of these schools the term must doubtless have been mainly used to mean the loss of individual identity and the paroxysm of energies which derive from being possessed by the

47 Cf. TĀ I.173cd-174ab and Jayaratha's commentary.

48 Cf. Sanderson 1986: 177 n. 33; on this term see also Hara 1966.

XXXII

terrible divinities linked to the funerary and sexual transgressions, evoked in the Kāpālika rites49. In the later phase represented by the Trika and the elaboration of its teachings, the prevalent use of the term in its active meaning sometimes appears as a conscious declaration of new intentions - even a kind of damnatio memoriae of a past that has become uncomfortable for those aiming at new integrations50. Utp. uses āviś- with the active meaning in the first of the final verses of the ĪPK ('... entering in the Śiva nature ...'). Also in the first verse of the ŚD51 samāviṣṭah is definitely active, but, since it is referred to Śiva, the result is that it is my form that is penetrated by the God. As Utp. explains, it is Śiva who has become one with my form, in the sense that he has entered realities such as the vital breath etc., which in themselves are not different from the other inert things such as the jar and so on, in order to make them into centres of individual conscious life; indeed, the more intensely he penetrates them, the more the extraordinary powers emerge (p. 3 yāvatyā ca mātrāyā samāveśas tāvanmātrasiddhisambhavah). And, therefore, he who seeks these powers must increasingly deepen this 'entry' into himself. However, Utp. concludes, we can imagine that both Śiva and man enter and 'are entered', since the movement towards unity is the same in both. These considerations have left their mark in the way Abh. develops the laconic text of the kārikā, tatsamāveśa°, which is only glossed in the vrtti with tacchaktisamāveśa°. Abh. says (ĪPVV III p. 327) that this word is intended to refer to the two forms of liberation in

49 In the Śaivaparibhāṣā āveśa-samāveśa is considered to be the form of liberation peculiar to the Kāpālikas (Sanderson 1985: 213 n.90, Brunner 1992: 46-47 n.65, Brunner 1986: 518-19). Says Śivāgrayogin (p. 342): kāpālikāḥ samāveśena sāmyam upagacchanti / tathā hi - yathā grahāḥ puruṣeśv āviśanti tatheśvaragunā mukteśv āviśanti; what makes such a conception of liberation unac­ceptable in the eyes of a Śaivasiddhāntin is its dangerous resemblance to possession by evil spirits and the subject's loss of identity and autonomy (p. 346 tathā sati bhītāv iṣṭasyevaśvātan­tryāpattyā puruṣārthatvāyogāt). These themes had already been touched on in the older dualistic Śaiva tradition (cf. Paramoksanirāsakarikā v.13b āveśenāsvatantratā; Rāmakanṭha's țikā p. 286 bhūtāviṣṭasyeva pumso muktāv asvātantryam eveśvarasa­mateti; ibid. p. 283 anyais tu tadā parameśvaragunaih svāśrayasthair eva siddhah samāviśyate grahair iva puruṣa ity āveśapaksah).

50 This process reaches its completion with Kṣemarāja, for whom samāveśa as a technical term seems to have lost any specific identity (PH p. 45 samāveśasa­māpattyādiparyāyaḥ samādhiḥ).

51 asmadrūpaśamāviṣṭah svātmanātmanivārane / śivah karotu nijayā śaktyā tatātmanā //

XXXIII

life. In the one case it has the meaning of 'complete and perfect entry into the own true nature' (satyasvarūpe samyag ā samantāt praveśa-lakṣanam; active meaning) and corresponds to the state beyond the fourth in which the adept, after becoming aware of the supreme nature of the I, becomes as though withdrawn from the knowable which formed his fictitious identity. By contrast, in the other the various components of the levels of the limited subject are gradually penetrated by the elixir of the I, until they become, so to speak, transfigured, removed from their nature of knowable realities. This experience, which corresponds to the fourth state, can be extended further, until it flows into the state beyond the fourth, where the components of limitation, including samskāra, are totally dissolved and incorporated in the I. The passive sense of the second explanation of samāveśa and its relation with the above passage of the ŚDVṛ become evident from what Abh. adds later (ibid. p. 330), this time apparently referring to tacchaktisamāveśa° in the vrtti: tasya samviddharmāḥ śaktiviseṣāḥ samyag āviśanto vibhūtīr utthāpayanti (cfr. above n. 49)52.

The fourth state and the state beyond the fourth can be attained through the practices centred on the vital breaths, described in the Āgamas. Utp. briefly touches on this point at the end of the Āgamādhikāra. The duality of the prāṇa and apāna that characterizes the waking and dream states is overcome with the entry into the samāna, which suspends and unifies them, corresponding to the state of deep sleep. The subsequent level, represented by the udāna, marks the end of the limited condition, which culminates - with the all-pervasive presence of the vyāna - in the supreme condition of Śiva.

52 Therefore, there is a difference between the two turyātītas delineated by Abh. The first pertains to the jñānin, who reaches it instantly thanks to an intense and flashing 'peak experience'; the second pertains to the yogin and is the fruit of the repeated practice of this samāveśa . Here the adept does not halt when his self enters into the nature of Śiva; this is merely his starting point. He slows down this identification and intensifies it at the same time, extending it gradually to all the levels of the fictitious I (body, mind etc.), which the jñānin, on the contrary, only abandoned. This second turyātīta. reached in two stages - says Abh., following Utp.'s tīkā - pertains to the viśeṣadarśanas, namely to the schools that prevalently teach a śāktādvaita, i.e. that stress the 'power' element in the self (p. 331). These are opposed by the sā mānyadarśanas, which place the emphasis on the subject in his transcendence and whose form of turyātīta - the one described first - tends to cancel the knowable and aims for pure consciousness (this is the vyatireka-turyātīta

XXXIV

In the Tattvasamgrahādhikāra the path to liberation passes through the overcoming of the activity which, throughout the ĪPK, is considered as the very core of the condition of limited subject: the vikalpa - the dualizing, discursive thought, that unfolds through antinomies and organizes reality in terms of separation - in all its forms. The two different methods taught by Utp. give the same result - liberation in life - the one by the gradual attenuation and elimination of mental constructs through the cultivation of their opposite (the nirvikalpa), the other by allowing them to flow and progressively purifying their contents. I shall return to these later.

The whole of Utp.'s work, however, is built around teaching the 'recognition' of the Lord, which also the two above-mentioned forms are part of (ĪPVV III p. 354 pañcabhih [the last five stanzas] pratya- bhijñātmakam moksatattvam). This new and easy way is merely the triggering in the devout of an act of identification, which does not reveal anything new but only rends the veils that hid the I from himself; a cognition is not created but only the blur that prevented its use, its entering into life, is instantly removed. The way by which the master creates the premises for this to occur may, on the contrary, be gradual; this is what Utp. does with his work, which aims through a series of arguments at bringing to light the powers of the I and those of the Lord, until identification is triggered. The practice of such a linear (avakra) path is enough to enter into the nature of Śiva and achieve the condition of liberated in life, which may also be accompanied by the extraordinary powers (ĪPVV I p. 111 īśvarapratyabhijñākṛto 'bhyudayo jīvanmuk- tivibhūtilakṣaṇaḥ). This 'penetration' still presupposes a body and, in any case, a state in which the distinction between the I and Śiva has not yet been dissolved53. This occurs within everyday reality just as it is. The light of liberation does not cause its colours to fade, does not cover them but brightens them, performing the miracle of eliminating otherness whilst maintaining the richness of individual flavours.

Says Utp.:

"Dwelling in the midst of the sea of the supreme ambrosia, with my mind immersed solely in the worship of You, may I attend to all the

to which the avyatireka- of the viśeṣadarśanas is opposed). Utp. and Abh.'s inclinations lie towards the latter.

53 ĪPV II p. 258 dehapāte tu parameśvara evaikarasah iti kah kutra katham samāveśah.

XXXV

common occupations of man, savouring the ineffable in every thing." (ŚSĀ XVIII.13)

"May my desire for the objects of the senses be intense, O Blessed One, like that of all other men, but may I see them as though they were my own body, with the thought of differentiation gone". (VIII.3)

"May I aspire to liberation by worshipping You, without withdrawing from experience and the world, and without even seeking the dominion [of the extraordinary powers], but becoming intoxicated with the abundant liquor of devotion". (XV.4)

The householders of Kashmir, whose desire for an intimate experience of communion with the absolute, though profound, withdrew before the abysses of transgression and loss of identity, were not to remain indifferent to the insinuating message of the Pratyabhijñā.

THE FUTURE OF THE PRATYABHIJÑĀ

If Utp.'s teaching did not remain an isolated, though brilliant, fruit of the Śaiva tradition, this is probably due to Abh., who took it to constitute the theoretical and speculative side of his great synthesis of the principal currents of non-dualist Kashmir Shaivism. The privileged position that the Pratyabhijñā occupies in the paddhati of the Trika (in the broad sense), not only in the treatment of the bimbapratimbavāda in the third āhnika of the TĀ, but in the work as a whole - as well as, in various ways, in the MVV, the PTV, the Paryantapañcāśikā etc. - was so to speak the sounding board that amplified and spread its teaching. It is virtually impossible not to find traces of the characteristic themes of the Pratyabhijñā in the Śaiva authors who came after Abh., as well as in later scriptures such as those of the Tripurā (Yoginīhrdaya, Tripurārahasya etc.). The doctrine of the ābhāsas, consciousness as the synthesis of prakāśa-vimarśa, ahamtā, camatkāra, the four levels of the limited subjectivity occupy - either isolated or combined - an important place even in some Vaiṣṇava scriptures, like the Ahībudhnyasamhitā and the Lakṣmītantra. Nor must one think that this influence was limited to Kashmir, since it was very strong also in south India, where the Trika, Krama and Tripurā schools were widespread (Sanderson 1990:80-82). This is also evident from the inclusion of the Pratyabhijñā in the SDS of the Southern Indian Mādhava (14th c.) and the presence of MSS of works by Utp. and

XXXVI

Somānanda54 in southern scripts (for example, it is to be noted that the only complete MS of the ĪPvrtti is in Malayalam characters).

The doctrine of the upāyas, stated briefly and as though in passing in the MV (II.20-23), is taken up again with great emphasis by Abh. and comes to constitute one of the main structures of the Trika, serving as a device to incorporate diverse spiritual traditions in this great synthesis, without levelling away their individuality. I will spare the reader yet another exposition of the doctrine of the upāyas55 (just as I spared him that of the tattvas) and confine myself to some considerations particularly on the first two: the non-means (anupāya) and the divine means (śāmbhava). Firstly, we must remember that, according to Abh., there is no real difference between anupāya and śāmbhava – the former only representing the higher plane of the latter56 – and that the śāmbhava is the sole 'direct' means (sāksāt, TĀ I.142a), whereas, of the two others, the ānava is only a means of access to the śākta and the śākta to the śāmbhava (142b and Jayaratha's commentary). In the non-means57 the adept immediately establishes himself at the centre of the supreme consciousness without the need of a jumping off point. The teaching of a master pronounced once only, or a series of arguments, or faith in the scriptures are sufficient (TĀ II.49ab). Even if these causes may be endowed with succession, the phenomenon they trigger is not; in fact, in particularly privileged cases, it may arise by itself, without any of these detonators. The intensity of this state of penetration is such that even a mere chance encounter with a subject who has entered it may in certain cases determine instantaneous liberation (II.40). Everything appears dis-

54 Which shows that the early Pratyabhijñā texts continued to be studied also independently of Abh.'s interpretation.

55 One of the most concise and clear enunciations is to be found in the Tantroccaya: (anupāya) p. 161 tatra sa parameśvaraprakāśah kasyacid avirataprabandhatayā punar upāyanirapekṣatayā bhāti ekavāram guruvacanam īdrśam samyag avadhārya; (śāmbhava) p. 163 kasyacit tu parameśvaraprakāśas tadīyāyām svā-tantryaśaktau nirjñātāyām prakāśate [...] ahamsvātantryaśaktibhasitaviśvabhāsah parameśvaro vibhur eko nitya iti muhur muhur bhāvayan sa eva; yasya tv ittham apy avikalpakah samāveśo nodeti sa savikalpabhāvanayā sat-tarkarūpayā samkurute vikalpam; (ānava) p. 167 yas tu sattvarkarena na nirmalasvabhāvam āsādayati tasya dhyānādikam ucyate.

56 TĀV I p. 182 sa eva parām kāsthām prāptaś cānupāya ity ucyate.

57 The expression 'non-means' is not to be taken literally; it alludes to the far more subtle nature of this means – which, however, remains such – with respect to the others (cf. TĀ II.2-3 and Jayaratha's commentary).

XXXVII

solved in the Bhairavic fire of consciousness to him who has entered this state (II.35).

While this kind of samāveśa58 concerns the innermost aspect of consciousness, in the divine means the stress falls on its power of Freedom (TSā p. 10, Tantroccaya p. 163), or its manifesting as the universe. Here there is indeed a means-end relation, though in a very subtle form. The adept can start from any aspect of the knowable, taken in its dimension that is closest to consciousness and not yet differentiated from it, and see it as an image reflected in a mirror - distinct but inseparable from its support. This experience, strictly devoid of discursive representations characterized by the scission between subject and object, rests on the Lord's power of Volition, in which the manifestation is still completely enclosed within the desire to create: "All is born of me, is reflected in me, is inseparable from me" (TĀ III.280ab). Unlike the anupāya where everything happens in an instant, once and for ever, this experience is to be repeated essentially the same and progressively extended to all the aspects of reality (TĀ III.274 ff.). Through repetition and succession (TĀV II p. 1 śāmbhavopāyam kramaprāptam, TĀ III.271a bhūyo bhūyah, Tantroccaya p. 163 muhur muhuh) the adept reaches liberation in life, the fourth state. One can only speak of liberation in life with respect to the divine means; in the anupāya it cannot even be said that one is somehow liberated from something; in it, means and end coincide (TĀ III.272-73).

The Pratyabhijñā is traditionally connected with the anupāya. Abh. does not say this explicitly; he just refers, towards the end of the treatment of the anupāya, to two authorities: the Ūrmimahāśāstra and masters such as Somānanda etc. Regarding the latter, Jayaratha's commentary only mentions the ŚD and quotes the well-known passages VII.101cd and VII.5-6, the second of which, moreover, also appear through in the final stanza of Āhnika II of the TĀ. In fact, it is Somānanda who expressly mentions ekavaram (VII.5c), which constitutes one of the essential features of the anupāya (which always takes place sakrd 'once and for ever'). The central teaching of the Pratyabhijñā seems less extreme: we might say a śāmbhavopāya in the broad sense which leaves room within it for both experiences - sovereignty

58 Even though the modern authors sometimes tend to use upāya and samāveśa as synonyms, an obvious distinction between them remains (cf. TSā p. 9 pārameśvara samāveśo nirupāyaka eva, etc.).

XXXVIII

and complete dissolution in the I, turya and turyātīta, seen as forms of liberation in life. The end of the Pratyabhijñā is, indeed, jīvanmukti(ĪPV I on IV.16 etatpariśīlanena śivatāveśāj jīvann eva mukto bhavati) - a condition which Abh. does not connect with the anupāya but with the śāmbhava (TĀ III.272). The Pratyabhijñā furnishes a series of preliminary arguments to the act of recognition, which, though with a different intensity, is at the basis of both the anupāya and the śambhava. In fact, also the anupāya requires 'purification through the composition of a series of arguments' (TĀ II.49a yuktipracayaracanonmārjanavaśāt)59; and, in the ambit of the śāmbhavopāya, Jayaratha (TĀV I pp. 239-40 on TĀ I.216) notes that it is only thanks to the teaching that leads to the recognition of the true nature of the self that these practices can be productive, while those same states exempt from discursive representations which from time to time are spontaneously produced in the unaware and fettered subject (paśu) remain sterile.

Also the practice outlined in ĪPK IV.11, with its emphasis on vikalpana and krama60, is typically śāmbhava. Kṣemarāja in PH p. 40 makes this clearly understood by quoting this passage to illustrate an experience he defines vikalpam akiñciccintakatvena praśamayan, which echoes the definition of śāmbhavasamāveśa in MV II.21.

But in the ĪPK other means are also taught. If we want to read it in the light of Abh.'s systemization, we will have to classify as ānava the practices and experiences centring on the vital breaths hinted at in III.2.19-20) and recognize in the experience described in IV.12 - 'He who, having all as his essence, thus knows: 'All this multiform deployment is mine', he, even in the flow of mental constructs, attains the state of Maheśa" - the elements that characterize the śāktopāya, i.e. the progressive purification of discursive representation and its transformation of the identity of the individual self and of the Lord, attained through the demonstration of his own theses and the invalidation of opposing ones, and therefore precisely to the ĪPK. The later Śaiva tradition was to connect the name of Utp. with the work of constituting these logical arguments; cf. Mahāyaprakāśa (Trivandrum) III.64b yuktitattvavicakṣaṇaiḥ.

59 Abh. is referring here, according to Jayaratha (TĀV I p. 41), to the recognition of the identity of the individual self and of the Lord, attained through the demonstration of his own theses and the invalidation of opposing ones, and therefore precisely to the ĪPK. The later Śaiva tradition was to connect the name of Utp. with the work of constituting these logical arguments; cf. Mahāyaprakāśa (Trivandrum) III.64b yuktitattvavicakṣaṇaiḥ.

60 This gradual process of attenuation and elimination of the vikalpas - explains the vrtti - is grafted onto an experience in which the creation of the Lord appears in all clarity and devoid of vikalpas, since the idea of differentiation does not arise (bhedānudayād vikalpojjhitā spaṣṭāvabhāsā ca; cf. ĪPV I pp. 34-35). Thanks to this practice the adept becomes anavaratanirvikalpakasamvedanaratnamālābhūṣitahrdayah (ĪPVV III p. 389).

XXXIX

mation into pure reflective awareness61 (Śivopādhyāya is explicit in con-

cluding his commentary on VBh 109, p. 95, where this verse is quoted

as an authority: iyam śaktī bhūḥ).

The teaching of the Pratyabhijñā, in both its aspects of philosophical

elaboration and spiritual achievement, was transmitted by Abh. and his

disciple Kṣemarāja, thus influencing the diverse currents of Tantric

Saivism which, either explicitly or implicitly, were to tend to model their

theoretical structure upon it.

The case of Kṣemarāja deserves special mention. With the PH he

composes a kind of catechism, of Pratyabhijñā in nuce, addressing the

many who would have been unable to deal with the dialectical subtleties

of the major work (the ĪPK and its commentaries). In the PH - a work

perhaps modest in itself, but important because of its wide circulation

and the influence it exerted especially over the Śaiva-śākta schools of

South India (cf. Maheśvarānanda; the texts connected with the Tripurā

school etc.62) - we witness a kind of apotheosis of the Pratyabhijñā:

from a single school accepted and valued within the great synthesis of

Abh.'s Trika it ends up by constituting the very framework, in which

Kṣemarāja also inserts teachings from other schools, especially the

Krama63.

The Īśvarapratyabhijñākārikā and its commentaries

The ĪPK - which made the name of Utp. famous - must have been

an early work, given the fact that all his other philosophical writings

presuppose it and contain specific references to it.

His other works include the following. ŚDVṛ, named Padasaṅgati in

the third introductory verse (Maheśvarānanda calls it by this name in the

MMP), which has come down to us only up to IV.74. ĪS and vrtti: a

demonstration of the existence of the Lord written from a Naiyāyika

viewpoint, arguing against Mīmāṃsā, Sāṃkhya and Buddhists. SS and

vrtti: the exposition of his own conception of relation against the oppos-

61 Cf. the vrtti: parāmarśamātrān aśeṣān vikalpān sampādayatah. See also the

commentaries on this verse by Abh. (ĪPVV III p. 393 vikalpā avikalpā bhavanti)

and by Kṣemarāja (PH p. 28 śuddhavikalpaśaktim ullāsayanti).

62 The Yoginīhrdaya-Setubandha and Lakṣmīrāma's commentary on the

Parātriṃśikā call it Saktisūtra.

63 On the possible connection of Utp. with the Krama see above p. XXX n. 43.

XL

ing theories, especially the Buddhist ones (see below pp. 95 ff.). APS: restatement of his conception of subjectivity and its relation with object, first expressed in the ĪPK64. The ŚSĀ deserves a separate mention65: twenty mystical hymns composed by Utp. at different times during his life and later collected and 'edited' by his disciples66. In them, moments of exaltation alternate with moments of discouragement, enriching and cracking the mask of the impeccable theologian of the major works67.

Utp. devoted two commentaries to the ĪPK, a short one (vrtti) - edited and translated here - and a long one (vivrti or tīkā)68, of which only fragments have come down to us (Torella 1988). Neither of these is really and truly a word for word commentary. The vrtti confines itself to presenting the content of the stanzas in another form and briefly pointing out its implications; the vivrti takes the kārikās and the vrtti as its starting point and often develops into far-reaching excursuses. The link between the sūtras and the vrtti is a particularly close one. Despite the fact that they are presented as being artificially differentiated, they substantially constitute a single work, since - according to Abh. (ĪPVV I pp. 16, 183) - they were composed at the same time. Proof of this lies in the fact that the namaskāra, present in the sūtras, is not repeated at the beginning of the vrtti, whereas it is in the tīkā (ibid. pp. 2-3). This makes the vrtti specially important for grasping the 'original' meaning of

64 The commentary that Utp. certainly composed on this work (see MMP p. 133) has not come down to us.

65 Śāstraparāmarśa v. 8 (cit. Pandey 1963: 163 n. 3) "There are, it is true, everywhere thousands of rivers of beautiful verses, but none of them resembles the divine river of the Stotrāvalī".

66 Two other verses attributed to him, but not included in the ŚSĀ, are quoted in the Mahādayaprakāśa of Śitikaṇṭha p. 12 (yat kimcid eva bhūtānām bhāvada-ranam prati ! na kimcid eva bhaktānām bhavadāvaranam !!; sarvasamvin-nadībhe-dābhinna-viśrāntimūrtaye ! namah pramātṛ-vapuṣe śivacaitanyasindhave !!, see also Bhāvopahāra-vivarana p. 43); Yoginīhrdaya-dīpikā p. 287, after quoting the latter verse as belonging to the Parāpañcāśikā, quotes another one (yatra yatra militā marīcayah ! tatra tatra vibhur eva jrmbhate !!) and attributes it to the Stotrāvalī (is it that by Utp. ?).

67 The MS given as 'Mantrasāra by Utpaladeva' in the Bhandarkar Oriental Research Institute Library, indicated by Rastogi (1979:134) as a possible unpublished work of Utp., actually contains the PTLV.of Abh., whose second verse says, in fact, sadāgamapravāheṣu bahudhā vyākrtam trikam / śrīmad-utpaladevīyam tantrasāram tu varnyate !!.

68 Its length must have been equivalent to 6000 or 8000 ślokas (Torella 1988: 141).

XLI

the stanzas. In the tīkā, which was composed later on, we already find multiple interpretations of the same verse, all considered equally possible but evidently being the outcome of further reflections.

The kārikās, the vrtti and the tīkā represent three distinct and relatively autonomous stages in the exposition of the Pratyabhijñā doctrine and, accordingly, it is as though Utp. has taken three different roles (bhūmikā) - kārikā° or sūtrakāra, vrttikāra and tīkāakāra (ĪPVV I pp. 2-3) - almost three different persons, capable of dialoguing and disputing with each other, and yet remaining within the higher unity of the granthakāra or śāstrakāra.

The kārikās, the vrtti and the tīkā are seen by Utp. in the tīkā - as we can gather from the ĪPVV - as respectively linked to the three planes of Paśyantī, Madhyamā and Vaikharī on which the Supreme Word (Parāvāc) is articulated.

On the supreme plane the essence of the doctrine still lies in the consciousness of its author Utp., undivided from Śiva, and it is given the name of 'heart', precisely because of its quintessential nature.

On the Paśyantī plane it takes the form of sūtra, insofar as a differentiation begins to be outlined (āsūtranā), which is necessary if it is to become an object to be taught and explained to other men.

Then, on the Madhyamā plane, the essence is disclosed (unmīlita) in the vrtti and finally, on the Vaikharī plane - which corresponds to the vivṛti -, having now become clear through the refutation of the opponents' possible objections, it is fully differentiated and thus accessible to the comprehension of all (ĪPVV I p. 16).

To this distinction between the planes of the doctrine corresponds the diversity of levels in the recipients of the teaching.

Both the sūtras and the vrtti are aimed at those who are not impregnated with the latent traces of their earlier adhesion to other doctrines.

The study of the sūtras-vṛtti as a whole prepares them for the subsequent phase which consists in the study of the sūtras alone, independently of the vrtti.

On the contrary, those who are under the influence of erroneous conceptions deriving from the profession of different doctrines, are obliged to follow the whole course beginning with the tīkā.

When the tīkā has removed these errors, they can have access to the sūtras accompanied by the vrtti, and subsequently to the sūtras taken alone, until they reach the heart of the doctrine in identity with the consciousness of the śāstrakāra (ibid.).

Abh. devoted two of his most important works to the doctrines of Utp.

The one (Vimarśinī) is a commentary on the stanzas alone, the other (Vivṛtivimarśinī) a commentary on the vivṛti or tīkā, which, as we have seen, was in its turn a commentary of the complex kārikās-vṛtti.

No commentaries predating Abh.'s have come down to us, even though an XLII

exegetic tradition must already have existed (cf. ĪPVV II p. 210), which was probably only oral. From Abh.'s sporadic annotations we come to know that the texts of the Pratyabhijñā69 were already beginning in his time to present some doubtful readings and variants for some points were already circulating.

Information on the chronological order of Abh.'s two commentaries is contradictory. What would appear to be the more natural sequence - first the lengthy commentary on the ṭīkā and then a reduction of the vast and varied material to a smaller size and a more concise and incisive form - seems to be confirmed by what Bhāskara says at the beginning of his sub-commentary (I p. 3)70. But Bhāskara has not proved to be a reliable source in this respect, since he later (Bh I p. 316) does not hesitate to invert the order71. However, a passage from the ĪPVV, which unequivocally mentions the Vimarśinī72, is decisive on this point.

Abh.'s contribution to the Pratyabhijñā - in the light of what can be inferred from the hints at and quotations of the ṭīkā contained in the ĪPVV and other passages of the ṭīkā which have been handed down to us in various ways - appears far less original than is generally thought. Despite the inevitable difficulty in reading a text like the ĪPVV, which is an extensive and diffuse commentary on a work that has not come down to us, it seems clear that the majority of the themes and subjects that Abh. touches on find their direct correspondence - or at least their starting point - in Utp.'s ṭīkā. In fact, the ĪPV itself, which according to Abh. is intended to be a commentary on what the kārikās are in themselves73, accomplishes its task through a carefully gauged and considered systematization of a rich speculative material whose early origin is to be glimpsed in the ṭīkā. In his interpretation of the stanzas Abh. essentially echoes the interpretation or interpretations of Utp., sometimes enriching them with new arguments. He only diverges from Utp. in four instances vyākhyātā.

69 The later authors often refer to one or the other of the three texts (ĪPK, vrtti and vivrti) only by the mention 'iti pratyabhijñāyāmi'.

70 On this basis Pandey 1963: 32 maintains the priority of the ĪPVV.

71 The first passage, too, was incorrect in considering the ŚDVṛ as the first of Utp.'s works, whereas, on the contrary, it often refers to the ĪPK and its commentaries.

72 III p. 230 anenaiva āśayena asmābhiḥ sūtravimarśinyām ekarasatvena idam sūtram vyākhyātam, iha tu yāvadgati vicārayitum ittham vibhāgena apeksā vyākhyātā.

73 ĪPV I p. 39 asmākam tu sūtravyākhyāna eva udyamamah.

XLIII

(I.3.4b bhrānter arthasthitih katham; I.4.4ab naiva hy anubhavo bhāti smratau pūrvo 'rthavat prthak; I.7.1c akramānantacidrūpā; II.4.10d sthi-rasvarthakriyākaram) – in the last instance this is explicitly declared; see p. 179 n. 14 - but in none of them is the overall thread of discourse significantly altered. Discrepancies are sometimes to be found between the text of the kārikās and vrtti as has been established here on the basis of

the MS tradition and Abh.'s quotations from it. This is often due to the fact that these are not actual quotations, but paraphrases, which sometimes expand and sometimes contract the text; at other times, it is a question of errors that have crept into the MS tradition of the ĪPVV or of errors already present in the text of the vrtti that Abh. had before him (see the Apparatus).

That the ĪPV of Abh. ended up by eclipsing all the other commentaries - including those by Utp. himself74 - and that it has become established as the standard commentary, thanks both to its intrinsic quality and to its author's great prestige, may also be seen from the fact that it is the only one for which sub-commentaries were composed, such as the vyākhyā of Bhāskarakanṭha or the anonymous commentary in the Madras Government Oriental MSS Library75, and others only extant in fragmentary MSS, like the ĪPV-vyākhyā in Trivandrum Library (No.15413C) or the ĪP-anvayadīpikā (by one Sādhānanda?) in Mysore Library (No.B.187 of the old Catalogue)76.

74 This is proved by the scant MS tradition of these works (only one complete MS of the vrtti and a short fragment of the țikā have come down to us) and the very few quotations from them. While Kṣemarāja and Jayaratha (obviously, besides Abh.) show their knowing the țikā, also independently of the ĪPVV, the only mentions of the vrtti I am aware of are in late works and by authors belonging to South India (Maheśvarānanda and Kṛṣṇadāsa; see text notes p. 1 n. 9 and p. 63 n. 218.

75 I have learnt from N. Rastogi that the cdition of this text was prepared long ago by K.C.Pandey and himself, but its publication has been as yet stopped by problems concerning precisely this joint authorship.

76 There are also at least two other (late) commentaries on the kārikās alone, one by Nāgānanda in the Adyar Library (No.28. B. 24, mentioned in Raghavan 1981: 31) and the other by the Kashmiriian Bhațțāraka Sundara (ĪPKaumudī; Research Department Library, Srinagar, No.1083). The colophon of a MS of the vrtti (sec below p. LIII) mentions Bhațțāraka Sundara as he who would have composed a commentary to take the place of the final part of Utp.'s vrtti (missing in all the Kashmiriian MSS). During my last visit to Srinagar, I tried - without success - to see the MS of the Kaumudī, possibly misplaced.

XLIV

Bibliographic Note

The ĪPK with Utp.'s vṛtti has been edited in the KSTS (see Bibliography); the text of the vṛtti, corrupt in many points, breaks off at the beginning of III.2.9. The text of the kārikās is obviously included also in the two editions of the ĪPV77 and in that of the ĪPVV.

The kārikās have been integrally translated into English (twice: Pandey 1954, together with the ĪPV; Kaw 1975-76); partially - Jñānādhikāra and āhnika I of the Kriyādhikāra - into German (Frauwallner 1962). The vṛtti has never been translated.

77 If I chose to quote the ĪPV according to the edition which also contains Bhāskara's sub-commentary, it was not because of my considering it the more correct (it is rather the opposite!), but only in order to be able to give a single reference for both texts.

XLV

ABOUT THIS EDITION

This edition is based on nine MSS (and the comparison with the in-complete text edited in the KSTS). Of these, eight are in Śāradā script and are indicated with the letters from Gh to T (the MSS K Kh G are those used for the KSTS ed.; the reference to them is limited to the readings quoted in the foot-notes of that edition). The ninth MS, marked with the siglum T, comes from Kerala and is in Malayalam characters.

A first division is to be made between the MS T - on the one hand - which is the only one complete and generally gives better readings, and the Śāradā MSS - on the other - which all break off at the same point (beginning of the vṛtti on III.2.9), have lacunae and corrupted passages and, in general, less satisfactory readings (cf. text notes Jñānādhikāra 35, 41, 59, 72, 92, 279, 293, 371, 384, 388, 381, 408, 417, 461; Kriyā° 8, 89, 148, 151, 155, 167, 170, 181, 186; Āgama° 9, 39, 46, 64, 74, 77; etc.).

Moreover, within the Śāradā MSS (dating from XVII to XIX century) we can distinguish between a very homogeneous group of seven - Gh Ñ C Ch Jh Ñ T (to which we may add K Kh G) - and J. This MS, though sharing many of the features common to the seven others, is the only one that shows a considerable amount of readings identical with, or sometimes similar to, those only found in T (cf. text notes Jñāna° 53, 105, 237, 242, 263, 281, 322, 383, 404, 427, 432, 444, 445, 449; Kriyā° 21, 43, 70, 87, 116, 120, 130, 159, 165, 218; etc.).

Already on the basis of these first considerations it is possible to attempt to outline a stemma codicum, bearing however in mind the factors of uncertainty determined by the peculiar features of the Indian MS tradition, and particularly of the Kashmiri one. The transmission of texts in Śāradā script (as pointed out by Bühler 1877) is normally the work not of simple scribes or devotees, unaware of the content of what they are copying, but of learned pandits, of scholars, who mostly do not limit themselves to passively copying the MSS, but compare them with others, record various readings in the margin, choose between alternative readings found in the excemplar, make - explicitly or not - emendations of their own. Moreover, in the case in question, they often have before

XLVI

them - besides other MSS of the vṛtti - also other commentaries on the

ĪPK, first of all the ĪPV which seems to be felt as the standard one, as

the inclusion in the text of glosses drawn from it and sometimes the par-

tial or complete reproduction of its colophons shows (see below)1. In

spite of these difficulties, met with in any contaminated tradition, the

picture that comes out is relatively clear, at least in outline.

We start from a work which probably did not have a large circula-

tion and was reproduced in few copies. The scrutiny of the extant MSS

permits us to assume an archetype [a] in Śāradā characters, complete,

which however some corruptions have already crept into (at least those

indicated in Jñānādhikāra text notes 92 and 167, Kriyā° 155 and 205)2.

This MS, or a faithful apograph of it, was brought from Kashmir to

1 N even reproduces large portions of the ĪPV in the margin and between the

lines. In other cases (Kh G) it is the vṛtti that may be found in the margin of a MS

of the ĪPV.

2 Abh. in the ĪPVV had already noticed various readings in the MSS of the vṛtti

at his disposal.

XLVII

Kerala, probably around XII-XIII century when the Trika school spread over South India. T derives from it, although not directly. The exemplar of T (which belongs to XVIII century) was already in Malayalam script, for some mistakes in T can be explained as misreadings of Malayalam characters. The good quality of T is a further example of the high standard of the Kerala MS tradition. From [a] at least another MS also derives, which was probably the only one that later survived Muslim invasion, although with the loss of the last portion; we shall refer to it as the hyparchetype [b]. From [b], incomplete and with a certain number of corruptions but still relatively correct, derive J and the hyparchetype [c].

In this manner it is possible to explain, on the one hand, the many readings that J, alone, shares with T and, on the other, the still more substantial affinity of J with the other Śāradā MSS. As regards the latter, which constitute a very compact family, we may assume a common ancestor, i.e. the hyparchetype [c], which further corruptions have crept into. Besides these corruptions shared by the whole family, there are some others that are peculiar to single MSS or groups of them3. Probably, it would be possible to go farther on and identify genetic connections within the family K Kh G Gh Ṅ C Ch Jh Ṅ T, too. But, at this point, the contamination of the MSS and the scribes' work of collation and emendation4 makes this attempt rather complicated and its result, anyhow, doubtful. These considerations and, above all, the awareness that such further investigation is bound not to have any direct bearing on the establishment of the text have induced me to refrain from it.

For this edition the testimony of ĪPVṛV, which sometimes quotes literally, sometimes paraphrases or simply gives the contents of the vrtti, has been largely made use of. Since it mostly proves invaluable in confirming or discarding a reading, choosing among equivalent ones, or even helping to restore passages corrupt in all MSS, I have decided to include it in the critical apparatus on a par with the MSS, though being well aware that in so doing I have departed from the established usage of classical philol-

3 A passage from the Jñānādhikāra seems particularly fit to illustrate the progressive corruption from [a] to the above group of Śāradā MSS (cf. text note 405): o jñānānubhāvena na bhinnasyātītasya rūpyajñānasyāprāmāṇyam (correct reading. only found in T), o jñānānubhave jñānasyāprāmāṇye sā J, T (C Jh then corrected into the reading J).

4 See, for instance, the isolated concordances of single MSS with correct readings only found in T (T Jñāna° n. 253; C Jñāna° n. 395; Ch Jh Kriyā° n. 118; etc.).

XLVIII

पुरुषमवछमनंछवाराजांभूयपरिबनि

भूयमतयुगपन्मकृतेनभूनेम्भ

भूममभृतःमोचत्रःप्रमात्रन्नीनउ

याकन्डते ॥

प्रतकरपतलभूनाॅमेक्त्रं

सप्तदिनांमा कडकलमु

मित्तुन्डक्रमात्रुप्रभुप्र ॥

कडकल्मानतवक्त्रपिप्रवक्

त्रथल्नमूरक्रमात्रुप्रथनममब्रचमु

गड्भमत्रिला:प्रवकरपत्ल्मुः

क्रमिभवुविभधमात्रल्नमपिकीली

नचत्रपक्रमित्रमालः

भूताॅचूर्मुमंखित्रः:मालां

बृक्रमभूवे पचतुर्थमफ्रवे

मालनेॅमवनाथ्रम) ॥

प्रकारवर्मुमंखेदनमुॅति

मर्वशन्मुखमभरवत्रेक्त्रमभय

प्रमालुॅम्परालन्मभौ

MS Ch, f. 19a

ogy. But I have preferred to put together in a single apparatus what (MSS and testimonia) is in this case equally important for the constitution of the text, so that the reader is not compelled to continually go from one apparatus to the other.

Abh.'s commentary composed around 1015 represents the oldest evidence for our text - older than any extant MS. This constitutes its unique importance; however, I have made a cautious use of it and followed it only when I could find some kind of support in the MSS. For in many a case, even when ĪPVV seems to quote literally, it is apparent from the MSS evidence that Abh. is giving a text either approximately remembered, or reworded, or mixed with his own glosses. In these cases, if all the MSS agree on a different but equally satisfactory (and often more concise) reading, I have accepted the latter.

I have followed the division of the text into adhikāras and āhnikas, as in the ĪPV and ĪPVV, though the MSS vary greatly on this point, some of them even showing: no division at all. I have given the colophons in the simplest form, like that often found in T. There is some doubt as regards the Tattvasamgrahādhikāra, which is taken by the ĪPV as consisting of a single āhnika5 whilst the ĪPVV divides its commentary on the adhikāra into two vimarsas (third and fourth, first and second being devoted to the two āhnikas of the Āgamādhikāra6). In the respective introductions the ĪPVV refers separately to 15 (ĪPVV III p. 355) and 3 ślokas (ibid. p. 400). Considering that no MS of the vrtti shows a division into two āhnikas7 and that this division might have been introduced by the tīkā - which the ĪPVV directly refers to -, I have taken this adhikāra as a unitary section.

In the text, < > angle brackets enclose what has been added to the transmitted text by conjecture; [ ] square brackets indicate editorial deletions. No mark has been used for conjectural emendations of transmitted readings; they are acknowledged as such and accounted for in the Apparatus. Sandhi has been normalized8.

5 ĪPV II p. 280 [...] āgamārthasamgraham ślokāṣṭadaśakena darśayati "svātmaiva sarvajantūnām" ityādinā "utpalenopapāditā" ity antenaikenāhnikena.

6 Of the MSS, only J T correctly divide the Āgamādhikāra into two āhnikas.

7 In fact, it is only C Ñ that mention the name of the Tattvasamgrahādhikāra.

8 I have only occasionally dealt with the peculiar varieties of sandhi, scribal mistakes etc. in the Śāradā and Malayālam MS tradition.

XLIX

The Manuscripts1

Gh

The MS belonged to the personal collection of Pt. Dinanath Shastri,

Srinagar, who, with great kindness and generosity, allowed me to

photograph it. This MS no longer exists, having been destroyed along

with the entire collection in the course of the recent Muslim desorders in

Srinagar (Dinanath Shastri's personal communication).

The MS is on paper, written in Śāradā characters, consisting of 28

folios numbered on the verso from 1 to 28, with 16 lines per page, measur-

ing cm. 21,4 × 17,6. On the first page the vrtti on the first seven kārikās,

omitted in the text, was added in the margin and on a small streap of

paper pasted in the left margin; the hand is apparently the same.

Beginning: om namaḥ sarasvatyai // om namaḥ pratyakṣasva-

rūpāyātmane //

End: kartryoge 'pi bodhānāṃ karmottīrṇānāṃ vidyeśvarākhyānāṃ

bhinnavedyabhāktvena māyāmalam // Then, kārikā III.2.10.

There is only a division into adhikāras; within each of them the

numeration of the kārikās is continuous. Rare corrections in the margin.

N

Research Department Library, Srinagar, No. 824.02.

The MS is on paper, written in Śāradā characters, consisting of 48

folios numbered on the verso from 2 to 49, with 11 lines per page,

1 The MSS K Kh G are described in KSTS Ed. as follows: "(K) Ms. belongs to

Pandit Harabhatta Shastri. On Kashmiri paper in Sharada character. Size 10" × 6". It

is in the Talapatra form. (Kh) and (G) Ms. - These are, in fact, the Manuscripts of

the Ishvarapratyabhijna-Karikas with the Vimarshini and their margins contain the

Vrtti in question. The Ms. (Kh) belongs to Rajanaka Maheshvara. In Sharada charac-

ter, on old Kshmiri paper. Correct. Size ¼ foolscap, paper-bound. The (G) Ms. - Got

on loan from Pandit Rama Kokila, a well known Pandit of Srinagar. Size ¼ foolscap,

bound in leather, almost without mistakes, on old Kashmiri paper, in beatiful Shara-

da character." (p. 3). The printed edition is referred to in the Apparatus as E.

L

measuring cm. 18,7 × 14,1. It has been profusely annotated by a different hand with single glosses or long passages (mostly drawn from the ĪPV) between the lines and often entirely covering the margins. There is only a division into adhikāras; within each of them the numeration of the kārikās is continuous. Between ff. 35 and 40 there are a few pages without glosses. A passage based on the ĪPV is inserted in the text between f. 2a line 8 and f. 2b line 6.

Beginning: (The first page is missing) oheśvare / ajadātmā ... (kār. 1.1.2).

End: same passage as in Gh; then the remaining kārikās without commentary. Finally: samāpteyam pratyabhijñā / śubham astu sarvajagatām.

C

Research Department Library, Srinagar, No. 892.

The MS is on paper, written in Śāradā characters, consisting of 23 folios numbered on the verso from 1 to 23, mostly with 21 lines per page (sometimes 17-19), measuring cm. 25,6 × 17,1. Sporadic corrections and glosses by other hands in the margins. There is only a division into adhikāras; within each of them the numeration of the kārikās is continuous; sometimes a different hand notes the end of the āhnika in the margin.

Beginning: om namah śivāya.

End: same passage as in Gh; then the remaining kārikās without commentary. Finally: tattvasamgrahādhikārah // iti śrīmadutpaladevācāryaviracitā pratyabhijñāvṛttih samāpto (sic) śivāya namah // āditah ślokāni 190.

Ch

Research Department Library, Srinagar, No. 1672.

The MS is on paper, written in Śāradā characters, consisting of 30 folios numbered on the verso from 1 to 22 only, measuring cm. 25,2 × 16,5. It is divided into three portions written by different hands, apparently in different periods.

First part: ff. 1-11a. 13 lines per page. It begins with: om namo gurave / om. Some glosses by different hand in the margin only in few pages. The vṛtti on each kārikā is introduced by atra vṛttih. Division into āhnikas. The colophons are as follows. iti pūrvapakṣavṛttih; paradarśanānupapattih; smrtiśaktitimirāpanāṃ catuṣṭham.

L1

Second part: ff. 11b-22b. 14-18 lines per page, neither glosses nor corrections. F. 13 is missing, lacuna from °śakālasparśāt (vrtti I.5.14) to ātmaśabdāc cāham iti (vrtti I.5.17). Colophons: iti śrīmadācāryābhinavaguptaviraciitāyām pratyabhijñāsūtravimarśinyām jñānaśaktinirūpanam pañcamam āhnikam; iti śrīmadācāryābhinavagupta° etc. apohaśaktinirūpanam ṣaṣṭham āhnikam. The colophons of the six āhnikas are derived from ĪPV; the portions of ślokas found at the beginnning of each āhnika also come from the opening ślokas in the ĪPV. No division between āhnikas VII and VIII.

Third part: ff. 23-31a. 23-28 lines per page. Some glosses in the margins drawn from the ĪPVV. Neither division into āhnikas nor numeration of verses. End: same passage as in Gh; then the remaining kārikās without commentary. Finally: sampūrneyam (sic) pratyabhijñāśāstram.

J

Research Department Library, Srinagar, No. 1662.

The MS is on paper, written in Śāradā characters, consisting of 37 folios numbered on the verso from 1 to 37, with 21 lines per page, measuring cm. 13,4 × 16,9. Corrections in the margins or between the lines, generally tending to replace the readings that J shares with T with those found in the other Śāradā MSS. The numeration of the kārikās is continuous, but there is a division into āhnikas within each adhikāra; the colophons are derived from the ĪPV.

Beginning: om namah śivāya.

End: same passage as in Gh; then, kār. III.2.10.

Jh

Research Department Library, Srinagar, No. 860

The MS is on paper, written in Śāradā characters, consisting of 22 folios numbered on the verso from 1 to 24, with 16-17 lines per page (from f. 13 onwards, 20 lines), measuring cm. 21,9 × 14,8. F. 17 (from II.3.1 idam etādrg ity evam to II.3.7 aviruddhāvabhāsānām eka°) and f. 20 (from II.4.9 °kramena janayed ghaṭam to vrtti II.4.9 tataḥ pari°) are missing. Some folios are misplaced. Notes and corrections by different, much more recent, hands in the margins and between the lines. The numeration of the kārikās is continuous; no division into āhnikas.

Beginning: om namo mahāmāyeśvaryai // om namaḥ śivāya saśivāya śivadāya namo namaḥ.

LII

End: same passage as in Gh; then the remaining kārikās without commentary. Finally: śrīmatsundarabhaṭṭārakeṇa pratyabhijñāvrttih sam-pūrnīkrteti śubham astu devādīnāṃ mānusyakitānām /

Ñ

Research Department Library, Srinagar, No. 2352

The MS is on paper, written in Śāradā characters, consisting of 16 folios numbered on the verso from 1 to 16, with 22 lines per page, measuring cm. 26 × 17,9. Neither division into āhnikas nor numeration of verses. Rare corrections in the margins.

Beginning: om śrīgurur jayati om śrīmān paramaśivo jayatitamām //

End: same passage as in Gh; then the remaining kārikās without commentary. Finally: tattvasamgrahādhikārah iti śrīmadutpaladevācārya-viracitā pratyabhijñāvrttih

T

University of Baroda, Central Library, Sanskrit Division, No. 807.

The MS is on paper, written in Śāradā characters, consisting of 56 folios numbered on the verso from 1 to 55 (no number on the last folio), with 12 lines per page. Neither division into āhnikas nor numeration of verses. Rare corrections by other hand in the margins.

Beginning: om śrīganeśāya namaḥ om namaḥ śivāya (to be noted: f. 14a in the margin: om namo bhagavate vāsudevāya).

End: same passage as in Gh; then the remaining kārikās without commentary. Finally: sampūrṇeyam pratyabhijñāśastram / śubham astu / śivāyom namaḥ.

T

Oriental Research Institute and Manuscripts Library, Trivandrum, No. 8900A.

The MS is in palm-leaves, written in Malayalam characters2, consisting of 21 folios (corresponding to ff. 4-24 of the whole codex), with 11-12

2 P.L. Shaji, MSS Assistant, provided me with a devanāgarī transcript. The collation of the MS, however, is based on my own reading of the Malayalam text. In this connection, I would like to tell a very instructive story, which should caution against

LIII

lines per page, measuring cm. 27,2 × 7,5. According to the information kindly supplied by P.L. Shaji, the MS was collected by Ramasubha Shastri, Kadinamangalam, from Chirakkal Palace Library on 9/9/1941. On paleographic evidence, it is not later than XVII-XVIII century. The ĪPVr begins on f. 4b, 1.2, after an incomplete anonymous work dealing with the theories of error. The palm-leaves have become dark, especially at the edges, from which small fragments are missing here and there; in some places they have been eaten by insects. No numeration of verses; there is a division into āhnikas, with some inconsistencies (prathamam āhnikam; dvitīyam ... etc.; iti jñānādhikārah; navamam āhnikam; daśamam āhnikam; kriyādhikāre trtīyam; dvādaśamam āhnikam; kriyādhikārah; āgamādhikāre prathamāhnikam; āgamādhikārah; neither division into āhnikas, nor mention of the title for the last adhikāra). The MS often has voiceless stops instead of (correct) voiced stops, evidently due to the absence of the latter class of sounds from the scribe's own language (i.e. Malayalam).

Beginning: kathamcid āsādya .... (kār.1.1.1) End: samāpteyam īśvarapratyabhijñāvrttih /3 bhīrūnām abhayaprado bhavabhayākrandāsya hetus tato hrddhāmni prathitabhīravarūcām īśo 'ntakasyāntakah / bheram vāyati yas suyoginivahas tāsya prabhur bhairavo viśvasmin bharanādikrd vijayate vijñānarūpah parah / śubham.

The sigle E in the Apparatus refers to the printed edition.

the widespread practice of working with South Indian MSS using devanāgarī copies. Prof. H. Alper (whose untimely death in 1987 saddened all of us) came across this MS, too. Having realized its importance but not knowing Malayalam script, he asked for and obtained - through the good offices of Prof. K. Kunjunni Raja - a transcript ĪPVr, he was so kind as to send me his transcript. The comparison of the pandit's transcript and mine was startling. The differences were so many as to make me doubt about my actual understanding of Malayalam. Thus, I carefully checked some portions of both transcripts with my photographs of the original MS. It soon became clear that the astute pandit - besides definitely having an imperfect knowledge of Malayalam script - had prepared his transcription keeping the KSTS Ed. before him: sometimes he (badly) copied from the MS, sometimes found it more practical to copy from the printed edition. Also his transcription of the last portion of the vrtti, not available in the KSTS Ed., is very inaccurate and full of mistakes.

3 This verse is quoted in MMP p. 173 and ascribed to Kṣemarāja.

LIV

TEXT

ईश्वरप्रत्यभिज्ञाकारिका: वृत्तिसहिता:

ज्ञानाधिकार:

प्रथममाहिकम्

कथंचिदासाद्य महेश्वरस्य दास्यं जनस्याप्युपकारमिच्छन् । समस्त'संपत्समवाप्तिहेतुं तत्प्रत्यभिज्ञामुपपादयामि ॥ १ ॥

1 sampat omitted in C. 2 parameśvaraprasādādev eva cit. and commented on in ĪPVV I p. 24; cf. also ibid. p. 38 vrttau ca gurubhattārako 'pi parameśvaraśabdena nirdiṣṭah. 3 The compound is partly cit. and commented on ibid. p. 21 vrttau ca sūcito 'yam artho "dāsyalakṣmīḥ" ity anena / dāsyaprabhavā lakṣmyah samastāḥ sampadah parameśvara-prasādāt labdhā yeneti; on the same page cit. (from the țikā) atyantadurlabham dāsyam. 4 aham ekākisampadā lajjamanāno cit. ibid. p. 27. 5 janam apīmam: janasamīpam E Gh Ñ C Ch J Jh Ñ T, which probably derives from IPV I p. 33 upaśabdah samīpārthah, tena janāsya parameśvaradharmasamīpatākarṇam atra phalam (sa and ma are almost identical in śāradā script). 6 sva°: svam T N J. 7 vaksyamānopāyena cit. ibid. p. 32.

1

I 1.1

ṇopāyen' 'pratyabhijñāpayāmi 'yen tasyāpi paramārthalābhena pari- tuṣyeyam् ॥ १ ॥

kartāri jñātari svātman'yādisiddhe maheśvare । ajahatma niṣedhaṃ vā siddhiṃ vā viddhīt kaḥ ॥ २ ॥

'sarveṣāṃ svātmanḥ ''sarvārthasiddhisamāśrayasya tat'tatsarvārthasādha- nānyathānupapatterya 'krodīkṛtasiddheḥ svaprakāśasya pramātrekapuruṣaḥ: 'pūrvasiddhasya ''purāṇasya jñānaṃ kriyā ca । स्वसंवेदनसिद्धम् 'svaphalatvam्, tenaivaṛtmasya siddhā niravaṛtane च ''jādānamatrodyaṃ: ॥ ३ ॥

8pratyabhijñāpayāmi cit. and commented on ibid. pp. 23, 31-2. 9tasyāpi paramārtha°cit. ibid. p. 30; cf. the paraphrase ibid. yena asyāpi lābhena tuṣyeyam; parituṣyeyam cit. and commented on ibid. pp. 33-4. The whole passage is quoted in Kṛṣṇadāsa's Anuttara- trimśikālaghuvṛttivimarśinī (ORI and MSS Library, Trivandrum, MS No. C2108 D, f. 152b). 10sarveṣāṃ svātmanah cit. in ĪPVV I p. 50; svātmanah is indicated as a subjec- tive genitive depending on jñānam kriyā ca (kartari eṣā śasthī / tatkartrkham hi jñānam karanam ca, na tu saṃhandhasasthī). 11sarvārthasiddhi°cit. ibid. p. 49; °samāśrayasyacit. ibid. p. 52 sarve tasmāt pramātur anye ye 'rthāḥ teṣāṃ yā siddhiḥ prakāśamānutā, tasyā yatah sa pramātā nibandhanam viśrāntisthānam vimarśarūpatayā samāśrayah. 12°sarvārtha°omitted in IE. 13krodīkṛtasiddheh cit. and commented on ibid. p. 67 krodīkṛtasiddhikatvam yad uktam vṛttau ādisiddhatvasphuṭika- ranābhiprayena ... tena pramātā svātmano nijasyā rūpasya siddhiḥ prakāśaḥ prameya- siddhyuiva krodīkṛtā kropalaksitahrdayarūpatām gatā kṛtā hrdayangamīkṛtety arthah. 14pūrvasiddhasya cit. ibid. p. 67. 15purāṇasyacommented on (and differentiated from the close pūrvasiddhasya) ibid. p. 51 idantādeśakālasvarūpaniyamārūḍham prathanam ya- sya nīlasukhādikasya tasyā bhāvas tannisedhopalaksitāsun pramātā prakāśate ca prākāśistha ca prakāśisyate ceti yāvat / ata eva purāṇaḥ pūrvasiddhaś ceti na paryāyau, purāṇo 'pi hi na pūrvvasiddho 'pi na purāṇo dhūma iveti ubhayok- tịh. 16T omits eva. 17tenacit. ibid. p. 50. 18jādānam evodyamcit. ibid. p. 49 and commented on p. 78.

2

किंतु मोहवशावस्मिन् हृटेडप्यनुपलक्षिते ।

शक्त्याविष्करणेनेय प्रत्यभिज्ञोपदर्श्यते ॥ ३ ॥

केवलमस्य १९स्वसंवेदनसिद्धस्या २०ईश्वरस्य २१मायाग्यामोहाद् २२अहदयङ्मतवाद् २३असाधारणप्रभावाभिज्ञानरह्यापनेन २४दृढनिश्चय-

रूपं २५प्रत्यभिज्ञानमात्रम् २६उपदर्श्यते ॥ ३ ॥

तथा हि जहभूतानां प्रतिष्ठा जीवदाश्रया ।

ज्ञानं क्रिया च भूतानां जीवतां ज्ञानं

जीवनं मतम् ॥ ४ ॥

२७वस्तूनां २८जडाजडभेदेन २९द्वैविध्यम् । ३०तत्र जडस्वरूपस्य ३१जीवन्निष्ठा सिद्धेः , जीवतां पुनर्जीवत्वं ३२जीवनं ज्ञानक्रिये

एव ॥ ८ ॥

19 svasamvedanasiddhasya cit. and commented on ibid. pp. 79-80 svasamvedana-siddhasya ity api vrtitau gaunam drastavyam sukhādivat anapahnāvanīyatām nirūpayitum.

20 E Gh N C Ch J Jh N T F omit api.

21 Cf. the paraphrase ibid. p. 80 māyākrto vyāmoho māyā'yyāmohah.

22 ahrdayangamatvāt cit. ibid. p. 80, where it is remarked that this causal ablative glosses the locative anupalaksite, which is therefore a nimittasaptamī.

23 The compound is paraphrased and commented on ibid. p. 86 asādhāranam aniśvare yad asambhāvyaṃ jñātvakartṛtvādi, tad eva abhijñānam ekadharmaitmakasvarūpamukhena anantadharmacitravastusvarūprapraveśābhyupāyah, abhitah svarūpasarvasvasvikāreṇa va-stu jñāyate yeneti, tasya khyāpanena prathanāya. This confirms °abhijñāna° T Gh N J Jh T, instead of °abhijñāpana° E C N and °abhijñānākhyānakhyāpanena Ch.

24 Cf. ibid. p. 86 "iyam" iti drdhaśabdena, "pratyabhijñā" iti bhāvasādhanam niscayasabdena.

25 Quo-tas tadvayavahāramātram api neha sādhayate, api tu upāmśu cchannam sthitam sat, upa-samīpe iti hrdayasamgamaneṇa pradhānatāya darśyate ... tena vrttau sūtrānusāreṇaiva "upadarśyate" ity uktam.

27 vastūnāṃ cit. ibid. p. 88.

28 Cit. ibid. p. 95 vrttau ca jadājadabhedapadam.

29 dvaividhyam: dvaividhye C Jh T.

30 tatra cit. ibid. p. 96.

31 jī-van°: jīva° T; cf. ibid. p. 97 "anena" iti jadānām antarāle jivaniṣṭhatāni-rūpanena.

32 Cf. ibid. p. 97 "jīvanasambandhah" ity anena kriyāsambandho 'tra pravrttinimittam iti hrdi nidhāya samāsakṛtaddhiteṣu sambandhābhidhānām iti manyamānah śatrantam jī-vacchabdam sādhayāmāvyuparatajīvanakriyāvadabhidhāyakam sūcayati / jīvatām ity upakramya jīvanam jñānakriye ity ekavacanena samvidaikyam pārārthikam āha. jīva-nam jñānakriye eva cit. ibid. p. 104 E Gh N C Ch J Jh N T invert: jīvanam jīvattvam.

33 Gh N add yo janāti karoti ca jīvatīry ucate, being a gloss drawn from ĪPV I p. 67.

I 1.5

तत्र ज्ञानं स्वतः सिद्धं क्रिया कापाश्रिता सती ।

परैरप्युपलक्ष्येत तयान्यज्ञानमूर्छति ॥ ५ ॥

जीवतां क्रिया कायपरिस्पन्दपर्यन्तीभूतान्यत्रापि प्रत्यक्षा ,

ज्ञानमातमवेद्यं परत्रापि क्रिययैव प्रसिद्धमिति सिद्ध एव

स्वसंवेदनसंवेद्यतया स्वपरयोरेश्वरोऽहं प्रत्येय आत्मा, तस्य

मायाशक्त्या निर्गीतिरोधानाद् एवं विमतिः ॥ ५ ॥

इत्युपोद्घातः ॥

इति ज्ञानाधिकारे प्रथममाह्निकम् ॥

34 tatra: tathā T. 35 Instead of jīvatām ... ātmavedyam T, E Gh Ñ C Ch J Jh Ñ T have a text that is evidently corrupt: tatrāpi pratyakṣāj (pratyaksa° C Ch) jñānam ātma-samvedyam. The text of T is supported by ĪPVV: jīvatām cit. p. 103; kriyā is mentioned p. 104; kāyaparispandaparyantibhūtā appears through the paraphrase p. 104 kāyaspanda-rūpatvam prāptā satī and p. 105 vrttau ca ayam arthah pūrvāvavasthāsambhavasūcakvipratyayaprayogena (referring to °paryantībhūtā) svīkrtā; anyatrāpi can be understood from p. 104 ... yat pramātṛvyatiriktasya pramātrantarasyā darśanasparśanasphuṭa-pratyāvaseyeti. 36 api omitted in Gh. 37 Cit., in a slightly different form, ibid. p. 108 siddhatīty api vrttau svaprakāśatāmśaprādhānyāya. 38 Cf. ibid. p. 108 etac ca vrttau "siddha eva na tu sādhyaḥ" ity anena darśitam (na tu sādhyaḥ being most probably a gloss). 39 °samvedyatā: °siddhatayā E Gh Ñ C Ch J Jh Ñ T, equally good. 40 Cf. the paraphrase ibid. p. 111 tasyā iti karmani ca kartari ca śleṣeṇa śasthī / yah svaparayor ahammatrāpratyayatayā nirbhāsanayogo 'ta eva sa māyāśaktitirohitair mūrkhair vimanyate uktarūpaviruddhena rūpeṇa abhimanyate, sa eva ca īśvaro vipratipadyate tasyāiva kartrtvāt / ubhayatra māyāśaktikrtam āvaraṇāvabhāsanam hetuḥ. °pratyaya: E Gh Ñ C Ch J Jh Ñ T. 42 ātmā: svātmā J. 43 evam vimatih: eva vimatih E, Gh; evam iti N C J Jh Ñ; eva mitih Ch; evam probably cit. ibid. p. 111 and glossed with vaksyamānapūrva-paksarūpatayā. 44 Omitted in E C Ñ; upodghātaḥ is glossed ibid. p. 112.

4

द्वितीयमाह्निकम्

ननु स्वलक्षणाभासं ज्ञानमेकं परं पुनः ।

साभिलापं विकल्पाख्यं बहुधा नापि तदृश्यम् ॥ १ ॥

नित्यस्य कस्यचिद् द्रष्टुस्तस्यात्रानवभासतः ।

अहंप्रतीतिरप्येषा शरीराॅधवसायिनी ॥ २ ॥

ज्ञानमेकं स्फुटा'वभासस्वलक्षणानुभवरूपं 'निर्विकल्पकं ते कथयन्ति । अपरं तु शब्दारूषणया स्मृतिसंशयोत्पेक्षादिबहु भेदं विकल्प'संज्ञम् । 'उभयमप्येतन्न युज्यते 'बोधरूपज्ञानातिरिक्त- स्यान्यस्यै संबन्धतया 'तस्यानुपलब्धेः । कौसौ 'स्थिररूप आत्मा । 'साभिलापाहंप्रत्ययेनापि शरीराॅदिकवेद्यवस्तूत्पर्णो 'वेदयिता न 'कश्चिल्लोकैर वधार्यते ॥ १-२ ॥

45 T adds atha pūrvapakṣaḥ (Ch om pūrvapakṣatayeti). 46 nityasya: anyasya T. 47 °avabhāsa°: °avabhāsam T J, but cf. ibid. p. 118 tata eva sphuṭo deśakālākārani yato 'tra ortho bhāti. 48 nirvikalpakam is probably to be preferred to nirvikalpam T (cf. ibid. p. 116, where it is said that in the țīkā the form vikalpakam is used in order to indicate, by the agent suffix, the autonomy of this function and the fact that it does not require a permanent entity as a subject; however, in the case of nirvikalpaka, the suffix might also be understood as a mere samāsānta). 49 °samjñam: °samjñakam Jh. 50 The quotation iti creates some doubt about the reading etan, on which, however, all the MSS agree. 51 ĪPVV I p. 119 quotes as vrtti only bodhātiriktasya, which, however, might be a deliberate abbreviation by Abhinavagupta. 52 tasya: asya T; tasya cit. and commented on ibid. p. 119 tasya iti ātmano 'nupalabdheḥ, tasyeti vā ubhayasya anyasambandhitayā anu- palabdhirtir vrrttiyojanā (the meaning does not change; a twofold interpretation of the kārikā had already put forward ibid. p. 115). 53 sthirarūpa: sthira E Gh Ñ C Jh Ñ, equally good. 54 Cf. ibid. p. 121 vrttāv api sahaśabdena ahampratīter abhilāpāvyabhi- cāro darśitaḥ and p. 120 nanu ahampratyayaśabde sābhilāpagrahaṇam viśeṣaṇam vrttau kimartham iti. 55 vedayitā na: na vedayitā T Gh Ch J Jh Ñ. 56 kaścit cit. ibid. p. 119; lokair cit. ibid. p. 120

I 2.3

अथानुभवविध्वंसे स्मृतिस्तदनुरोधिनी ।

कथं भवेत् नित्यः स्वादात्मा यधानुभावकः ॥ ३ ॥

⁵⁷स्मृतिकाले ⁵⁸पूर्वानुभवनाशात् कथं ⁵⁹पूर्वार्थानुभाव- ष्टम्भधर्मा ⁶⁰स्मृतिज्जायेत यदि ⁶¹तदापि तदनुभवबोधो ⁶²नानु- वर्तेत, यश्च ⁶³अनेककालस्थायी बोधः स एवात्मानुभविता

॥ ३ ॥

सत्यप्यात्मनि हृड्नाशात्तद्वारा हृ्टवस्तुषु ।

स्मृतिः ⁶⁴केनाथ यत्रैवानुभवस्तत्पदैव सा ॥ ४ ॥

अर्था⁶⁵भासनाशात् तन्मुखा⁶⁶वलम्बनीयो विषयोऽप्यत्र नास्ति स्मृतेरित्येकबोधात्मनि ⁶⁷सत्यप्यात्मनि सा निर्विषयैवेतिन्य- वहारोच्छेदः । असतो⁶⁸प्यर्थानुभवस्य विषयेण सा विषयवती

⁶⁹यदि ॥ ४ ॥

⁵⁷ smrtikāle cit. ibid. p. 122. ⁵⁸ pūrva°: pūrvārtha° T, but cf. the paraphrase ibid. p. 122 anubhave pūrvabhāvini; purvānubhavābhāvāt N. ⁵⁹ E Gh Ñ C Ch J Jh Ñ T omit °artha°, but cf. the analysis of the compound ibid. p. 122 pūrvo 'rthavisayo 'nubhavas tasya yo 'vastambhas tadrūpaparitāgah sa dharmo yasyaḥ. ⁶⁰ smrtir jāyate yadi: smrtau di Ñ. ⁶¹ tadāpi cit. ibid. p. 122 “tadāpi” iti smrtikāle smrtirūpatāyām ity arthah. ⁶² nānuvarteta: nānuvartate Ñ, nāvarteta E C Ñ. ⁶³ aneka°: pūrva° Ñ. ⁶⁴ kenātha: ke- nāpi T. ⁶⁵ °ābhāsa°: °avabhāsa° Gh Ch. ⁶⁶ °avalambanīyo viṣayo 'py artho °ālambanīyo 'py artho E Ñ, °ālambanīyo viṣayo 'py artho Gh N C Ch J Jh T (viṣayo is added in the margin in C and deleted in J). ⁶⁷ saty apy ātmani omitted in Ch. ⁶⁸ 'pi only found in T: asato 'pi cit. ibid. p. 126. Cf. the paraphrase of the whole passage in the țīkā, cit. ibid. p. 126 “katham sa nāśasya viṣayena viṣayavān?” iti. ⁶⁹ yadi cit. ibid. p. 126 yadiśabde-na athaśabdāsya sambhāvanām dyotyarvena āha.

6

यतो हि पूर्वानुभवसंस्कारात् स्मृतिसंभवः । यद्येवमन्तर्गहना कोऽर्थः स्पात् स्थापिनातमनः ॥ ५ ॥

अनुभवात् संस्कारः संस्काराच् च स्मृतिजन्मानना तं पूर्वानुभवं अनुकुर्वत्येवावगाहितविषयं तमनुभवमाभासयति । एवंमात्मस्थैयेण क्रियमाणयोगिना संस्कारस्योत्मवादप्यनुकीरणात् तेनैव च सिद्धे: ॥ ५ ॥

ततो भिन्नेषु धर्मेषु तत्त्वरूपाविशेषतः । संस्कारात् स्मृतिसिद्धौ स्पात् स्मर्ता दृष्टेव कलिप्तः ॥ ६ ॥

सुखदुःखज्ञानादिभिन्नधर्मा श्रयत्वेनापि आत्मनोऽनुपयोगो भिन्नैर धर्मैरसंभिन्नस्य अनुपजातविशेषस्य स्मृताव् अन्यापृते: । तद् दृष्टवत् कल्पनामात्रमेतदात्मा स्मरंति ॥ ६ ॥

70 ca omitted in J. 71 anukurvarty eva: anukurvaty E Gh Ñ Ch J Jh Ñ T, anukurvī-ta C. 72 tam anubhavam ābhāsayati: tam anubhavam avabhāsayati E C Ñ, tam anubha-vam ābhāsayati Ch, tam anubhāvayati T. The reading accepted (Gh Ñ J Jh) agrees with the quotation ibid. p. 126. 73 evam: yady evam Ch, evam corrected to eva Gh. 74 Cit. ibid. p. 126 "tenaiva ca" iti samskārena; ca omitted in Ñ. 75 sukhaduhkha0 cit. ibid. p. 131. 76 odharmaśrayatvena: odharmatvena T Ṭ. 77 'nupayogo: nopayogo E Gh Ñ C Ch J Jh Ñ T; cf. the paraphrase ibid. p. 132. 78 dharmair, omitted in E Ñ and added in the margin in C, cit. ibid. p. 131 "bhinnair dharmair asambhinnasya" iti alabdha-tadāveśasyeti yāvat. 79 anupajāta0: anujāta0 Ñ. 80 Cit. and commented on ibid. p. 132 (avyāprteh is interpreted as a bahuvrīhi depending on ātmano), where a portion of the whole passage is also paraphrased "smrtāv avyāprteh" iti vrttigrantham saṣṭhyantam ātma-vāt pratisamdhāv ātmano 'nupayogaḥ smrtav avyāprtatvāc ca. 81 tad draṣṭvat cit. ibid. p. 132.

I 2.7

ज्ञानं च चित्स्वरूपं चेतदनित्यं किमात्मवत् । अथापि जडमेतस्य कथमर्थप्रकाशता ॥ ७ ॥

'चित्स्वाभाव्ये ज्ञानस्य 'चेत्यधर्मदेशकालावेशयोगाद् 'आत्मन इव नित्यतादिप्रसङ्गः । जडत्वे कथं अर्थस्यासौ प्रकाशः ॥ ७ ॥

अथार्थस्य यथा रूपं धत्ते बुद्धिरतथात्मनः । चैतन्यमजडं सैवं जाड्ये नार्थप्रकाशता ॥ ८ ॥

'ज्ञानं बुद्धिः । सा जडापि यथा विषयरूप'च्छायां धत्ते तथात्मनोडपि चैतन्याच्छायाम्, 'अतोऽसावर्थप्रकाशः । तथा च 'तस्यापिचित्स्वरूपतां स्यात् । 'एवं ज्ञान'स्यापि 'ज्ञान्यस्य सम्बन्ध, अनुपपत्तेः । क्रिया तु 'न स्वरूपपेऽस्ति 'ज्ञान्य- सम्बन्धितया ॥ ८ ॥

82 cit°: jñānacit° T. 83 cetyadharma° cit. ibid. p. 136. 84 ātmanā iva nityatādiprasangah cit. ibid. p. 135. 85 jñānam buddhih cit. and commented on ibid. p. 154. 86 °cchāyām cit. ibid. p. 154. 87 ato 'sāv arthaprakāśaś cit. and commented on ibid. p. 154 "atah" iti ubhayavacchāyādhāranāt / asau buddhih / arthasya prakāśas tatphalaparvanto 'dhigama ity arthah. 88 tathā ca: tathāpi Ch; tathā ca cit. ibid. p. 174 vrttau tathāpīty arthenā ca ity enābhyupagamavādābhidhāyinī (cf. also p. 158). 89 tasyāś cit-svarūpatā syāt cit. ibid. p. 174. 90 evam cit. and commented on ibid. p. 176 evam iti sūtravadayoktābhir yuktibhiḥ yaḥ prasaṅga uktaḥ, tadviparyāyeṇa anityarūpasya ananyāśri-tasya jñānasyā siddhir ity arthah. 91 sad api cit. and commented on ibid. p. 176 sad api iti vrttāv apiśabda uttarakārikāvakāśādānārthaḥ / apiśabdo hi kāmakārarūpam sattve tāvad anvavasargaṃ dyotayann anyasya" iti "na adhigantum śakyam iti prakatayati. 92 Even though it is not found in any MS (nānyasambandhitayā E Gh Ñ C Ch J Jh Ñ T, nānyasya sambandhinaḥ T), nānyasya sambandhi is the reading that the context seems to require. This is confirmed by the quotation ibid. p. 182 tad āha "na anyasya" iti "sambandhi" iti tādātmyasamavāyādineti arthah. Then the ĪPVV cites the words that the ṭīkā uses to gloss anyasya: "anyasya" iti vrttim vivrnoti pāthapūrvakam "ātmākhyasyā" iti "vastunaḥ" iti.

8

क्रियाप्यर्थस्य कायादेशतत्देशादिजातता ।

नान्याह्टेर्न साप्येका क्रमिकैकस्य चोचिता ॥ ९ ॥

क्रियापि पूर्वपर्यभूतावयवैका कारकन्यापाररूपा न युक्ता क्रमिकस्यानेककालस्पृशः स्वात्मैकयोगात्, नापि कालक्रमण्यापी चैकरस्वभावरुच तस्या आश्रयो युक्तः, केवलं गमनपरिणामादिरूपा सा क्रिया ।

कायादीनां तु तत्तद्रिदेशकालगतापूर्वसत्तामात्रम् एतदतिरिक्तस्य अन्यस्यानुपलम्भात् ॥ ९ ॥

तत्र तत्र स्थिते तत्तद्वतीत्येव दृश्यते ।

नानयन्तनान्योऽस्ति सम्बन्धः कार्यकारणभावतः ॥ ९० ॥

पूर्वसिमन्सति परस्य सत्तेत्येतावन्मात्रेऽनुभवः ।

क्रियावच्च न क्रियाकारकसंबन्धः कश्चिदिन्नस्स्यास्यानुपलम्भात् ।

कार्यकारण-

of the vrtti is recalled at the beginning of the ĪPVV on the following kārikā (ibid. p. 183): svarūpena anyasambandhitayā ceti ... / yuktyā pramānavyāpāreṇa bādhitatvam svarūpa-ta uktam "na sāpy ekā kramikā" iti, anyasambandhitayā tu uktam "ekasya kramikā katham bhavet". 95 T omits api, which is confirmed by the quotation ibid. p. 183 kriyāpi na yuk-tā iti vrttir vyākhyātā. 96 pūrvāparībhūtāvayavaikā (cf. SDVr p. 91) kārakavyāpārarūpā paraphrased ibid. p. 184 kārakasya phalasiddhihetubhūtakriyālimititasya svakriyāpekṣayā vā kartuh sambandhī tatsamavāyī tadatāmā vā vyāpāro rūpam svam laksanam yasyāh. sā kriyā yasyāh sambandhino bhāgāh pūrvāparībhūtāh kramavittayah iti yāvat.

oikākāravyāpāra° T Ch. 97 na omitted in E. 98 Cf. the paraphrase ibid. p. 191 ekā katham kramikā iti sūtram sphutayitum vrttau kramikatvam vyācaṣṭe / ekakālasprktvam ekaś ca tatkalāś cetī svarūpatas tāvad aikyaṃ kramikāṃ bādhitam. 99 Cf. ibid. p. 193 āśraya-bādhakanirūpaṇam sūtrāṇiṣam vyācaṣṭe vrttikrt "nāpi kālakramayāpi ca" iti / kārakāra-vanyonyavirodhadyotakau.

100 Cit. ibid. p. 194 kevalam gamana ity ādina vrttigranthenānyasya. 101 rūpā: °rūpam T. p. 196 parināma iti (cit. by the tikā, which comments on it). 102 kāyādīnām ... mātram paraphrased ibid. p. 195. 103 tu omitted in T J Ñ. 104 Ch omits anyasya. 105 kriyāvac ca na kriyākārakasambandhah kaścit cit. ibid. p. 198: kriyākārakādi° E: Gh N C Ch Jh Ñ T.

I 2.10 भावादते १०६नान्यद् १०७वस्तूनां ज्ञातेयम् ॥ 90 ॥

दृष्टस्यानेकरूपत्वात्सदृस्यान्यानपेक्षणात् । पारतन्त्र्यादयोगाच्च तेन कर्तृपि कल्पितः ॥ 91 ॥

संबन्धो दृष्टो न १०८चैकेनात्मनोभयत्रावस्थितिरयुक्ता १०९न च द्वयोः सिद्धयोर् ११०अन्योन्यापेक्षात्मा नापि स्वात्ममात्रनिष्ठयोः १११पारतन्त्र्यरूपः सम्बन्धः । ततो ११२यथा ज्ञातृत्वं कल्पितं तथा कर्तृत्वमपि ११३इति कथमात्मा सर्वेश्वर इति ॥ ९९ ॥

११४ इति ज्ञानाधिकारेऽ द्वितीयमाह्निकम्॥

106 Cf. ibid. p.198 nānyad iti atra pūrvasya hetutvam vrttāv uktam bhinnasya vya-tiriktasya asya sambandhasya kriyāyā iva anupalambhād ity anena: T omits asya. 107 Instead of vastūnām T has an almost illegible word (perhaps iha, which would equally good). 108 Cf. ibid. p. 203 ekatvam hi abhinnarūpatvam, tadviruddhena ca bhinnarūpa-tvena vyāptam dvayar yugapat svarūpasarvasvena avasthānam; °sthitir E Gh N C Ch J Jh Ñ T. 109 na ca dvayor cit. ibid. p. 204; 110 anyonyāpekṣā° cit. ibid. p. 204; E Gh Ñ Ch J omit °ātmā, °apekṣanāpi C Ñ, °apekṣe T. 111 pāratantryarūpah sambandha" ity atra ittham yojanā - pāratantryam bhavad api sambandho na siddhyati. °svarūpah T. 112 yathā ... kalpitam ... kartrtvam api cit. ibid. p. 206. where the whole passage is commented on. 113 iti katham ātmā sar-vesvarah cit. ibid. p. 206: maheśvarah T. 114 T adds iti purvapakṣo, Ch iti pūrvapakṣavi-vrtih. J iti pūrvapakṣanirūpanam.

तृतीयमाह्निकम्

सत्यं किंतु स्मृतिज्ञानं पूर्वानुभवसंस्कृते: । जातमात्रात्मनिष्ठं तन्नोध्यानुभववेदकं ।। १ ।।

पूर्वानुभवसंस्कारप्रबोधजन्मापि स्मृतिर् आत्ममात्रनिष्ठत्वात् स्वरूपसंवेदनैव न तु पूर्वानुभवा वेशाभावात्पूर्वानुभूता-र्थन्यवस्थापिका घटते ।। १ ।।

एकस्वाभासैव नान्येन वेद्या रूपदशेव हृक् । रसे संस्कारजत्वं तु तत्तुल्यत्वं न तत्र ते: ।। २ ।।

सर्वां हि ज्ञप्तिं: स्वसंवेदनैक रूपानन्यासंबंधि वेद्या , रूप-रज्ज्ञानयोरेर अन्योन्यवेदनेऽन्योन्यविषयवेदनमपि स्पात् ततश्च इन्द्रियनियमाभाव: । पूर्वानुभव संस्कारजत्वेन तत्सादृश्यमात्रं

115 T adds atha paradusśanānupapattinirūpanam. 116 pūrvānubhavasamskāprabodhajanmāpi cit. ibid. p. 210; oprabodha° omitted in T (in Gh it is added under the line); °jāpi E. 117 ātmamātranișṭhatvāt cit. ibid. p. 210. 118 Cf. ibid. p. 211 tat katham uktam “ātmanișṭhasvarūpasamvedikaiva” iti (from the ṭīkā). On the same page the compound “svavisayasamvedikaiva” iti; samvedakaiva Ch. 119 na tu vyavasthāpikā ghatate cit. ibid. p. 213; °vyavasthāpikāpi T; °vya-vyasthāpitā N. 120 °āveśābhāvāt: °āveśāt J. 121 sarvā hi jñaptīh cit. ibid. pp. 216 and 222; sarvāpi T. 122 Cf. the quotation from the ṭīkā ibid. p. 215 svasamvedanaikarūpatvāt, glos-sed with svasamvedanam ekan rūpam yasya iti; °svarūpā. Cf. ibid. p. 229 svasamvedanaikarūpā … yā kācit hi jñaptih, sa anyasamvidyedya nā bhavatīti and the quotation from the ṭīkā ibid. p. 222 “jñānam jñānāntaravedyam na bhavati” iti. 123 °rūpānanyā°: °rūpā nānyā° E Gh N C Ch J Jh N T. 124 °vedyā: °samvedyā° E. 125 anyonyavedane anyonyavisayavedanmapi syāt cit. ibid. p. 229; anyonyavedane omitted in N Ch Jh: anyonyasamvedane E Gh (added in the margin) C J N T. 126 tataś ca cit. ibid. p. 230. 127 Cf. the paraphrase ibid. p. 230 iti indriyani(y)amo hīyeta. 128 Cit. ibid. p. 214 upacāra iha vrttau sphuṭikṛtah “samskārajatvena” iti; °samskāratvena C. 129 Cf. ibid. p. 253 āha “sādrśyamātram” iti, perhaps cit. through the ṭīkā; Ch omits tat°.

11

I 3.2 न तु पूर्वानुभवावगतिः , तदभावात् तत्साहृश्यमपि नावसेयम्

॥ २ ॥

अथातद्विष्यत्वेऽपि स्मृतिरस्तदवसायतः ।

हृष्टालम्बनता भान्त्या तदेतदसमञ्जसम् ॥ ३ ॥

न चापि भान्त्या पूर्वानुभवं तद्विषयं च

शुक्तौ रजतम् इवास्वेद्यमानमध्यवस्यतीति स्मृतिस्तद्विषया ॥ ३ ॥

स्मृतितैव कथं तावद् भान्तेश्चार्थसिद्धतिः कथं ।

पूर्वानुभवसंस्कारापेक्षा च किमितीष्यते ॥ ४ ॥

पूर्वानुभवापकाशात् तद्विषयसंप्रमोषेऽध्यवसायमात्रात्स्मृतितत्त्वं

न युक्ततम् । न च भान्त्या पूर्वानुभवार्थव्यवस्थापनम् ,

पूर्वानुभवास्पर्शं च तद्विन्नयोगक्षेमाया भ्रान्तेः संस्कारजत्वे

को ग्रहः ॥ ४ ॥

130 na ca cit. ibid. p. 236 asamañjasam iti vrttau vyākhyātam neti samastavākyārtha-nisedhena / cakārena āśankādyotakenātha iti vyākhyātam. 131 E omits ca. 132 śuktau

cit. ibid. p. 236. 133 T reads iti vāo, which, if corrected to itīvao, would be equally acceptable. 134 The whole sentence up to the danda cit. ibid. p. 238. 135 osampramoṣe:

oasampramoṣe C. 136 pūrvānubhūtarthaocit. ibid. p. 239, where this point is dwelt on at length; cf. in particular na asyām [smrtau] kimcit prakāśate yena vaiparityam āśankate /

etadartham eva vrttau pūrvānubhūtāgrahanam anumānavikalpanādāv iva apūrvaprakāśa-tānirākanatātparyena prayuktam. 137 pūrvānubhavāspārśe ca tadbhinnayogakṣemāyā

paraphrased and commented on ibid. p. 240 purvānubhavasparśo 'sādrśyam anāveśaś ca saptamyā hetutvena nirdiṣṭo vrttau bhinnayogakṣematve sādhya iti vivrnoti “tataḥ” iti

yogo 'labdhalābhah. iha tu tadupalaksitam prakāśalaksanam svarūpam, kesam labdha-pariraksanam, tadupalaksitas tu iha visayāṃśaḥ / tadubhayam anubhavāt bhinnam smrteḥ

prakāśasya apūrvatvat visayasya ca avastutvāt iti. T omits ca. 138 bhrānteḥ saṃskārajat-ve ko grahaḥ cit. ibid. p. 240; p. 241 Abh. records a different reading (ityanye patanti):

o jatvam. 139 grahaḥ: brahmaḥ G Ch (in T brahmaḥ, deleted.precedes grahaḥ); hetuḥ K C Ñ.

भान्तितवे चावसायस्य न जहाद्विषयसस्थितिः ।

ततोऽजाडथे निजोल्लेखनिष्ठान्

नार्थस्थितिसतताः ॥ ५ ॥

१४०अध्यवसाय एव भान्त्या विषयण्यवस्थापको न तु स्वसंवित्,

१४१स च जडः कथमर्थन्यवस्थाया हेतुः । चिद्रूपोऽप्यतीतार्थ१४२मात्रमाभासयेदबाध्रस्वात्मोल्लेखमात्रप्रकाशो वा न तद्व्रतुः

॥ ५ ॥

एवमन्योन्याभिन्नानामपरस्परवेदिनाम् ।

ज्ञानानामनुसंधानजन्मा नश्येज्जनसस्थितिः ॥ ६ ॥

१४३ज्ञानानि स्वात्ममात्रपरिनिष्ठितानि१४४स्वसंविद्रूपतया १४५अपर-

संवेद्यानी । १४६तेषाम् १४७अन्योन्यविषयसंघट्टनामयः १४८परमार्थो-

संवेद्यानी ।

140 adhyavasāya eva cit. ibid. p. 242. The whole passage is paraphrased and commented on pp. 242-43 bhrāntyā bhramarūpena upalaksito yo visayavayavsthāpaka uktah, so 'dhyavasāya eva, na tu svasamvit / nāpi svasya prakāśamasyā rūpasya samvit nir-vikalparūpā. E (but not G) Gh omit sva°. 141 sa ca cit. ibid. p. 243, where the rest of vṛtti is also paraphrased, sa tarhi adhyavasāyo jado vyavatisṭhāpayiṣite bāhye 'rthe syā aprkāśarūpatvāt / yadi tu tatra prakāśarūpah syāt, yogina iva tatra pūrvasminn arthe vikalpanīābhimate sphutatayā pūrvaprakāśaḥ syād iti na smṛtitā. nāpi adhyavasāyatā bhaved anubhavarūpatvāt / atha na bāhye 'sau prakāśo yena jadah, kimtu svātmani svollekhe ca svākāre prakāśarūpa eva asau. tarhi punar api arthe bāhye jada eveti bāhyarthavyavastāyā hetuh syāt / sā hi samvinniṣṭhā bhavati, na tu kadācit jadanisṭheti / bhramarūpena yad abāhyabāhyayor ekīkaranam, na tena bāhyasya kimcit sprśtam tatparśe bhrānta-tvāyogāt / na hi raja-tajñānena satvarajatasya śukter vā vyavastā kācit iti tātparyeṇa vṛttyarthah / evam sūtre atītragrahanam ca vṛttau vikalpalaksanam utprekṣāvikalpaśyāpi pūrvānubhavopajivana-praveśena vastuto 'tītvād eveti.

142 mātram only found in T. 143 jñānāni svātmumātra° cit. in the ṭīkā (Torella 1988: 144); cf. also ĪPVV I p. 253. 144 svasamvidrūpatayā cit. in the ṭīkā (Torella 1988: 144). 145 aparasamvedyāni: nāparasamvedyāni E Gh N Ch Jh T, na parasamvedyāni C N, na parasamvedyāni T. I have accepted the reading as quoted and glossed ibid. p. 255; see also the ṭīkā (Torella 1988: 144). 146 teṣām cit. ĪPVV I p. 268. 147 anyonya° cit. in the ṭīkā (Torella 1988: 150); cf. ĪPVV I p. 268 vṛttau visayasamghattanāyāḥ prādhānyāt ... and the ṭīkā (Torella 1988: 150) vṛttau ca visayasaṃghatanāpeksayā saṣṭhī; saṃghattanā°: saṃghattanā° E Gh C N T (also possible). 148 paramārtha° cit. ibid. p 269 and in the ṭīkā (Torella 1988: 151).

I 3.6 पदेशपर्यन्तो लोकन्यवहार: १४९कथम् ॥ ६ ॥

न चेदन्तःकृतानन्तविश्वरूपो महेश्वरः ।

स्पादेकशिचद्रपुजानस्मृत्यपोहनशक्तिमान् ॥ ७ ॥

१५०चितत्वमेव १५१अतो 'तिरिक्तस्य चित्- १५२अस्यैव ज्ञानाङ्किका: शक्तयः ।

मतः स्मृतिज्ञानपोहनं चेति ह्युक्तम् ॥ ७ ॥

१५३इति ज्ञानाधिकारे तृतीयमाहिकम् ॥

चतुर्थमाहिकम्

स हि पूर्वानुभूतार्थो१५४प्लबधा परतोऽपि सन् ।

विमृशन्न इति स्वैरं स्मरत्यपपदिश्यते ॥९॥

१५५परचादपि पूर्वानुभूतार्था१५६नुभवितृत्वात् १५७पूर्वानुभूतार्थ-

149 katham cit. ĪPVV I p. 272. 150 cittatvam eva and ato 'tiriktasya cit. in the tika (Torella 1988: 152); riktasya Ch. 151 Cf. the quotation in ĪPV I p. 276 vrttikrd āha aseṣapadārthajñānānām anyonyānusamdhānam cittatvam iti. 152 asyaiva cit. ibid. p. 285. 153 °ādikāḥ: °ādi° T. 154 C Ch J T add paradarśanānupapattih. 155 paścād api cit. in ĪPVV II p. 4. 156 anubhavitrtvāt: °anubhavitrtayā E C Ñ (°tvāt is also the reading of Kh). 157 The whole passage (from pūrva°) is paraphrased and partly cit. (through the tika) ibid. p. 4 "tasyai kasya pūrvānubhūtatvena yah pratyavamarśaḥ" iti ubhayakālasamśparśī "atra" iti prānuparyastakadehādau samkucitāsya "pūrvānubhūtārthāsampra-moṣarūpā smṛtir nāma vyāpāraḥ" iti sambandhaḥ / ekatvāt hi sa ubhayakālasprk / kar-trtaiva ca vibhutā svairitvam iti vrttiyojanā; p. 11 atra iti vrttipadam; p. 13 vrttau ca uktam vibhoḥ iti. vibhoh kartuḥ Ch.

प्रकाशासंप्रमोषणम्, तस्यैकस्य विभो: कर्तुः स इत्पत्र पूर्वानुभूतत्वेन प्रत्यवमर्शः स्मृतिरन्वाम न्यापारः ॥ ९ ॥

भासयेच्च स्वकाले sर्थात् पूर्वाभासितमामृशन् । स्वलक्षणं घटाभासमात्रेणाथाखिलात्मना ॥ २ ॥

158 स्मृतिशक्त्या स इति पूर्वानुभूतं स्वलक्षणं परामृश्यन्नाभासयत्येव 159 अन्यथा 160 प्रकाशितस्य 161 परामर्शो न कृतः स्पात् 162 स्वसत्ताकाल एव च, 163 तेन स्मरणकाले 164 नष्टस्याप्याभासो न दुष्यति । कदाचित् त्वरितावशाद् घटकाच्चन्द्रद्रव्य 165 सत्ताद्य 166 नितमैकाभासरूपेणैवास्य 167 स्फुटावभासः, 168 अन्यदा तु 169 सर्वात्मनार्थितत्वेन तथैव 170 । अतिशय 171 निरन्तरावहितचेतसस्तु 172 दृष्टार्थ 173 प्रत्यक्षीकरण एव ॥ २ ॥

158 smrtiśktyā cit. ibid. p. 19. 159 anyathā cit. in IPVV II p. 22. 160 prakāśitasyā cit. and commented on ibid. p. 22 pūrvaprakāśitatvena tatprakāśanulagnaḥ katham parāmarśo bhavet / yadi ca na bhavet pūrvaprakāśitatvena parāmarśanam, tat sa iti bhūta-prakāśaparāmarśo na syāt / tena "prakāśitāsya" iti vrttau prakrtyarthe ca pratyayārthe ca bharah kartavyaḥ; cf. also p. 31. 161 parāmarśo: parāmarśo 'pi Ch. 162 svasattākāla cit. ibid. p. 31; ca omitted in J. 163 tena cit. ibid. p. 23. 164 Cf. ibid. p. 23 tathā naṣṭasya api abhāso na duṣyati cit. ibid. p. 25, where the whole argumentation of the vrtti, as developed in the tika, is summarized. 165 osattā°: osattva° Gh N J Jh T T. 166 oanyatama° cit. ibid. p. 29 ghatābhāsāsaśabdah sūtre kevalābhāsopalakṣanārthaḥ iti vrttau vyākhyātam anyatamapadena. 167 My conjectural emendation for asyāśphutāvabhāsah (Gh N C J Jh Ñ T T, asyāḥ sphuṭāvabhāsah E Ch. rellā 1988: 154-55. 168-69) and by the (probable) gloss sphuta eva added to the following tathaiva in Ch J T (see also the tika: tadā sphuta evābhāsah; Torella 1988: 155). The reading asyāḥ sphuṭāvabhāsah, to which asyāśphuṭāvabhāsah might be reduced by assuming the omission of the visarga before samyoga, could only be acceptable if we took asyāḥ as referring to smṛti (which does not seem likely). 168 anyadā tu cit. ibid. p. 29 and in the tika (Torella 1988: 155). 169 sarvātmanārthitvena: sarvārthitvena J. 170 T J (and Ch, in the margin) add sphuta eva (being probably only a scribal gloss of tathā eva). 171 Cf. ibid. pp. 29-30 antarayati sāntaram karotīti antaram vijñīyāṃśo malam, tadabhāvo nairantaryam; atiśayanirantarā° cit. in the tika (Torella 1988: 155). 172 Cf. ibid. p. 30 ane-na drṣṭārthaśabdo vyākhyātah. 173 °pratyakṣīkāra cit. ibid. p. 29.

I 4.3

न च युक्तं स्मृतेर्भेदे स्मर्यमाणस्य भाषनम् ।

तेनैक्यं भिन्नकालानां संविंदां वेदितैष सः ॥ ३ ॥

पूर्वानुभूतार्थोः १७४ snubhaven सह १७५ tātkālika १७६ स्मृतिप्रकाशे

सर्वभावमानः स्मृत्याभिन्न एव १७७ प्रकाशाद्विननेsय प्रकाशमानतां-पपत्ते: । एवं १७८ चानुभवस्मृत्यादि १७९ संविदामैक्यं स एव चात्मा

वेदक: । १८० तथा हि ॥३॥

नैव ह्यनुभवो भाति स्मृतौ पूर्वोऽर्थवत् पृथक् ।

प्राग्नुभूर्महमित्यात्मारोहणभासनात् ॥ ४ ॥

स्मृतौ स्मर्यमाणो १८१ snubhūtārthō yathā pūrvam bhūto १८२ bhāti न

तथानुभवः स्वात्मन् एवाहन्ताप्रत्येयस्यानुभवमयत्वेन पृथनात्,

१८३ yaśyānekakālo$havedyo$arthah स एवात्मा ॥४॥

योगिनामपि भासन्ते न दशो दर्शानान्तरे ।

स्वसंवेदेकमानास्ता भान्ति मेयपदेऽपि वा ॥५॥

१८४ सर्वज्ञानामपि १८५ प्रमात्रन्तरगतोपलम्भाः स्वसंविद्मात्रवेद्यास्व-

174 Cf. ibid. p. 34 anubhavena saha arthah smaraṇād abhinnah iti vrttyā tātparyadayam sautram vyākhyātam. 175 tātkālika°: tātkālike Ch. 176 °smrtiprakāśe cit. ibid. p. 33; T omits °smrti° and reads °prakāśo. 177 prakāśād bhinnasya cit. ibid. p. 34; cf. p. 33 prakāśād bhinnasya mā bhūt prakāśamānātā; prakāśyamānatā° T. 178 ca only found in T T. 179 °samvidām: °samvid° N. 180 tathā hi cit. ibid. p. 37 181 'nubhūta°: 'nubhūto T G. 182 bhāti: 'bhāti T. 183 yaś ca cit. ibid. p. 40; anekakālo, cit. ibid. p. 41, is quoted on the same page in a slightly different form (probably meant to clarify its meaning) nanu ca samvidah kālasparśa eva nāstīti katham vrttih anekakālagata iti / anekakālagateṣu vedyeṣu yad unmukhatvam, tāvanmātrād eva sa bodho 'nekakāla iti upacāreṇa acaryate ... tadunmukhaś ca ekah svasyamedanapramātā. 184 Cf. ibid. pp. 43-4 nanu yogī

kriyāśaktiprakarṣavān api iha darśane, tat katham jñānaśaktiprakarṣamātram asya vrttau nirūpitam; cf. also ibid. p. 47 sarvajñagrahanam ... . 185 pramātrantaragatopalambhāḥ

16

भावः १८६स्वात्मारूढा एव भासेरन्, १८७अतश्च तेषां १८८योगिनां I 4.5 परात्मतापत्तिरेव १८९तत्त्वम्। प्रमेयकक्ष्यायाम् अपि१९० घटादिवत् १९१प्रातिस्विकेन शुद्ध१९२बोधात्मना रूपेण १९३अवभासेरन् यदि १९४तथा संभावेत् ॥ ५ ॥

स्मर्यंते यद् हगासीन्मे सैवमित्यपि भेदतः । तद्‌ध्याकरणमेवास्या मया दृष्टमिति स्मृतेः ॥ ६ ॥

१९५मया दृष्टमिति प्रमाणतन्त्रगत१९६दर्शनपरामर्शान्तिमकाया एव स्मृतेःविभज्य १९७कथनम् १९८एतदेवमनुभवोऽसौ ममाभूति भेदेनापि १९९निर्देशः२०० ॥ ६ ॥

या च पश्च्याम्यहिमं घटोऽयमिति वाचा । मन्यते समवेतं साप्यवसातरि दर्शनम् ॥ ७ ॥

पुरःस्थितार्थविकल्पने‌डपि घटमिमं पश्यामि घटोऽयमिति वा प्रमातृमयमेव २०१दर्शनं प्रत्यवमृश्यते ॥ ७ ॥

186 svātmārūdhā eva cit. ibid p. 47 and in the țīkā (Torella 1988: 158). bhāseran cit. ĪPVV II p. 52; ābhāseran E, avabhāseran K G Gh N C Ch J Jh N T. 187 ataś ca: ata eva T. 188 yoginām parātmāpattir eva cit. in the țīkā (Torella 1988: 158) (as paramā°, which is also the reading of Kh Gh J). 189 tattvam cit. ibid. p. 47 “tattvam” iti vivṛṇoti “paramātmavedakam” ity anena tattvam paramātmavedakatvam ity arthah (cf. Torella 1988: 158, n. 54) [...] sarvajñagrahanam prāgupakrāntaparasamvedanopakramāt parātmaṁvedakaparāyāyatām etīti bhāvaḥ. 190 api: eva vā T, also possible. 191 prātisvikena cit. ibid. p. 52. 192 °bodhātmanā rūpeṇa: bodhena E, but cf. the paraphrase ibid. p. 52 ayam ca atra bhāvaḥ svasmin svasmin darśane yādṛśam bodhasya śuddha-rūpaṃ ... . 193 Cit. (as bhāseran) ibid. p. 52. 194 T omits tathā, but cf. ibid. p. 53 yadi tathā iti na tu etat sambhavatīti; tadā Ch. 195 mayā dṛṣṭam cit. ibid. p. 54. 196 In T °darśana° is added under the line; °parāmarśana° Gh N. 197 Cf. ibid. p. 54 kathanaśabdena (from the țīkā). 198 etad omitted in T. 199 T adds drśo, being probably a gloss. 200 Ch adds smṛter vyākaraṇam eva. 201 The text cit. ibid. p. 55 is a little different: ity evam vṛttau vyākhyāto “darśanam eva tat pratyavamṛśyate” iti vacanena.

I 4.8

तन्मया हृश्यते हृटोऽयं स इत्थामृशत्यपि ।

ग्राह्यग्राहकताभिन्नावर्थौं भातः प्रमातरि ॥ ८ ॥

तस्माद् 202ट्कपरामर्शेपुरःसारायां स्मृतौ विकल्पमात्रे वा

202एकशब्दानुविधं विनापि सु इत्ययमिति वार्थमात्र204निर्देशेन

204सर्वत्रैकप्रमातृलोनावेवानुभाग्यानुभावकौ 206मायाकृतवic्छिन्ना-

वभासाव् 207अपि 208प्रकाशते ॥ ८ ॥

इति ज्ञानाधिकारे चतुर्थमाहिकम् ॥

पञ्चममाहिकम्

वर्तमानावभासानां 205भावानामवभासनम् ।

अन्तःस्थितवतामेव घटते बहिरात्मना ॥ ९ ॥

प्रत्यक्षे210सपि यावदर्थानां भेदेनावभासः प्रमातृ211न्तर्लीनानामेव

सतां युक्तः ॥ ९ ॥

202 Cf. ibid. pp. 57-8 vivṛttā'v api ... dr̥k prādhānyena parāmṛśvate / smarane pratyakṣādhyavasāye ca ādāranam eva purahsarāpūrvādiśabdair darśitam. 203 In Ch °pa-rāmarśa° is added after dr̥k° under the line. 204 °nirdesena: °nirdeśe T N C Jh Ñ T; in Gh °na is added in the margin. 205 sarvatraika°: sarvatātraika° T. 206 Cf. ibid. p. 62 nanu māyākṛtavicchedā'v iti yad uktam ... 207 Cf. ibid. p. 57 apiśabdo bhinnākramah. bhinnā'v api ekatra ābhasate iti vṛttyanusāreṇa vā yojamā 208 prakāśete: prakāśeta J. 209 It is out of °question that bhāvānām is the original reading of the kārikā (cf. vṛtti: arthānām). Bhāskara holds, instead, that the text which Abh. had in front of him had bhin-nānām, glossed by him with vicchinnānām. Also the editors of Bh accept this thesis. I would say that the text commented on by Abh. also had bhāvānām and that bhinnānām - as well as the further explanation vicchinnānām - are meant to elucidate the sense of bahirātmanā; bhāvānām is simply understood in yeṣām ... teṣām ... bhinnānām etc. (ĪPV I p. 196). Furthermore, bhāva is mentioned in the paraphrase in ĪPVV II p. 68 atyakta-samvidabhedasyai'va bhāvasya kalpitapramātrapeksa'yā vicchedena yat prakāśanam ... (it is worth noting that bahirātmanā is here paraphrased as vicchedena prakāśanam; as bhe-denāvabhāsah in the vṛtti). 210 Cf. ibid. p. 68 vṛttā'v apiśabdo yāvacchabdaś ca sūtraśeṣatvena paṭhitāh. 211 °antar: °antara° N C Ñ.

18

प्रागिवार्थोंsप्रकाशः स्यात् प्रकाशात्मतया विना ।

न च प्रकाशो भिन्नः स्यावात्मार्थस्य प्रकाशता ॥ २ ॥

प्रमातृ²¹²संज्ञप्रकाशस्वरूपतां विना ²¹³यथादौ घटोऽस्य नावभातस तथा ²¹⁴ज्ञानकालेsपि स्यात्, ²¹⁵प्रकाशमानता चार्थस्य प्रकाशः स्वरूपभूतो न तु भिन्नः ॥ २ ॥

भिन्ने प्रकाशे चाभिन्ने संकरो विषयस्य तत् ।

प्रकाशात्मा प्रकाश्योऽर्थो नापकाशश्च सिध्यति ॥ ३ ॥

प्रकाश²¹⁶मात्रं चार्थी²¹⁷दिनं सर्वार्थसाधारणं ²¹⁷तस्य घटस्य प्रकाशोऽययं पटस्यैवेति विषयनियमो निर्निबन्धनः । तस्माद् ²¹⁸अर्थस्य सिद्धिः प्रकाशात्मतयैव ॥ ३ ॥

²¹² °samjñā°: °samjñaka° T, samjñā E. ²¹³ yathā: yadā Ñ C Ñ T. ²¹⁴ jñānakāle: jñānajñāne J. ²¹⁵ Cf. ibid. pp. 68-9 tathā ca vrttih - svarūpabhūto yah prakāśah so 'rthasya prakāśamānata na tu asau prakāśamāna-tātmā arthād bhinnah san prakāśah, so 'rthasya prakāśamānata-yukteti; p. 69, where a slightly different interpretation is given, atra pakse vrttir evam netavyā - arthasya yā prakāśamānātā, sa prakāśo 'nubhavah. na tu asau prakāśamānā-tātmā prakāśo 'rthasya prakāśamāna-tātmā prakāśo 'rthasya prakāśamānātā svarūpabhūtā 'rthaśarīramagnah. The passage ibid. p. 75 nanu prakāśamānātā svarūpabhūtā ghatusya iti vrttau kim etat, although presenting itself as a literal quotation is only a further variation on the same theme. ²¹⁶ Cf. ibid. p. 77 vrttau mātraśabdena abhinnatvamātram vyākhyātam, sarvārthasādhāranatvena samkarasa-nkāyāh pīthābandhah krtah. ²¹⁷ Cf. ibid. pp. 77-8 "tasya" iti prakāśamātrasya "viṣaya-niyamo nirnibandhanah" na kenacit hetunā upapadyate iti sambandhah / visayaniyam-ākāram darśayati "ghatasya" ity ādinā. In J tasya is followed by prakāśamātrasya, which is precisely the gloss of ÏPVV; ghataprakāśo J; T does not repeat ayam. ²¹⁸ Cf. ibid. p. 78 trtiyapādāsya artham prakāśāmatāśabdena vrttau vyācaśte "arthasyā siddhih" iti / "tat" iti vyākhyātam "tasmāt" ity anena; artha° E Gh Ñ C Ch J Jh Ñ T; prakāśāyat-tā T.

I 5.4

तत्तदाकस्मकाभासो बाह्यं चेदनुमापयेत् ।

न त्वभिन्नस्य बोधस्य विचित्राभासहेतुता ॥ ४ ॥

जड़ानाम् 219आभासमानतैव 220सत्तासिद्धिः सा च 221आभासत्वैव । ततस्तु बोधमात्रं 222एकमेवात्र तत्त्वम् । 223तस्याविशेषे ऽपि 224क्रमेण विशिष्ट्यमानो ऽर्थावभासस्ततो ऽन्यमप्रकाशमानं हेतुतया बाह्यामर्थमूहयेदिन्द्रियवत् ॥ ४ ॥

न वासनाप्नबोधो ऽत्र विचित्रो हेतुतामियात् ।

तस्यापि तत्पबोधस्य वैचित्र्ये किं निबन्धनम् ॥ ५ ॥

225वैचित्र्यवासनाप्रबोधः 226स्थाने 227बोधो ऽपि नः , तस्यापि वैचित्र्य को हेतुः । ततो बाह्य एवार्थो भिन्नाभासहेतुः ॥ ५ ॥

स्यादेतदवभासेषु तेष्वेवावसते सति ।

न्यवहारे किमन्येन बाह्येनानुपपत्तिना ॥ ६ ॥

आभासमानैरेव 228अर्थैर्न्यवहारः , ते चाभासात्मका: सन्तु 229एव क्षति: । तत्किं बाह्येन कार्यं 230तावता 231लोकयात्रा-

219 ābhāsamānataiva: avabhāsamānataiva T, Jh has ābhāsamātrahetutaiva then corrected to ābhāsamānataiva. 220 sattā omitted in C N. 221 ābhāsat mataiva: ābhāsamataiva J, ābhāsataiva corrected to ābhāsatmataiva Jh, bhāsamataiva Ch, ābhāsatātmaiva J, ābhāsataiva corrected to ābhāsasmataiva N. 222 ekam omitted in E Ch. 223 ta- syāviśese 'pi cit. ibid. p. 82; tasya viśeṣa T. 224 kramena viśisyamāno 'rtho cit. and commented on ibid. p. 89 "kramena" iti vrttau kramo 'pi na niyataḥ, api tu bahubhir bhe- dair ābhāsan viśiṣṭān karoti iti viśisyamānābdam vrttigatam tātparyena vyācaśte ... yo 'rthaśabdo vrttau ... 225 vicitro: citra C N. 226 na omitted in T. 227 Cf. ibid. p. 89 "bo- dhavilakṣaṇa" iti aprakāśamān tvad eva āśayaḥ. 228 arthair cit. ibid. p. 130. 229 kṣatiḥ: kṣitiḥ J 230 tāvatā: tāvaty eva T. 231 Cf. ibid. p. 130 teṣu [ābhāseṣu] samāpte vyavahāre ...

20

समाप्ते: । २३२बाह्यश्रार्थ: प्रमाणबाधित: सावयवो विरुद्ध-

धर्माध्यासादेर्‌नि‌रवयवश्च दिक्षट्कृतयोगादेर्‌बहुश: २३३ ॥ ६ ॥

चिदात्मैव हि देवो‌ऽन्तःस्थितमिच्छावशाद् बहिः ।

योगीत निरुपादानसर्थजातं प्रकाशयेत् ॥ ७ ॥

चित्तत्वमेव २३४ईश्वरत्वात् २३५स्वात्मरूपतया ३६उपपन्नाभासनम्

अनन्तशक्तितत्त्वाद् २३७इच्छावशान् २३८मृदादिकारणं विनैव

बाह्यत्वेन २३९घटपटादिकम् २४०अर्थराशिं प्रकाशयेत् ॥ ७ ॥

अनुमानमनाभातपूर्वं नैवेष्टटमिन्द्रियं ।

आभातमेव बीजादेर्‌आभासाद्धेतुत्वस्तुनः ॥ ८ ॥

पूर्वं २४१धर्मान्तरस्थित एव २४२अर्थे २४३नान्तरीयकार्थदर्शन-

232 Cf. ibid. p. 130 bāhyaś cārthaḥ sāyayavaniravayavobhayavilakṣaṇo viruddha-dharmādhyāsādikam dūsanam avalambya pramāṇābādhitaḥ ... "niravayavaś ca paramānurūpo 'rthaḥ pramānābādhitō bahubhir bahubhir prakārair saṭkayōgādikam vikalpya" iti vrttigranthasya yojanā. T reads yugapadarthadikṣṭkayōgād eva (yugapad -rat satkena yugapāt yogāt paramānoḥ śadamśatā). 233 Ch adds iti syāt. 234 īśvaratvāt: īśva-rāt T; cf. ĪPVV II p. 146 "hi" iti yata evam devatvāt cidātmā .... 235 svātmarūpatayā cit. ibid. p. 146. upapannābhāsanam: upapannāvabhāsanam T, upapannābhāsarūpam E Gh C Ch J Jh Ñ T, upapannābhāsam Ñ. We have accepted the reading quoted and glossed ibid. p. 146 tasyā yad ābhāsanam, tat tāvad upapannam na prayatnasādhyam; ta-dabhedinaś ca bhāvarāśeh prakāśātmatvād upapannābhāsanam iti. 237 icchā°: icchādi° E Gh Ñ C Ch Jh Ñ T. 238 mrdādi°: aparā° T. 239 ghatapatādikam: ghatādim T. cf. ibid. p. 155. 241 °avabhāta°: °avabhāsa° E Gh Ñ C Ch Jh Ñ T T, but cf. ibid. p. 156 prathamam sūtrabhāgam samyamānirthanirūpanena sphuṭayati vrttikrt "pūrvāvabhātā ity ādinā 'anumanam" ity antena i "pūrvāvabhātantahsthitah" iti. 242 arthe: artho E Gh Ñ C Ch Ñ T, artha T. The reading arthe, required by meaning, is only found in Jh, as a correction of a previous artho; J has arthe, then corrected to artho. 243 nāntarīyaka° cit. ibid. p. 156, where the rest of the sentence is paraphrased: pūrvaśa-dopādāne 'pi hi antaḥsthitatvasyā avarakālatāvṛttir vyangyā / na ca vimarśanamātram anumanam, api tu praptipraṛttiyogyavasatuniścayarūpam, ity āha "nāntarīyaka" ity ādi. On the meaning and the necessity of tattaddeśakālādiyojanayā see also ibid. p. 162.

15.8 वशात् ततद्देशकालादियोजनया विमर्शनमनुमानम् । इन्द्रियं २४८अप्यनुमीयते किंचिन्मात्रं निमित्तं तच्च बीजाद्याभासाद् २४९आभासितमेव ॥ ८ ॥

आभास: पुनराभासाद्दाह्यास्यासीत् कथंचन । अर्थस्य नैव तेनास्य सिद्धिरनुमानतः: ॥ ९ ॥

२४६घटाद्याभासाद् बाह्यस्यानुपपत्तेर्नासीदाभासः , २४७ततस्तत्र नानुमानादपि सिद्धः ॥ ९ ॥

स्वाभिनश्चात्मसंस्थस्य भावजातस्य भासनम् । अस्त्येव न विना तस्मादिच्छामर्शः प्रवर्त्तते ॥ ९० ॥

चिदात्मनश्चेश्वरस्य २४८आत्मनीवाभेदेनार्थेऽपि प्रकाशोऽस्त्य- न्यथा प्रतिभासमानार्थे²⁴९कविषयो निर्मातृतामयो विमर्श इच्छारूपो न स्यात् ॥ ९० ॥

स्वभावमवभासस्य विमर्श विदुरन्यथा । प्रकाशोऽथोंपरक्तोऽपि स्फुटिकादिहोपमः ॥ ९१ ॥

प्रकाशस्य २५०मुख्य आत्मा प्रत्यवमर्शः , २५१तं विना २५२अर्थ- भेदिताकारस्य २५३अप्यस्य स्वच्छतामात्रं न त्वजाड्यं २५४चमत्कृ- तेरभावात् ॥ ९१ ॥

244 apy omitted in C. 245 ābhāsitam: ābhātam T 246 ghatādi° cit. ibid. p. 165 vrttau "ghatādi" iti ābhāsasya viśeṣanam. 247 tatas cit. ibid. p. 165. 248 ātmanīva cit. ibid. p. 167. 249 eka° cit. ibid. p. 167 ekasabdena avaśyam esaniyantritavā icchayā bhavitavyam ity āha. °arthakalāpaika° G Ch, also possible. 250 mukhya ātmā cit. ibid. p. 175. 251 tam vinā cit. ibid. p. 175 "artha" iti ! bhedita- grahanena pratikarma vyavasthāmātram sidhyatu atah. na tu ajādyam. 253 apy. only found in T T, is cit. ibid. p. 177 sūtre vrttau ca ayam apiṣṭadah svabhavāntarasamuccayārtha iti yavat. 254 camatkṛter abhavat cit. ibid. pp. 177-9.

आत्मात् एव चैतन्यं चित्क्रियाचितिकर्तृता- । तात्पर्येणोदितस्तेन जहात् स हि विलक्षणः ॥ १२ ॥

२५५आत्मदृश्यस्य २५६भावात्मकमप्येतज्जडाद् २५७भेदकतया विम- शरियं २५८मुख्यं रूपमुक्तं चैतन्यं २५९दृशिशक्तिरिचितिरिति । सा चेतनक्रिया चितिकर्तृतैव ॥ १२ ॥

चितिः प्रत्यवमर्शात्मा परावाक् स्वरसोदिता । स्वातन्त्र्यमेत्तन्मुरझयं तदैश्वर्यं परमात्मनः ॥ १३ ॥

२६०अभिन्नवाच्याद्या वाग् २६१एषा २६२नित्यचित्स्वरूपत्वेनान- यन्ता २६३परतन्त्रा , भावान्तरानपेक्षं शुद्धमेतत् स्वातन्त्र्यमैश्वर्यसंज्ञं ॥ १३ ॥

सा स्फुरत्ता महासत्ता देशकालाविशेषिणो । सैषा सारतया प्रोक्ता हृदयं परमेष्ठिनः ॥ १४॥

२६४स्फुरद्रूपता स्फुरणकर्तृता २६५अभावाप्रतियोगिनी २६६अ- भावऽयापिनी सत्ता भवत्ता भवनकर्तृता २६७नित्या देशकाल- २६८स्पर्शात् सैव २६९प्रत्यवमर्शात्मा चितिक्रियाशक्तिः । २७०सा स्फुरत्ता महासत्ता हृदयं परमेष्ठिनः

२५५ Cf. ibid. p. 184 vrttau dravyapadam ātmasāmānādhikaranyena uktam; o dravya-svao T. २५६ bhāvātmakam cit. ibid. p. 186. २५७ bhedakatayā: bhedatayā T. २५८ mukhyam: muktam T. २५९ drśśaktiś citir: citiśaktir drśir E Gh N C Ch Jh Ñ T (also possible). २६० abhinnavācyā cit. ibid. p. 190. २६१ eṣā: ameyā E. २६२ nityao: nityā Ñ, omitted in T; cf. ibid. p. 195 aham iti yat nityam citsvarūpam taddeśakālākārāraktrābhyām ādyantābhyām śūnyam nityoditam. २६३ aparatantra: aparatantrao E Gh Ñ C Ch Jh Ñ T, paratra G. २६४ sphuradrūpatā: sphurattā E Gh Ñ C T (In Ch sphurattā is corrected to sphura-drūpatā); cf. ibid. p. 206 vrttāv api sphuradrūpatā. २६५ abhāvāpratiyoginī: abhāvapra-tiyoginī T, bhāvapratiyoginī Kh. Ch has abhāvāo corrected to abhāvāc; abhāvāpratiyoginī is cit. and commented on ibid. p. 201. २६६ abhāvāo: bhāvao T. २६७ nityā cit. ibid. p. 201. २६८ oasparśāt oasparśā T. २६९ pratyavamarśātmā: pratyavamarsā C Ñ. २७० sā viśvātmanaḥ: sāro viśvātmano 'pi T.

I 5.14 विश्वात्मनः परमेश्वरस्य स्वात्म271प्रतिष्ठारूपा 272हृदयमिति तत्त्रागमे निगद्यते ॥ १४ ॥

आत्मानमित एवायं जेयीकुर्यात् पृथक्स्थिति । शेयं न तु तदोन्मुख्यात् खपुष्पवदेतस्य स्वातन्त्र्यता ॥ १५ ॥

273एतादृशशुद्धस्वातन्त्र्यवशान्नैष274 पृथग्वै लब्धप्रतिष्ठं वस्त्व् 275अवैति , अपि त्वज्ञेयमात्रानमनियन्न्रिततप्रभावतया जेयीकरोति । भिन्नज्ञेयसापेक्षत्वे ज्ञानकर्तृता म्लायेत् ॥ १५ ॥

स्वातन्त्र्यामुक्तमात्मानं स्वातन्त्र्यादद्यात्मनः । प्रभुरीशादिसंकल्पनिर्माणाय व्यवहारयेत् ॥ १६ ॥

276अत एव वैदैकीभावलक्षणपूर्णतामयात् स्वातन्त्र्यात् 277तदानीन्तनमेव वेदकमात्मानम् ईश्वरः शिवो वेदितेत्येवमादि-विकल्पैर् 278आभासयति भावनादिग्यवहारार्थम् ॥ १६ ॥

नाहन्तादिपरामर्शभेदादस्यान्यतात्मनः । अहंमृश्यतयैवास्य सृष्टेरितड्वाच्यकार्मवत् ॥ १७ ॥

279भावे नाहम्प्रत्यवमृश्यस्य प्रमेयत्वेनदन्ता । विम-र्शभेदे चाभासभेदे च प्रमातैवेशरशब्देन सृष्ट्टोशहं280परामर्श-

271 °pratiṣṭhārūpāi cit. ibid. p. 206: °rūpam T. 272 hrdayam iti cit. ibid. p. 206. 273 etadrśa°: etadr̥ga° E Gh N C J Jh N T. 274 E Gh N C J Jh N T add tadā, which seems unnecessary. 275 avaiti: vetti T. 276 ata eva cit. ibid. p. 209. 278 ābhāsayati cit. ibid. p. 211. 279 °bhāvenāhitapratyavamrśyasya: E Gh N C J Jh N T. 280 °parāmarśa°, conjectural emendation for °parāmarśo E Gh N C J Jh N T. °pratyavamrśyaparāmarśa° T.

व्यवधानेन केवलं, यथा क्रियादिशब्देन ²⁸¹पचत्यादिवाच्योऽर्थ: I 5.17

यथाहु: क्रियागुणजातिसंबन्धादिशब्दैर्न क्रियादय उच्यन्ते पच-

त्यादिमुखेनाभिधानात् । ईश्वरशब्दाद् आत्मशब्दाच्चाहमिति परामृशन्न् ²⁸²एवात्मनमवैति न साक्षात्, ²⁸³प्रतीतिस्वरसतयेव

स्मृत्यादिवासनावशेनमुखेनानुमेयत्वात् ॥ १७ ॥

मायाशक्त्या विभो: सैव भिन्नसंवेद्यगोचरा । कथिता ज्ञानसंकल्पाध्यवसायादिनामभि: || १८ ||

प्रकाशात्मन: ²⁸⁴परमेश्वरस्य मायाशक्त्या स्वात्मरूपं ²⁸⁵विश्वं भेदेनाभास्यते । ²⁸⁶तत: सैव ²⁸⁷चितिज्ञानम्²⁸⁸ ²⁸⁹अध्यक्षम् ।

²⁹⁰न तस्यैव भिन्नस्रोतस्तया स्मृति: ²⁹¹संकल्पाद्यवसाया मनोबुद्धिवृत्तिरूपत्वेsपि ²⁹²चितिरेव ॥ १८ ॥

साक्षात्कारक्षणेऽप्यसित विमर्श: कथमन्यथा । धावनाद्युपपद्येत प्रतिसंधानवर्जितम् || १९ ||

²⁹³साक्षात्कारलक्षणे ज्ञानेऽपि चितोऽर्थप्रत्यवमर्शोऽस्ति सूक्ष्म: ,

281 pacityādivācyo 'rthah | yathāhuḥ kriyāgunajātisambandhādiśabdair na kriyādaya ucyante pacityādimukhenābhidhānāt : pacityādivācyo 'rtha | yathāhuḥ kriyāgu-

najātisambandhādiśabdena kriyādaya ucyante pacityādivat | pacityādimukhenety abhidhānāt E Gh Ñ C Ch Jh Ñ T.

J, kriyādaya ucyante pacityādivat pacityādimukhenety abhidhānāt E Gh Ñ C Ch Jh Ñ T. 282 eva only in T. 283 pratitis tv asty eva : pratīti<ḥ>vsasyaiḥva T. 284 Cf. ibid. p.

218 yataḥ pratyagātmā vrttau parameśvara iti kathitaḥ. 285 viśvaṃ bhedenābhāsyate cit. ibid. p. 218; avaḥbhāsyate T. 286 tataḥ: tadā T, also possible. 287 citir: citikriyā T.

288 J adds upy. 289 adhyakṣam Cit. ibid. p. 217. 290 tasyaiḥva bhinnasya cit. ibid. p. 218. 291 The rest of the vrtti is paraphrased ibid. p. 218 nanu pratyagātmano manovrttir saṃkal-

paḥ, buddhivrrttir vā jñānam, manobuddhiyaksānāṃ ca jadatvam iti jñānādayaḥ katham ci-tivabhāvāḥ. - iti codyam garbhīkṛtya vrttāḥ apīśabdo yaḥ prayuktāḥ ...: rūpatvenāpi E Gh Ñ C Ch Jh Ñ T (J has otve corrected to otvena). 292 The quotation ibid. p. 218 is a little different: iti vrttau darśayati 'citisaṃskṛtiḥ' iti. 293 sāksātkāralakṣaṇe jñāne:

sāksātkārakṣanajñāne E Gh Ñ C Ch J Jh Ñ T.

I 5.19 294वाचनधावनादौ शीघ्रक्रिया तततदृश्यमानदेशाद्युपादित्साजिहासानुसंधानेन हि भवेत् ॥ १९ ॥

घटोऽयमित्यध्यवसा नामरूपातिरेकिणी । परे शक्तिरात्मेव भासते न तु वन्ध्या ॥ २० ॥

295अयमिति घट इति 296वाध्यवसायो भिन्नप्रकाशमाननामरूपा297तिरिक्तचितिशक्ति298मय एव298तमेवाभेदेन 300अवभासते

केवलं भिन्नसंवेद्यदेशकालानुरोधतः । ज्ञानस्मृत्यवसायादि सक्रमं प्रतिभासते ॥ २१ ॥

चित्तरत्वस्य मायाशक्त्या भिन्नं घटादि संवेद्यं ततददेशकालभिन्नं प्रकाशते 301यस्याभेदेनावष्टम्भाद 302विभिन्नदेशकाला303दिना ज्ञानस्मृत्याद्याभासते ॥ २१ ॥

इति ज्ञानाधिकारे पञ्चममाहिकम् ॥

294 The rest of the vrtti is cit. and commented on ibid. pp. 232-3 atha dhāvanasya ādipadasya ca prakārarthasya vyākhyānāya yat śīghrapadam, tasyā vyāvrtyāha “kriyāpi” iti śayanāsanādikā niranusandhānatratrotanādikā vā ! [...] i anyat deśarūpam yat dravyam, ādipadena śuklapītādilakṣaṇo gunah, yadi vā anyadeśam anyasvabhāvam ca vat dravyādi, tadviṣayā / “tad āha” iti etad vastu drśyamānasabdēna āhety arthah / anusandhānaśabdam vivṛnoti “tādrgicchāyojanena” iti lakṣaṇe trīyā / idam iṣṭam idam iṣyate idam ca eṣanīyam iti hānopādeccchānām anusandhānēna upalakṣitam yad eṣanīyatvam, tena vinā “tat” iti hānam upādānām ca pratikṣaṇam yad, tat na upapadyate iti sam-bandhaḥ. 295 ayam iti ghata iti cit. ibid. p. 241. 296 vā: ca T. 297 atiriktāḥ: “atirikta” Gh Ñ C Ch Jh Ñ T. 298 ātmeva: ātmaiva Gh Ñ Ch Jh Ñ T. 300 avabhāsate: ābhāsate Gh Ñ C Ch Ñ, bhāsate J. tīkā, glossed ibid. p. 271 “bodhyasyāvaṣṭambhaḥ” iti karmani saṣṭhī bodhyam avaṣṭabhnaīti, bodhaḥ svam rūpam tadekalīnam karotīti yāvat / “athavā” iti atra vyākhyāyām kartari saṣṭhī. bodhasyeti tu karmani. 302 Cf. ibid. p. 272 evam ca atra yo jan - jñānadirupataya ekajātiyatve 'pi vibhinnadēśakālādīnā uparaktam jñānam api jñānāntarāt smaraṇam ca smaraṇād anyad ābhasate iti samastasūtrayitītiatparyam samkṣipya [tīkākrṭ ...] 303 Cf. ibid. p. 269 esa ca artho vrttikṛtā spṛṣṭa ādipadena; p. 270 deśakālapadasya ca utra pakṣa upalakṣaṇatvam vyākhyātam.

षष्ठमाहिकम्

अहंप्रत्यवमर्शो यः प्रकाशात्मापि वाग्वपु: । नासौ विकल्पः स भुक्तो द्वयाक्षेपि विनिश्चयः ॥ १ ॥

304प्रकाशस्यात्मनि यह्मिति 305परावाग्रूपत्वात् 306साभिलापोऽपि 307स्वभावभूतः प्रत्यवमर्शो 308न विकल्प इत्युच्यते, 309स हि 310प्रतियोगिनिषेधपूर्वो निश्चयो न च 311अत्र प्रतियोगि-संभवः ॥ १ ॥

312तथा हि भिन्नयोरवभासो हि स्याद्घटाघटयोरितोः । प्रकाशस्येव नान्यस्य भेदिनस्तत्त्ववभासनम् ॥ २ ॥

प्रकाशाद् 313द्वितीयस्य भिन्नस्य प्रतियोगिनोऽपकाश 314संज्ञस्य 315अनवभासने प्रकाशे 316तरत्वं 317न स्यात् । तस्य 318अनाभासे न्यपोहनायोगादिकल्पताहानि: 319॥ २ ॥

304 prakāśyasyātmani: prakāśātmani C J Ñ. Cf. the quotation ibid. p. 274 aham ity ātmani. 305 parāvāgrūpatvāt cit. and commented on ibid. pp. 274-5; para° T. 306 sābhi-lāpo: sābhilāśo (then corrected to sābhilāpo) Ch J. 307 svabhāvabhūtah cit. and commented on ibid. pp. 278-9. 308 na kalpah cit. ibid. p. 274. 309 Cf. ibid. p. 274 "sa hi" iti vikalpah. 310 Cf. the paraphrase ibid. p. 274 pratiyoginisedhapūrvatvād eva niścayaśab-davācyatā iti yojanīyam. niṣedha° cit. ibid. nisedhaśabdena niraśanam vyācaṣṭe. 311 Cit. ibid. p. 274 "atra" iti viśudthe prakāśe and (through the tīkā) iti vṛttiga-tam vyācaṣte "viśuddhe" iti viśeṣaṇam māyāpramātṛ vyavacchedāya. 312 tathā hi, only found in T, cit. ibid. p. 280. 313 dvitīyasya bhinnasya cit. ibid. pp. 282-3. 314 Cf. ibid. p. 284 vyapadeśa iti tuccheva hi atuccam eva vyapadiśyate ity anena samjñāpadam vyākhyātam; samjñakasya E. 315 anavabhāsane: ābhāsane T, anavabhāsena Ch. 316 Cf. ibid. p. 287 itaraśabdena iti vṛttau. 317 na syāt cit. ibid. p. 288. 318 anābhāse: ana-vabhāse E Gh; cf. ibid. p. 288 pratiyogyanābhāse vā iti (from the tīkā). 319 Ch adds iti.

I 6.3 तथ च

तदतत्पतिभाभाजा मात्रैवातद्यपोहनात् ।

तन्निरूपणमुक्तो हि विकल्पो घट इत्यपम् ॥ ३ ॥

प्रमातर् एवं स्वतत्र्यात्रलीन एव ततदर्थाभासस्य प्रमातरि घट इति निश्चयो विकल्पो नाम न्यापारः ॥ ३ ॥

चितत्वं मायया हित्वा भिन्न एवावभाति यः ।

देहे बुद्धावथ पाणे कल्पिते नभसीव वा ॥ ४ ॥

प्रमातृत्वेनाहमिति विमर्शो$न्यन्यपोहनात् ।

विकल्प एव सु परप्रतियोग्यक्भासजः ॥ ५ ॥

चित्तत्वरयैवेश्ररसय मायाशक्त्या भेदावभासिनी शरीरेऽस्मिन्

बुद्धाव आन्तरे वा स्पर्शे तदुतीर्णे वाकाश इव शून्य एव

विकल्पिते$हमिति प्रमातृभावेन विमर्शः , तत्तदाभासमानशरीरादि

रादि प्रतियोग्यपोहनकरणाद घटोडयमित्येवद् विकल्प एव

॥ ४-५ ॥

320tathā ca, only found in T, cit. ibid. pp. 290 and 280. 321eva cit. ibid: p. 291.322otadatc: otadE Gh Ñ C Ch Jh Ñ T (in J otadatadis corrected to otad).323atadapohanena cit. ibid. p. 293; oyapohanaE Gh.324oavabhāsini: oavabhāsinīÑ.325āntare vā sparśecit. and commented on ibid. pp. 295-6; vāomitted in E Gh Ñ C Ch Jh T, āntarasparśaC, āntarasparśeÑ.326vikalpate: kalpateE Gh Ñ C Ch Jh Ñ T; kalpitekār.(kā.r.) is glossed p. 297 withvikalpanamātraparamārthah.

Cf. the phrase tasmin nijaniyatatūpe sarirādau yāvad ābhāsamānatvam, tāvad eva tasyasarīrāder avaśyam pratiyoginah sakasād apohah kriyate iti yo vrttryarthah ...; tad0T.p. 299 tasmin nijaniyatatūpe sarirādau yāvad ābhāsamānatvam, tāvad eva tasyasarīrāder avaśyam pratiyoginah sakasād apohah kriyate iti yo vrttryarthah ...; tad0T.

328pratiyogyapohanakaranātcit. ibid. p. 299; opratiyogyo 'pohanaoÑ.329ghato 'yamitivatcit. ibid. p. 299.330Ch adds iti.

कादाचित्कावभासे या पूर्वाभासादियोजना ।

संस्कारात्कल्पना प्रोक्ता साऽपि भिन्नावभासिनि ॥ ६ ॥

३३१विच्छिन्नविच्छिन्नेषु ३३२शून्यदेहाद्याभासभेदेषु पूर्वाभास-

३३३हंकाराख्ययानाम् ३३४ऐक्ययोजनान्तःपूर्वा३३४भासस्थिति३३६लक्षण-

संस्काराश्रिता पमातृव्यापाररूपा कल्पनैव ३३७प्रत्यभिज्ञान्या

॥ ६ ॥

तदेवं न्यवहारेऽपि पभुर्देहादिमाविशन् ।

भान्तमेवान्तरथोंऽघमिच्छया भासयेद्वहिः ॥ ७ ॥

३३८आदिसर्गे ३३९वा न्यवहारेऽपि वा ३४०महेश्वरो ३४१माया-

शक्त्या देहादिमातमत्स्वेन ३४२अभिनिविश्य प्रमातरं कुर्वन्ननन्तःस्थितं

विभान्तमेव ३४३तं तर्मथे ३४४क्रमेण बहिरूपं ३४५कर्तृशकृत्या

३४६भासयति । ३४७तथाभासनमेवोत्पादनम् । अस्य देहाद्यनविष्ठस्य

३४८तु स्वतो युगपदमिदमिति सर्वार्थप्रकाशः ॥ ७ ॥

३३१ T omits vicchinna°, but the repetition is confirmed ibid. p. 310 vṛttau vipsayā sarvatriko viccheda uktaḥ. ³³² śūnya° cit. ibid. p. 315. ³³³ E Gh Ñ C Ch J Jh Ñ T insert the last two members of the compound, but the paraphrase ibid. p. 310 follows the order of T: pūrvābhāsena yā aikyayojanā ādigrahanād ahamkāreṇa tatsamānādhikaraṇa ca caitra ityādināmnā ca, sā kalpanaiva iti vṛttau sambandhaḥ. ³³⁴ aikyayojanā cit. ibid. p. 312; aikyam C Ch. ³³⁵ °ābhāsa°: avabhāsa° E Ñ Ch J Jh T. ³³⁶ °laksana°: °laksanam C Ñ pratabhijñā cit. and commented on ibid. p. 314; pratabhijñāna° ³³⁸ ādi-sarge vā vyavahārakaṇe 'pi vā cit. ibid. p. 324. ³³⁹ vā: omitted in Jh T. ³⁴⁰ maheśvaro cit. ibid. p. 317. ³⁴¹ māyāśaktyā cit. ibid. p. 316. niveśya Ch. ³⁴³ tam tam: tattad° E Gh Ñ C Ch J Jh Ñ T. ³⁴⁴ krameṇa cit. ibid. p. 316. ³⁴⁵ kartrśaktyā cit. ibid. pp. 316 and 323. ³⁴⁶ bhāsayati cit. ibid. pp. 317 and 324. ³⁴⁷ tathābhāsanam evotpādanam: yathāvabhāsanam evāpūrvam evotpādanam E Gh J Jh T (in Jh evāpūrvam is added under the line), yathāvabhāsanam evāpūrvam evodbhāvanam Ch; I have accepted the reading T, which is confirmed by the quotation ibid. p. 324. ³⁴⁸ tu: tv asya T, tu omitted in T.

I 6.8

एवं स्मृतौ विकल्पे वाप्यपोहनपरायणे ।

ज्ञाने वाप्यनतराभासः स्थित एवेति निश्चितम् ॥ ८ ॥

सर्वसंवित्सु 349 सर्वार्थावभासः प्रमातृसंलक्ष्यतत्त्ववद 350 आनुरूप्येण प्रकाशमानः 351 सदैव स्थित एव ॥ ८ ॥

किंतु नैसर्गिको ज्ञाने बहिराभासनात्मनि ।

पूर्वानुभवरूपस्तु स्थितः स स्मरणादिषु ॥ ९ ॥

352 ज्ञाने बहिराभासनरूपे 353 सहज एव 354 चित्तत्त्ववस्यान्तरर्था 355 वभासः , स्मृत्यादौ तु 356 पूर्वानुभवात्मा । अत एव स्मृतिः संस्कारजोच्यते ॥ ९ ॥

स नैसर्गिक एवास्तित विकल्पे स्वैरचारिणि ।

यथाभिमतसंस्थानाभासनाद् बुद्धिगोचरे ॥ १० ॥

357 स्वतन्त्रस्तु विकल्पश्चक्षुराद्यगोचरमपि बुद्धिविषयतां 358 पादनेन यथारुचि पूर्वानुभूतत्वाविमर्शनने नवमेव तं तमर्थमाभासयति संनिवेश 359 विशेषं च । 360 तत्रासावर्थः सहज एवास्तित ॥ १० ॥

349 sarvārthāvabhāsah: sarvārthābhāsah Gh Ñ Jh T, sarvābhāsah J. 350 ānurūpyena: ānupūrvyena E Ñ, ānurūpena T; cf. ibid. p. 328 iyatā ānurūpyapadam vivṛtam. 351 sa-dā sthita eva: sthita eva Ñ, sthita eva sadā C. 352 jñāne bahirābhāsanātmani rūpe Ch Jh T. 353 sahaja cit. ibid. p. 329 (through the ṭikā) "sahaja" iti bhagavata eva svatantryād ullāsitah iti yāvat. 354 cittattavasyāntar cit. ibid. p. 330. 355 avabhāsah: ābhāsah Gh Ñ J Jh T. 356 pūrvaḥ pūrvārthaḥ uktam. 358 āpādanena yathāruci pūrvānubhūtatvāvimarśanena navam eva tam tam artham. 359 viśeṣaṃ ca. viśiṣṭam T. 360 tatra cit. ibid. p. 335 tatra iti vṛttau svatantravikalpo nirḍiṣṭah.

अत एव पथाभीष्टसमुल्लेखावभासनात् । ज्ञानक्रिये स्फुटे एव सिद्धे सर्वस्य जीवतः ॥ ९१ ॥

अपूर्वार्थनिर्माणज्ञान³⁶¹सामर्थ्याच³⁶२च विकल्प एव सर्वस्य सर्वज्ञत्वं³⁶३ सर्वकर्तृत्वे³⁶३च स्फुटम् ॥ ९१ ॥

इति ज्ञानाधिकारे षष्ठमाह्निकम् ॥

सप्तममाह्निकम्

या चैषा प्रतिभा तत्त्वपदार्थकमरुषिता । अकममनन्तचिद्रूपः प्रमाता स महेश्वरः ॥ ९ ॥

तत्त्वपदार्थकमाच्छादयितरश³⁶४चैषोऽन्तःस्थित एव³⁶४²आभासः³⁶६ सर्वसंवित्कालगाऽप्यक्रममानन्तचिन्मय आत्मसंज्ञः प्रमाता स्वाऽऽभूते प्रमेये निर्मातृतया महेश्वरशच ॥ ९ ॥

तत्तद्विभिन्नसंवित्तिमुखैरे कप्रमातरि । प्रतितिष्ठत्सु भावेषु ज्ञाते यमुपपद्यते ॥ २ ॥

³⁶७अनेकसंवित्स्रोतोमुखैरे कप्रमातृसिन्धुमुपलীয় भावभेदाः का-यकारसादृश³⁶८तात्पर्य्यं भजन्ते ॥ २ ॥

³⁶¹ C adds °kriyā°. ³⁶² ca cit. ibid. p. 336. ³⁶³ sphuṭam: sphuṭam J, cāsphuṭam Ñ. ³⁶⁴ ca omitted in T. ³⁶⁵ ābhāsah: avabhāsah TJ, but cf. ibid. p. 341 ābhāsa eva ātmā iti vrttigrantham .... ³⁶⁶ sarvasamvitkālavyālavāpī cit. ibid. p. 341. ³⁶⁷ Cf. ibid. p. 342, where the exposition of the kār. merges with that of the vrtti: ... api tu tās tā nilasukhādi-viṣayeṣu vibhinnā yā samvido niścayāntāḥ. tāny eva mukhāni upāyās tair nadīsrotahsthānīyair nenīyamānā hata-vāhenaiva aham ity ekamahāsamvitsamudre ābhimukhyena viśrāntim̐ yānti, tadā eṣām jñāteyam parasparam svasvāmitākārya-kāraṇabhāvādih samanvayo vyavahārātmā upapatyā ghaṭate. vahāram Ñ. ³⁶⁸ °vyavahāro°: °vya-

I 7.3

देशकालक्रमजुषामर्थानां स्वसमापिनाम् । सकृदाभाससाध्योऽसावन्यथा कः समन्वयः ॥ ३ ॥

३६९स्वरूपमवभासनं च भावानां स्वात्मपरिनिष्ठितमेव । युगपदेकाभासनिबन्धनरचैषां समन्वयः । सो३७०भिन्नः प्रमातृलीनतया कल्पते ॥ ३ ॥

प्रत्यक्षानुपलम्भानां तत्कदिनांशपातिनाम् । कार्यकारणतासिद्धिहेतुतेकप्रमातृजा ॥ ४ ॥

कार्यकारणभाव इव तत्सिद्धिरपि प्रत्यक्षानुपलम्भैरेप्रमातृमुखेन समन्वयमागत्य ३७१क्रियते, असमन्विताः प्रत्यक्षानुपलम्भाः क्रमिक३७२स्वविषयमात्र३७३ज्ञानक्षीणा नान्योन्यापेक्षोपलक्षणक्षमाः ॥ ४ ॥

स्मृतौ चैव स्वसंवित्तिः प्रमाणं स्वात्मसंभवे । पूर्वानुभवसद्भावे साधनं सैव नापरम् ॥ ५ ॥

पूर्वानुभव३७४स्वसंवेदनस्य ३७५अभावात ३७६स्मृतिस्वसंवेदनमेव साधनम्

369 svarūpam: svasvarūpam E T, but cf. ibid. p. 344 nanu kimartham ayam vrttau bhedena nirdeśah krtah / āha "svarūpa" iti svarūpe ābhāsane prakāśananiścayanarūpe yat bhedapratipādanam .... . 370 'bhinnaḥ: 'bhinna° T J Ñ. The latter would convey a meaning slightly different but cqually acceptable (however, as I have already pointed out, the omission of the visarga before a samyoga is frequent in T). The quotation ibid. p. 344 so 'bhinna itvādivrttih is not useful, since 'bhinna might be quoted as being the first part of the compound; but cf., in a similar context, ibid. p. 351 pramātrlagnatvena yad abhin-narūpam .... . 371 kriyate only found in T. E C Ñ have samanvitā and the same reading is cit. ibid. p. 362. but meaning requires asamavitāh T (cf. also āgatyāsamanvitā Gh N Ch Jh, °tāh T; in J agāryā is corrected to āgarya). 372 °jñāpa-nakṣīṇā: °jñānaparikṣīṇā Ch Jh, °jñānaparikṣīṇā corrected to jñānaparikṣīṇā T. 373 °jñāpa-samyedanāsya abhāvāt smrttisvamedanam evātraika°; °svasamvedanam evātraika° Ñ. 375 abhāvāt cit. ibid. p. 367. 376 smrtī°: smrtīh T.

32

अत्रैकार्थाभासमयप्रमातृरूपं स्मृतिस्वरूप इव प्रमाणम् । स्मृते: पूर्वानुभवाभासाभावे कार्यकारणभावासिद्धेर्‌न् कार्य-लिङ्गता ॥ ५ ॥

बाध्यबाधकभावाडपि स्वातन्त्र्यावेशाद् ज्ञाननाम । उदियादेकप्रमातृपरिनिष्ठते: ॥ ६ ॥

भिन्नस्वाभासमात्रनिष्ठानां ज्ञानानां को विरोधः , तत्कथं बाध्यबाधकत्वम् । एकप्रमातृविश्रान्तौ तु युक्तम् ॥ ६ ॥

विविक्तभूतलज्ञानं घटाभावमितिर्यथा । तथा चेच्छुक्तिकाज्ञानं रूप्यज्ञानाप्रमात्ववित् ॥ ७ ॥

इह भूतले घटो नास्तीति घटाभावज्ञानं केवलभूतलज्ञानमेव शून्यभूतलस्य घटाभावरूपत्वात् । तथैव यदि शुक्तिकारजतयोरपरस्परातमत्वाच् छुक्तिकाज्ञानं रजताभावज्ञानमिति प्रत्यक्षं बाधकम् ॥ ७ ॥

377 atra cit. ibid. p. 368 "atra" ity asya vyākhyānam "pūrvānubhave" iti. 378 smrti-svarupa iva pramānam / smrteh omitted in C Ñ. 379 The whole sentence is paraphrased and commented on (with some literal quotations) ibid. pp. 368-9 pūrvānubhavakāla eva samanantaraksane smrter udbhavo nāsti: smrtisamaye tu pūrvaksane yadi anubhavah prakāśeta vahnidhūmayor iva, grhyetāpi karyakaranabhāvah. smrtipūrvaksane tu na kathamcid anubhavasya prakāśah, na tu atra smrtikālā eveti mantavyam sahabhāvena prakāśe kāryakārana-tvāsiddheh / etadartham api ca vrttau pūrvagrahanam / smrter na kāryalingatā yatah pūrvāsya anubhavasya ābhāso nāsti yena tasyā smrtiṃ prati kāranabhāvah siddhyet iti visangatih. pūrvā°: pūrvārtha° T. 380 Cf. ibid. p. 372 vrttau ca virtau ca "svaś ca ābhāsaś ca" iti "svaś ca visayaś ca" iti yojyam. 381 omitir: omatir Ch Jh (this is also the reading found in ĪPV and ĪPVV). 382 chuktikā°: chukto Gh Ñ Ch J Jh T. 383 rajatābhāvajñānam: rajatajānanāpramātva-vedanam J, omitted in E Gh Ñ C Ch J Jh Ñ T; cf. ibid. p. 381 atra āha "tac ca" iti pratyaksam eva bādhakam.

33

I 7.8

नैवं शुद्धस्थलज्ञानात् सिध्येत् तस्याघटात्मकता ।

न तूपलब्धियोग्यस्याप्यत्राभावो घटात्मनः ॥ ८ ॥

केवलभूतलज्ञानादूतलस्याघटात्मकता सिध्यति , न तु तत्नाधारे भिन्नो दर्शयोःपि घटो नास्तीति ॥ ८ ॥

विविक्तं भूतलं शश्वद्रावानां स्वात्मनिष्ठते: ।

तत्कथं जातु तज्ज्ञानं भिन्नस्याभावसाधनम् ॥ ९ ॥

भूतलं 385अभूतलविविक्तं सदैव तत्कथं तज्ज्ञानं कदा-चिदेव तत्न भिन्नघटाभावं साधयेत् । भिन्नघटविविक्तता 387च भूतलस्य कादाचित्कं रूपं स्याद्यादि घटसहिततापि कदाचित् स्वरूपं भवेत् , न त्वेवम् । पदार्थौ द्वावेव स्वात्मपरिनिष्ठितौ , 388साहित्यं न तदतिरिक्तमुभयात्मकम् 389एकरूपम् । ज्ञानमेकं तूभयाभाससंसरगात्मकमप्यु 391एकाभासज्ञानान्तरा392भावरूपम् । वस्तु पुनः स्वात्मनिष्ठमेव परिच्छिन्नतीति न वस्तुबलेन 393प्रदेशदर्शनात् प्रदेशसिद्धिवद्घटाभावसिद्धिः । उभया394भावै-

कज्ञानात्मककार्यै395भावात् स्यात् । न चैवं न्यवधानेन 396प्र-तीति: प्रदेशदर्शनादेव 397तत्सिद्धे: ॥ ९ ॥

385 abhūtala: abhūtalam Ch. 386 T adds hi. 387 ca: api T. 388 sāhitya na: sāhitve-na E Gh C Ch J Jh Ñ T, sāhitye na N. 389 ekarūpam: ekam rūpam T. 390 tūbhayābhāsasamsarga: tūbhayāsamsarga T, tūbhayātmakasamsarga (corrected to tūbhayābhāsasamsarga Jh, tūbhayātmakam ekarūpam jñānam ekam tūbhayāsamsarga Ch. 391 ekābhāsa: ekābhāsam E Gh N C Ch J Jh Ñ T. 392 abhāva: bhāva Ñ. 393 pradeśadarśanāt: pradarśanāt C. 394 abhāsa: ābhāsātmakai ka E Gh N Ch Jh. ābhāsakaika corrected to ābhāsātmakai J and to ābhāsātma T. 395 abhāvat: abhāvavat E Gh Ch Jh Ñ, abhāvāt corrected to abhāvat J (vice versa in C T). 396 pratītih: pratīti C. 397 tat: tath (also possible) E Gh N C Ch Jh Ñ T.

किंत्वालोकचयोऽन्धस्य स्पर्शों वोष्णादिको मृदः । तत्रास्तित साधयेत्सस्य स्वज्ञानमघटातमताम् ॥ ९० ॥

प्रदेशेऽ्वलोक ३९८पूरं सन्तमसे मृद्मुष्णादिकं स्पर्शं वा घटरूपसस्पर्शाभावात्मकमनुभूयालोकादि घटाभावोऽत्रास्ति , घटो ना-स्तीति ३९९न्यवहर्तु युक्तम् ॥ ९० ॥

पिशाचः स्पादानालोकोऽप्यालोकाभ्यन्तरे यथा । अदृश्यो भूतलस्यान्तरनिषेध्यः स सर्वथा ॥ ९१ ॥

न चैवमालोकस्य पिशाचान्यत्वात् ४००तत्र पिशाच्चनिषेधप्रसङ्गः , स हृदृश्यो ४०१सन्यत्वे४०२पि यथा ४०३मृद्गोलकस्याप्यन्तरनिवार्यस्तथालोकस्यान्तरे । ततस्ततस्य ४०३अन्यमत इवासमन्वतेऽपि ४०४नादृश्यत्वादभावसिद्धिः ॥ ९१ ॥

एवं रूप्यविदाभावरूपा शुक्तिमतिरभवेल् । न त्वाद्यरजतज्ञप्तेः स्यादप्रामाण्यवेदिका ॥ ९२ ॥

शुक्तिज्ञानमेव रजतज्ञानाभावरूपं सिध्यति , तदानीन्तनशुक्ति-४०५ज्ञानानुभवेन न भिन्नस्यातीतस्य रूप्यज्ञानस्याप्रामाण्यम् ॥ ९२ ॥

३९८पूरम्: प्रसराम् E (Kh पूरम्) Gh N C Ch J Jh Ñ T. ३९९ व्यवहर्तुम्: व्यवaharanam Gh. ४०० tatra cit. ibid. p. 389 ... iti tridhā tatraśabdo vyākhyātah; tatrāpi E. ४०१'nyatve: 'nyathā'tve Kh; cf. ibid. p. 389.anyatvam saptamyantam yat vrttāv uktam. ४०२ mrdgolakasyāpy: bhūgolakasyā N Ch Jh T, भūगोलस्या E Gh C J Ñ. ४०३ anyamata: anyatra C. ४०४ nādrśyatvād abhāva: nādrśyatvābhāva: nādrśyatvābhāva E Gh N Ch Jh T, na drśyatvād abhāva C Ñ. ४०५ jñānānubhāvena na bhinnasyātītasyā rūpyajñānasyāprāmānye sā E Gh Ñ T (this is also the original reading of C Jh, then corrected to ijñānānubhavena bhinnasyā rūpyajñānasyāprāmānye sā: the latter is the reading of J), jñānānubhave jñānasyāprāmānye sā Ñ, jñānasyād aprāmānye sā Ch.

I 7.13

धर्म्यंसिद्धेरपि भवेद् बाधा नैवानुमानतः ।

स्वसंवेदनसिद्धा तु युक्ता सैकप्रमाातृजा ॥ १३ ॥

शक्तिकाञ्ञानकालेऽ च न पूर्व रजतज्ञानमस्ति । ततः स धर्मी न सिद्ध इति नानुमानेन बाधा । एकप्रमाातृ° स्व-संवेदने त्वेकदेशावष्टम्भयुबय° ज्ञानमयसंबन्धभासनात् सिद्ध्यति । पश्चात्° संवादः प्रत्यक्षस्वसंवेदने पूर्वस्यापि तस्य ° भासनादेकं प्रमाणमितरदन्यथेति भवति । संवादोऽप्येक° प्रमातृकृतः ॥ १३ ॥

इत्थमत्यर्थभिन्नार्थावभासखचिते विभौ ।

समनो विमलो वापि न्यवहारोऽनुभूयते ॥ १४ ॥

मायाशक्त्या भेदविषयोडयं सर्वो न्यवहारस् तथाज्ञानिनां शुद्धोऽज्ञानान्धानां तु मलिनस्तत्तद्° भिन्नार्थावभासभाजि भगवति ° संभाव्यते ° अनुभवेन ॥ १४ ॥

इति ज्ञानाधिकार सप्तममाह्निकम्॥

406 °mayā° omitted in C. 407 °svasamvedane: °samvedane E Gh Ñ C Ch J Jh Ñ T. N, °jñānamavasambandhabhāsanāt: °jñānam asambaddham E Gh J Ñ, °jñanasambaddha° N, °jñānasambaddham Ch T, °jñānam asambaddham corrected to °jñānasambandha-sambaddham C, °jñānasambandham corrected to jñānam asambaddham Jh. 409 °samvādah pratyakṣasvasamvedane: sasamvādapratyakṣe svasamvedane E Gh C Ch J Jh Ñ T, samvādapratyakṣe svasamvedane N. 410 bhāsanād: ābhāsanād E Ch Ñ T. 411 °pramāt°: °mātr° J. 412 tathājñānīnām: tathā jñānīnām E. but cf. ibid. p. 405 tathājñānīnām iti vrttau samastam padam. 413 tu omitted in C Ñ. 414 °bhinna°: °vibhinna° E C Ñ. 415 sambhāvyate cit. and commented on ibid. p. 405 sambhāvyate iti vrttāv uktam mūdhānāṃ jaghiti tathābhimānābhāve 'pi uktopadeśapariśīlanadiśā prakāśate eva ayam arthah iti darśayitum. 416 'nubhavena: 'nubhavane Ñ.

36

अष्टममाहिकम्

तात्कालिकाक्षसामक्ष्यसापेक्षा: केवलं क्वचित् । आभासा अन्यथान्यत्र त्वन्धान्धतमसादिषु ॥ १ ॥

आभासाः कदाचित् सन्निहितप्रत्यक्षैकक्षितो घटादयो मातृ⁴¹⁷न्यवहारहेतवः , अन्धतमसादौ तु पूर्वानुभवेधिताः ॥ १ ॥

विशोष sथावभासस्य सत्ताऽपि न पुनः क्वचित् । विकल्पेषु भवेद्विभवद्नूतार्थगामिषु ॥ २ ॥

स्मृत्युपेक्षारूपेषु प्रत्यक्षपृष्ठपातिषु ⁴¹⁸स्वतन्त्रेषु वान्येषु ⁴¹⁹विकल्पेषु कालत्रय⁴²⁰विषयेष्वर्थावभासो ⁴²¹'न्तस्तल्य एवाव-स्थितः ॥ २ ॥

सुखादिषु च सौख्यादिहेतुष्वपि च वस्तुषु । अवभासस्य सद्रावेऽप्यतीतत्वात् तथास्थितिः ॥ ३ ॥

सुखदुःखाद्याभासास्तत्साधनाभासाश्च सदैवान्तः सन्तोडपि न तथाह्हादादिमयों स्थितिं कुर्वन्त्यतीतत्वात् बहिस् ⁴²²तदानीमभावात् ⁴²³तदात्वाविशिष्टानां च तथाकारित्वात् ॥ ३ ॥

⁴¹⁷ vyavahārahetavah: vyavahārahetor E Gh N C Ch J Jh Ñ T; vyavahāra cit. ibid. p. 408. ⁴¹⁸ svatantreṣu vānyeṣu: asvatantreṣu vānyeṣu C Gh Ñ T, asvatantreṣu nyeṣu N, va svatantreṣu vānyeṣu J, vā svatantreṣv anyeṣu T. ⁴¹⁹ vikalpeṣu omitted in Gh. tika) tadānīm ity anena hi vartamānā bāhyaniṣṭhaiva iti darśitam. ⁴²² Cf. ibid. p. 410 (perhaps from the ⁴²⁰ ovisayeṣv: oviśeṣv J. ⁴²¹ 'ntas: 'nta eva T. ⁴²³ tadātvaviśiṣṭānām: tathātvam aviśiṣṭānām T; cf. ibid. p. 410 tadātvam vartamānātā

गाढमुल्लिखितमाने तु विकल्पेन सुखादिके ।

तथास्थितिसतथैव स्यात् स्पुटमस्योपलक्षणात् ॥ ४ ॥

⁴²⁴कुतश्चित् प्रयत्नविशेषात् ⁴²⁵स्वतन्त्रविकल्पोल्लिखितं

स्पुटमेव सुखादि ⁴²⁶जातं विकासादिहेतु: ॥ ४ ॥

भावाभावावभासानां बाध्रतोपाधिरिष्यते ।

नात्मा सत्ता ततस्तेषामान्तराणां सतां सदा ॥ ५ ॥

⁴²⁷सर्वेषामाभासानां ⁴²⁸भावाभावविष्याणामभिरूपत्वेऽपि सत्ता-

स्त्येव स्मृत्यादौ , बाध्रत्वं हि तेषामुपाधिर्न स्वरूपम् । अभाव-

⁴²⁹भासस्यान्तःसत्तयामपि ⁴³⁰बहिरभावातथात्वम् ॥ ५ ॥

आन्तरत्वात् प्रमाैक्ये नैषां भेदनिबन्धना ।

अर्थक्रियापि बाध्रत्वे सा भिन्नाभासभेदतः ॥ ६ ॥

अन्तश्च सर्वेषामेव नीलसुखाद्या ⁴³¹भासानां ⁴³²सदा सत्त्वेऽपि

प्रमातृमात्ररुपत्वात् कार्यकारणादिभेदा ⁴³³श्र्या नार्थक्रिया , प्रमा-

तुर्भेदे ⁴³⁴अपि ⁴³⁵बौद्धचाक्षुषत्वादिभेदनाभासभेदादाभासाश्रितार्थका-

रितापि ⁴³⁶भिद्यते रूपादीनाम् ॥ ६ ॥

⁴²⁴kutaścit ibid. p. 410. ⁴²⁵svatantra cit. ibid. p. 410 svatanagrahaṇam sma-

rananirākarana. ⁴²⁶jāitam could also be considered as being compounded with sukhādi

with the meaning of 'multitude, collection', without this involving a notable variation of

the general meaning. ⁴²⁷sarveṣām ābhāsānām: bhāvābhāvāvabhāsānām (which is the

same as the first pāda of the kār.) E Gh N C Ch Ñ T, bhāvābhāvāivabhāsānām corrected

to sarveṣām ābhāsānām Jh. ⁴²⁸bhāvābhāvavisayānām is paraphrased and commented on

ibid. p. 412 nanu evam vrttā abhasānām bhāvābhāvā viṣayatvena katham uktaḥ ... and

pp. 414-15 nanu ca abhāsā viṣayaḥ, yato grāhyāḥ, tat katham teṣām vrttau bahuvrīhin

viṣayitvam darśitam. ⁴²⁹abhāsasya: °abhāvasya Ñ. ⁴³⁰bahirābhāvāt: bahirābhāsāt T;

cf., in some context, ŚDVr p. 130 hāye prakāśanābhāvād ... . ⁴³¹ābhāsānām:

°ābhāsabhedānām T (also possible). ⁴³²sadā sattve: sadasattve E Gh Ñ C Ch Jh Ñ T

⁴³³āśrayā nāśthe: °āśrayanāsthe Cḥ. ⁴³⁴'ni: hi E Gh C Ñ, ni corrected to hi Gh.

⁴³⁵bauddhac: boddhya T. ⁴³⁶bhidyate: vidyate T.

चिन्मयत्वे सभासानामन्तरेऽ स्थितिः सदाः । माययाभासमानानां बाह्यत्वाद बहिरप्यसौ ॥ ७ ॥

चित्स्वरूपत्वेनाभासानां सदान्तस्तत्त्व एव स्थितिः । । मायाशक्त्या बहिः ³⁹प्रत्यक्शत्वेन ³⁸प्रकाश्यमानेषु भावेषु प्रकाशा³⁹न्यतिरिक्तेषु ⁴⁰बहिराभास इत्युच्यते, ⁴¹तदापि तेषामान्तरतत्त्वमेव सिसृक्षादौ तु भावानामपि ⁴² ॥ ७ ॥

विकल्पे योस्यमुल्लेखः सोऽपि बाह्यः पृथक्पृथः । प्रमातृकात्म्यमान्तर्य ततः भेदो हि बाह्यता ॥ ८ ॥

विकल्पे घटाद्युल्लेखश्चक्षुराद्यगोचरोरपि पृथग्भासाद बाह्य एव । ⁴³अहंविमर्शो ध्यन्तरतत्त्वम्, इदमिति तु बाह्यता । एवं च घटादीनामुभयी ⁴⁵बाह्यता बाह्यान्तःकरणद्वयीवेद्यता, ⁴⁶सुखादेस्व ⁴⁷एकान्तःकरणवेद्यतैव ॥ ८ ॥

⁴³⁷pratyakṣa⁰ cit. and commented on ibid. p. 422. ⁴³⁸prakāśyamānesu: pra-kāśamānesu E Gh. This is also the reading cit. ibid. p. 421, but all the argumentation that accompanies it presupposes prakāśyamānesu: "ata eva iti yato 'bhimānamātrasāram māyīyam ābhāsate svātantryam eṣām nīlādinām, tato vrttau pāratantryam eṣām nirūpi-tam. svātantryam tu na sprśtam iti darśayati "antarnī tā" iti svatantrasyaiva hi pradhāna-kriyāyām tannisthavighnasambhāvanānirāsopayogipraisādikriyāntaraviṣayam pārata-ntryam visaye 'pi prasarativa mukhyakriyārūpe, na tu paramarthatah pradhānakriyāyām asambhavatsvātantryah praiṣādibhir abhisambadhyate ! nanu sūtrānusāreṇa vrttau "ābhāsyamānesu" iti vaktavyye "prakāśalśya?mānesu " iti kā vācoyuktitḥ ' .... ⁴³⁹ °avya-ktiriktesu: °vyatiriktesu T. ⁴⁴⁰bahirābhāsa cit. ibid. p. 422. ⁴⁴¹tadāpi cit. ibid. p. 423 and glossed with pratyaksatve 'pi. ⁴⁴²T adds iti. ⁴⁴³ahamvimarśo cit. ibid. p. 426. ⁴⁴⁴ca omitted in E Gh Ñ C Ch Jh Ñ T. ⁴⁴⁵bahyatā omitted in E Gh Ñ C Ch Jh Ñ T. ⁴⁴⁶sukhādes tu cit. ibid. p. 429, where also ghatādīnām is mentioned. ⁴⁴⁷ekāntah⁰: ekānta⁰ Ch.

I 8.9

उल्लेखस्य सुखादेश्च प्रकाशो बहिरात्मना । इच्छातो इभर्तृरध्यक्षरूपो क्षयादिभुवां यथा ॥ ९ ॥

उल्लेखसुखदु:खलज्जादीनां चाक्षुषाणां रूपादीनामिवेश्वरस्य शक्त्या साक्षात्करणरूपो बौद्ध: प्रकाश: ॥ ९ ॥

तद् ऐक्येन विना न स्यात् संवित् लोकपद्धति: । प्रकाशैक्यातदेकत्वं मातैक: स इति स्थिततम्

ततद्विभिन्नसंविदनुसंधानेन हि व्यवहार: । एकस्य प्रकाशात्मा तदनुसंधानरूप: स एव चैक: प्रमाता परमात्मसं: ।। १० ।।

स एव विमर्शवेन नियतेन महेश्वर: । विमर्श एव देवस्य शुद्धे ज्ञानक्रिये यतः ॥ ११ ॥

स परमात्मा चिद्रूपो विमर्शारूयेनैव मुख्यस्वभावेनाऽऽय्याभिचारिणा महेश्वर: । चित्तत्त्वस्य विश्वात्मन: शिवसंज्ञस्याह-

448 'ksya°: 'ksa° E. which is also the reading of IPV (KSTS Ed.), khyā° N; the reading 'ksya° occurs also in Bh and IPVV and is confirmed by the paraphrase ibid. p. 430 aksyādibhūmikādirūpānām ... . 449 The text of the vrtti in E Gh N C Ch Jh Ñ T appear at first sight corrupted and full of gaps: ullekhasya sukhaduhkhādīnām ceśvarasya sākṣātkārarūpo bauddhah prakāśah. I have accepted integrally the text of T, even though one would expect an explicit reference to bahirātmanā; the addition of °lajjā° is indirectly confirmed ibid. p. 430 "atadrupāc ca" iti lajjā hi na sukham, na duhkkham, cittattiviśesaś ca. The text of J is very close to T (only, it omits °lajjā° and reads sākṣātkara°). 450 aikyena: ekena Ch T. 451 na syāt: nāsti Gh N C J Jh T. 452 hi vaya-vaharah. my conjectural emendation: hi na vyavaharah T, vyavahāra E Gh N C Ch J Jh Ñ T. 453 ekas ca: ekas eva E (ekas ca also in K Kh). 454 ca omitted in E Gh N C Ch J Jh Ñ T. 455 paramatma° cit. ibid. p. 433. 456 °ākhyena: °ākhya° T. 457 °samjñasyāhamvi-marśanam: °samjñakasyāhamvimarśa J.

40

विमर्शनमेव 458शुद्धे ज्ञानक्रिये, भिन्नाभिन्नज्ञेयकार्य459गते

त्वीश्वरस्य 460शुद्धाशुद्धे, भिन्नार्थ461विषये तु पुंसः सत्त्व-रजो-

'वृत्तिरूपे प्रकाशप्रवृत्तिसंझे तमसा 462संकुचिते अशुद्धे एव

॥ 11 ॥

इति ज्ञानाधिकारेsष्ठममाहिकम् ।।

इति ज्ञानाधिकारः ॥

458 śuddhe: viśuddhe T. 459 °gate tv īśvarasya: °gatve tv īśvarasya C Ch Jh Ñ, °gatve (corrected to °gate) tv īśvarasya Ñ, °gatve tv īśvarasya J. 460 Cf. ibid. p. 437 evam ahantāyāḥ śuddhāśudhatvam vyākhyāya, aśuddhatvam vyācaṣṭe. 461 viṣaye tu °viṣayitve E Gh Ñ C Ch J Jh Ñ T. 462 samkucite: samkocite T.

II 1.1 क्रियाधिकार:

प्रथममाहिकम्

अत एव यदप्युक्तं क्रिया नैकस्य सक्रमा । एकेत्यादि प्रतिक्षिप्तं तदेकस्य समर्थनात् ॥ १ ॥

एकचित्तत्वसमर्थनादेकसंबन्धी व्यापार एक एवेति क्रियाप्यकृतदूषण। ॥ १ ॥

सक्रमत्वं च लोकक्रियाः क्रियाया: कालशक्तितः । घटते न तु शाश्वत्याः प्रभाव्न्याः स्यात् प्रभोरिव ॥ २ ॥

मायाशक्तेर्भिन्नभावावभासानां क्रिया कालशक्तिवशात् सक्रमा न त्वात्मविमर्शरूपानादिनिधना प्रभोः 'स्वभावभूता ॥ २ ॥

कालः सूर्यादिसंचारस्ततत्पुष्पाविजन्म वा । शीतोष्णे वाथ तल्लक्ष्यः क्रम एव स तत्वतः ॥ ३ ॥

सा सा प्रसिद्धा क्रिया कालः शीतादि वा तदुपलक्षितः

1 svabhhāva° : svaprābhava° T. 2 śītosne : śītosnam T; śītosne is also quoted in MMP p. 131. 3 There is a direct allusion to sā sā in ĪPVV III p. 6: ādiśabdopāttam ca yat sūtre tat sanksepavṛttau vīpsayā. vyākhyātam. 4 śītādi : śītādir T Ñ. 5 °upalakṣitah : °upa-laksita° T.

सर्वेभिन्ना6वभासमानभावोपाधि9भूतः कम एव वासौ तस्यै- II 1.3 वोपयोगात् ॥ ३ ॥

कमो भेदाश्रयो भेदोस्याभाससदसत्त्वत: । आभाससदसद्वे तु चित्राभासकृतः प्रभो: ॥ ४ ॥

अन्योन्याभासशून्येभिन्नभावावभासवैचित्र्यक्रियैव प्रभोर् 10भवेषु कमहेतु: ॥ ४ ॥

मूर्तिवैचित्र्यतो देशक्रममाभासयत्यसौ । क्रियावैचित्र्यनिर्भासात् कालक्रममपीश्वर: ॥ ५ ॥

11अनेकक्रियान्योन्यभेदाभासादेशक्रमः 12क्रियामुखेन कालक्रमोsपि । 13एकस्य तु भावस्य तत्जन्मसत्ता14विपरिणामादि-क्रियाभेदात् कालक्रम एव ॥ ५ ॥

सर्वत्राभासभेदोसपि भवेत् कालक्रमाकर: । विच्छिन्ननभास: शून्यादेर्‌मातुर्भातस्य नो सकृत् ॥ ६ ॥

सर्वत्र 14अर्थे विचित्रोसवभास: शून्यदेहादेरव प्रमातु: काल-क्रमात्‌

6 oavabhāsamānā :obhāsamānā Ch. 7 obhūtaḥ :obhūta T. 8 eva vāsau : evāsau E Gh Ñ C Ch J Jh Ñ T. 9 Cf. ibid. p. 11 vrttau bhinnāvabhāsavaicitryaśabdābhyām na-umsakaikaśeso 'tra dvayasya upayogitām darśayati. 10 bhāvesu cit. ibid. p. 11. 11 Cf. e paraphrase ibid. p. 12 anekasmin vastuni deśakramo grhāṇganavat, kālakramo 'hakāramallikādivat / ekasmims tu na deśakramo dūratāder abhāvāt / vaitatyam api 'pimsānām / tad ekatra kālakrama eva pacyamānāmraphalavat iti vrttitātparyam. 12 kriyāmu-khena cit. ibid. p. 15. 13 ekasya cit. ibid. p. 16. 14 oviparināmā : oparināmā E Gh Ñ C Ch J Jh Ñ T (cf. Nirukta I.2.8-9 sad bhāvavikārā bhavantīti ha smaha bhagavān varsayinih jāyate 'sti viparinamate vardhate 'paksīyate vinaśyati). ibid. p. 20 sūtrānupāttam arthapadam vrttau prakṣipya ...

II 1.6 क्रमाभास'हेतुः; स हि 'प्राक्कालो न तथा तदानीं 'भासते 'स्मृतिं विहाय 'स्ववर्तमानापेक्षया चासौ भूतभविष्यत्ते व्यवहरति । सकृद्विभातस्य तु 'भासनक्रियाविच्छेदाद् 'आवृत्तिगणनाभावात् नात्मनोऽर्थेऽपि कालभेदः ॥ ६ ॥

देशक्रमोऽपि भावेषु भाति मातुर्भितात्मनः । स्वातमेव स्वात्मना पूर्णा भावा भान्त्यमितस्य तु ॥ ७ ॥

परिमितं प्रमातारमपेक्ष्य भावास्ततो 'न्योऽन्यं च भिन्ना दूरादिव्यपदेशभाजश्च, प्रकाशैकरूपत्वेन . त्वेश्वरस्य न 'क्वाप्य-प्रकाशसंभवात् प्रमाणावपि । प्रकाश'घनस्य 'नात्मनो भिन्नं दूरं वाऽव्यवधानेन 'वा किंचिद्रभाति ॥ ७ ॥

किंतु निर्माणशक्ति: साप्येवंविदुष ईशितुः । तथा विज्ञातृविज्ञेयभेदो यदवभास्यते ॥ ८ ॥

एवं पूर्णतया प्रकाशमानस्य 'अपि परमेश्वरस्य सैषा सृष्टि-

16 hetuh : krt T. 17 prākka"lo : prākāle T; cf. ibid. p. 21 "śūnyadehādyābhāsa-prākkā"la" iti bahuvriih. 18 Cf. ibid. p. 21 "smṛtim" [from the tikā] iti smṛtau kevalam bhāsate ity arthah. 19 sva° omitted in T. 20 vibhātasya° : vibhāvasya T. 21 bhāsana° avyāvartti° Ch T, abhyāvartti° Jh, avyāvṛtti° Ñ; āvṛtti° occurs many times ibid. pp. 21-22. 23 'nyonyam ca bhinnā: 'nyam ca bhinnā T, 'nyonyam bhinnā E Gh C, 'nyonyam ca Ñ. 24 sambhavāt : sambhavah C Ñ, sambhavāt corrected to sambhavah Gh J (in the preceding prakāśaikarūpatvena tv īśvarasya na kvāpy a is added in the margin), svabhāvābhāvāt T. Also the reading sambhavah would be acceptable, but Abh. refers explicity to the ablative case: ibid. p. 23 "na kvāpi aprakāśa" iti vrttivākyasya iyam yojanā "sakṛdvibhāto 'yam ātmā pūrṇasyāsa na kvāpy aprakāśanasambhavah" iti sāra-svatasamgrahasūtraikadeśārtho [...] pañcamīm vidhāyā hetutayā vrttāv uktah. 26 ātmano na : ātmano E Gh Ñ C Ch J Jh T, ātmanā Ñ. 27 vā omitted in T. 28 api cit. ibid. p. 27.

शक्तितः सञ्ज्ञातृज्ञेयस्वभावा भावाः स्वतोन्योन्यं च ²विभागे-नावसीयन्ते, ³0न च तथावसायेन ³1तस्य स्वरूपस्थितिसितरोधीयते ॥ 1.8 ॥ ८ ॥

इति क्रियाधिकारे प्रथममाहिकम् ॥

द्वितीयमाह्निकम्

क्रियासम्बन्धसामान्यद्रव्यादिक्कालबुद्धयः । सत्त्या: स्थैर्योपयोगाभ्यामेकानेकाश्रया मताः ॥ १ ॥

³2क्रियावदन्येsपि सम्बन्धादय एकानेकविषया अपि ³3सत्या-भासा:, ³4सर्वदोपयोगिनामेषामर्थवत्त्वेनापरिहरणीयत्वात् ॥ १ ॥

तत्रैकमान्तरं तत्त्वं तदेवेन्द्रियवेद्यताम् । सम्प्राप्यानेकतां याति देशकालस्वभावतः ॥ २ ॥

³4अभिन्नमेव ³8तत्त्वमन्त ³9बोंहिराभासभेदादेकं, बहिर्देश-कालस्वभावभेदाभास ³6संभेदमयै ³7कैकस्वलक्षणाभासानामनेकत्वात् ॥ २ ॥

29 Cf. ibid. p. 26 anena vibhāgāvasāyena .... 30 na ca cit. ibid. p. 26. 31 tasya omit-ted in E Gh N C Ch Jh N T. 32 Cf. ibid. p. 29 kriyātra drṣṭāntatvena upāttā. 29 "sarvadopayoginām" iti sthairyam uktam "arthavattvena" iti arthakriyopayogah. 35 T antarabhinnam tattvam samvinmātratayā, māyāpramātuh punah antahkarāṇābhāsāt bahyakaranābhāsāc ca bahirbhedāt vastu phalata ekānekam / anekatve hetuh "bahih" ity āditi vrttisamgatih; J has etat instead of eva. 37 bahir° cit. ibid. p. 45. 38 °sambheda-mayakasvalakṣaṇābhāsānām : °sambhedabhāsānām Jh T. 39 °ekaika° : °eka° E Gh N C Jh.

II 2.3

तदद्धयालम्बना एता मनोडनुण्यवसायि सत् ।

करोति मातृण्यापारमयीः कर्मान्दिकल्पना: ॥ ३ ॥

40 मध्यस्थतयान्तर्बहिस्तत्वविषया मानस्य: क्रियादिकल्पना:

41 प्रमातृण्यापारूपा: ॥ ३ ॥

स्वात्मनिष्ठा 42 विविक्ताभा भावा एकप्रमातरि ।

अन्योन्यन्वरूपैक्ययुज: सम्बन्धधीपदम् ॥ ४ ॥

राज्ञ: पुरुष इत्यादिसम्बन्धधियोन्त: समन्वयादैकयं 43 बहि:

संबन्धिभेदं चालम्बनते ॥ ४ ॥

जातिद्रव्यावभासानां बहिरप्येकरूपताम् ।

न्यक्त्येकदेशभेदं चाप्यालम्बनते विकलपना: ॥ ५ ॥

गावश्चैत्र इति च मतयो बहिरपि गोमात्रैकघनपुरुष-

विशेषाकाराभासैवयं स्वलक्षणावयवाभासबहुत्वं च परामृशन्ति

॥ ५ ॥

क्रियाविमर्शविषय: 44 कारकाणां समन्वय: ।

अनध्यवसितस्वभावान्नध्यालम्बा दिगादिधी: ॥ ६ ॥

काष्ठस्थालीदेवदत्तौदानां पचतीत्यान्त: समन्वयाद बहिर्भेदाच्चै-

कानेकविषया क्रिया 45 मति: । देशकालक्रमोदपि 46 भावानामवृध्दय-

40 Cf. ibid. p. 47 kayor madhyastham mano vrttau uktam iti śankitvā grāhya-

grāhakayor iti darśayati. 41 pramātr° : mātr° E Gh Ñ C Ch J Jh Ñ T. 42 viviktābhā :

viviktā vā E. viviktatā (but in Malayalam script t and bh are rather similar) T.

43 sambandhi° : sambandha° E Gh Ñ C Ch Jh Ñ T. 44 kārakāṇām : kāraṇānām Gh N

C Ch J Jh T. 45 Cf. ibid. p. 62 evam kriyām ekanekarūpam upapāadya vrttigranthasya

"matih" ity antasya ... . 46 E Gh Ñ C Ch J Jh Ñ T add bhāti, which is very probably to

be expunged.

46

वधिमद्रूपाणामन्योन्यापेक्षः ४७संबन्धभेद ४८एव तथा ४९एवैकनेकमयः । जातिद्रव्याक्रियासंख्यादिमतयः सर्वा एव सम्बन्धविशेषसमवायविषया एव ॥ ६ ॥

एवमवैषयिकी स्थान्मातुरथोक्रियार्थेन । भेदाभेदवतार्थेन तेन न भान्तरीडशी ॥ ७ ॥

५०एकानेकरूपैरेव कियादिभिरेवमाभासानुगुण्यरूपः प्रमातुः तदर्थनोऽथक्रियासंबन्धः । ततोन तद्बुद्धयो भ्रान्ताः ॥ ७ ॥

; इति क्रियाधिकारे द्वितीयमाह्निकम् ॥

तृतीयमाह्निकम्

इदमेतदार्गिल्येवं यद्रशाद् न्यवतिष्ठते । वस्तु प्रमाणं तत् सोऽपि स्वाभासोऽभिनवोदयः ॥ ९ ॥

सोऽन्तस्तथाविमर्शात्मा देशकालाद्यभेदिनि । एकाभिधानविषये मितिरेवस्तुन्यनाधिता ॥ ९ ॥

यदायत्ता हि ५१वस्तुनः स्वरूपेण नित्यताविशेषणे ५२वा

47 sambandha° : samanvaya° E Gh Ñ C Ch J Jh Ñ T, but cf. ibid. p. 61 yā dikkāla-buddhih, sāpi ... sambandharūpatvapradhānaiva. 48 eva : evam T. 49 eva omitted in T. 50 In the KSTS Ed. of ĪPV (vol. II pp. 57-8) fn. 110 is constituted by a long passage drawn most probably from the țikā, where the beginning of the vrtti is quoted: tad āha vrttau - ekānekarūpaih kriyādibhir iti. 51 vastunah and viśeṣanair cit. ibid. p. 77. 52 vā : yā E.

II 3.1-2

53न्यवस्थाप्यमानता स प्रमातुः स्वत्पेनापूर्ववस्तु54वभास एव स्थितः प्रमाणम् । 45स एव चाभासोऽयमिति नित्य इति वा तथैव पृथक्तयाभिनवैवेन च प्रमातृयुपारूढो विमर्शरूपताम!पन्नो 46विमर्शभेदानुसार्येकैकशब्दवाच्ये पृथक्पृथगेव तिर:कृतदेशांदिभेदे तस्मिन्नाभात एव सामान्यरूपे वस्तुनि स्वकार्थंक्रियाप्राप्ते प्रमाणांतरेण 57अबाधित59स्थैर्य्या प्रमितिः प्रमातृण्यापारः ॥ १-२ ॥

यथारुचि यथार्थितवं यथान्युत्पत्ते भिद्यते । आभासोऽप्यर्थ एकसिन्न्ननुसंधानसाधिते ॥ ३ ॥

एकसिन्न्न 60एव चैकप्रत्यवमर्शसामथ्र्योण्पपादिते 61वस्तुनि स्वेच्छावशादर्थितवानिरोधाद्वा नैपुणवशाद्वाभासभेदः ॥ ३ ॥

53 A reference to vyavasthāpyamānatā can be found ibid. p. 74 pramātrādhinatām tasyā prakatīkartum vrttau nicprayogah. 54 avabhāsā eva sthitah cit. and commented on ibid. p. 76 "avabhāsā eva sthitah" iti vrttipadair ābhāsavyatiriktam na anyat sthitam rūḍham satyam kimapi asti - iti vadadbhir abheda ukta iti yāvat. 55 sa eva cābhāso 'yam iti nityam iti vā". tat pumnirdeśe ka iti nityā vā cit. and commented on ibid. p. 77 nanu "vastuno viśeṣanaih" iti upakrāntā vrttih, tata evam vrttyantarena bhāvitavyam "idam iti" "nityam iti vā", tat pumnirdeśe ka āśayah ! ayam iti āha "ābhāsa eva iti ! sa eva cābhāsah" iti hi vrttiganthe ya ābhāsah pramānatvena uktah, tam eva atīnikatām apeksya nirdeśaḥ prakrtah pramāṇaprameyayor abhedam tāttvikam ācaṣṭe ity arthah. In the KSTS Ed. of the ĪPV (vol. II pp. 69-70) fn. 46 is constituted by a passage from the ṭikā, where the same phrase from the vrtti is quoted: tad eva darśitam vrttau sa eva cābhāso 'yam iti nityā iti veti. 56 vimarśabhedānusāryekaikaśabdavācye paraphrased ibid. p. 78 iyatā vimarśaviśeṣam anusarati avaśyam ya ekaikah śabdas tadvācya iti yā vrttih. 58 abādhita° : abādhitā Ch T. 59 sthairyā : °sthairyam T, °sthairyah Gh N. 60 eva cit. and commented on ibid. p. 107 vrttāv evena apyartha evā ākhyāta ekatvam tatra na na yuktam ity ācakṣanena. 61 vastuni cit. ibid. p. 108.

48

तथा हि

दीर्घवृत्तोद्भवपुरुषधूमचानवनता3दिभिः ।

यथाभासा विभिन्ने वेशकाला4विभेदिनः ॥ ४ ॥

तथैव सदृटद्रव्यकार्चनोज्ज्वलतादयः ।

आभासभेदा भिन्नार्थकारणस्ते पदे ध्वनेः ॥ ५ ॥

एकस्मिन्नेव घटादावर्थे5स्वधित्सावशादिर्घता6त्र्यश्रतापारिमाण्डल्यादि7वाभाति, चित्रकयवधानच्छायामात्रार्थितायां पुरुष एव8उर्ध्वतामात्रं9प्रतिभाति, तत्कार्यार्थितायां10तु न्यक्षणेक्षमाणस्य पुरुषः । धूममात्रेव कस्यचिदाभासते,11तद्विदस्तु12तार्णतादिमणिरूप्यादिविशेष13इव ।14स च तथा भिद्यमानोऽपि देशकालभेदं न स्पृशति । ।15तथैव घट एव16सन्निधत्याभासो17न्या-

62 E puts tathā hi at the beginning of the vrtti. 63 ādibhiḥ : ādayaḥ T 64 yathā : yadā T; ābhāsā vibhidyante T (and ĪPV, Bh, ĪPVV): avabhāsā bhidyante E Gh Ñ C Ch J Ñ T. 65 avibhedinaḥ : vibhedinaḥ Gh Ñ C Ch T. Also the text commented on by Bhāskara has vibhedinaḥ and on these grounds the editors of Bh claim that this was the original reading. The correct reading, on the contrary, is certainly avibhedinah, as one can gather from the ĪPVV and from the ĪPV itself, if rightly understood, and as is evident from the thread of the argumentation. Lastly, the vrtti (tathā bhidhyamāno 'pi deśakālabhedam na sprśati) removes any doubt on this point.

66 'vadhitsā' : tathārthitā T, tathāvadhitsā J; cf. the quotation (from the tīkā) ibid. p. 110 avadhitsādi iti rucyādibhedatrayāśrayenetyaṛthah. 67 Cf. ibid. p. 109 'tryaśra' [from the tīkā] iti dvigur matvarthīyājantah. 68 vā : ca E Gh Ñ C Ch J Ñ T. 69 ūrdhvatā : ūrdhva E Gh Ñ C Ch J Ñ T. 70 pratibhāti : bhavati E Gh Ñ C Ch Ñ T, bhāti J. 71 tu E Gh Ñ C Ch Ñ T. 72 tadvidastu : taddhitavastu T. 73 tārṇatādi : tārṇatādi E, tārṇatādi Gh Ñ C Ch J Ñ T. 74 iva : bhāvah E Gh Ñ C Ch J Ñ T. 75 In the KSTS Ed. of the ĪPV (vol. II p. 90) fn. 112 is constituted by a passage of the Aṣṭasāhasrī (that is, Utp.'s tīkā), where this phrase from the vrtti is quoted: ata eva vrttau sa ca deśakālabheda- dam tathā bhidyamāno 'pi na sprśati; deśa na sprśati cit. in ĪPVV III p. 110. 76 The whole sentence (f.:m tathaiva) is quoted in a passage from the tīkā which constitutes fn. 114 of the KSTS Ed. of the ĪPV (vol. II p. 91). The text is the same as the one I have established, except for mātrakā (see below fn. 81, reading J) which should probably be expunged. 77 sann ity : santitry T.

II 3.4-5 हगेव पटाद्यन्तराशेषार्थसाधारण: सत्तामात्रसाध्यसत्परामर्शोर्धंकियाकारी, तत्रैव घट इति चाभासोऽन्तर्बहिष्करणज्ञेय: पृथुबुधनोदराकारार्थसामान्ये सन्य एव य: पटादिषु नास्ति, काञ्चन इति चापरो यो मृण्मयादिषु नास्ति स च तथार्थता दिवशादिन्द्रयक एव जायते । एकैकस्यासावाभास एकेन शब्देन व्यवहारार्थ तत्स्व नियतार्थक्रियार्थिभिर् अभिधीयते, घट इति न सत्स्वभिधीयते न च काञ्चन इति घटाभास: । चैत्र इति बाल्यादिसाधारणो देशादिरहितस्तथैव च । ततोऽर्थंक्रिया भिन्ना ॥ ४-५ ॥

आभासभेदादस्तूनां नियतार्थंक्रिया पुन: । सामान्याधिकरण्येन प्रतिबिम्बादभेदिषाम् ॥ ६ ॥

एकस्मिन्नेव स्वलक्षणे प्रत्याभासं नियतिशक्त्या कार्य नियमितं तथाभूतानेककार्यकृता भाषभेदाधिकरणमेकं च ।

78 pata° : ghata° E Gh Ñ C Ch Ñ T, pata° corrected to ghata° J. 79 arthasādhārana° : arthasārthasādhāraṇa° T, arthamātrasādhāraṇah J, padārthasādhāraṇa° Kh. 80 satta° : sattva° T. 81 sat° : tat° T. Cf. ibid. p. 111 nahi sā itī abhāsāntarvartī sālpa-rāmarśah, sa ca katham arthakriyeti utkah / arthakriyā hi kāryam na tu svabhāvah. 82 Cf. the (abridged) quotation in ibid. p. 111 nanu ca vrttau ... tatraiwa ghate ghata ity ābhāso yaḥ patādiṣu nāsti, kāñcana ity ābhāsāntar T. 84 jñeyah : jñeya° T. 85 sāmanyo cit. ibid. p. 111 anya-tā T. 86 'nya eva : anya-° E Gh C Ñ, yaḥ ghata° Ñ Ch T, yaḥ pata° corrected to yaḥ ghata° J. 88 E repeats the passage budhnodara° ... ghatādiṣu. 89 cāparo yo omitted in E Gh Ñ C Ch J Ñ T. 90 sa ca omitted in E Gh Ñ C Ch J Ñ T. 91 ādi° omitted in J. 92 ekaikena : ekaika° E Gh Ñ C Ch J Ñ T. 93 abhidhīyate ghata : aghata E Gh Ñ C Ch J Ñ T. 95 na ca kāñcana : na kāñcana E Gh Ñ C Ch J Ñ T, na kāñcanaghata C. 96 deśādirahitas : deśādi-virahitas G, deśādibhih abhihitas E, deśādir abhihitas C Ñ (in C then corrected to deśādi-rahitas), deśādirbhihitas T (then corrected to deśādirhitas T. 97 ekasmin cit. ibid. p. 114. 98 pratibhāsam : pratyabhāsan T; cf. ibid. p. 114 vrttitāldparyām vyacaśṭe “pratyabhāsam iti.” 99 kāryam : kāryān T 100 ābhāsa° : avabhāsa° J.

सामानाधिकरण्याभासवशाद् वस्तु । 101अनेकस्पैकता हि सामानाधिकरण्यम् ॥ ६ ॥

पृथक्दीपपकाशानां स्रोतसां सागरे यथा । अविरुद्धावभासानामेककार्या तथैकयधि: ॥ ७ ॥

भिन्ना: प्रदीपप्रभा 102अविभागेनावभासन्ते नदीप्रवाहश् 103च सिन्धुबुद्धौ पानके च ते ते रसा:, तथा शौक्ल्यमहरत्व- 104पटत्वाद्य् 104वभासा: परस्परानुपवेशक्षमा:, न तु नील-पीताद्यवभासास्तत्तदेककार्यैकद्रव्याभासभावेन कल्प्यन्ते प्रत्यक्ष 106एव । 107तदेतत् सामानाधिकरण्यम् ॥ ७ ॥

तत्राविशिष्टे वद्यादौ कार्यकारणतोषणत- । तत् छेदोऽर्थाद् भातमा प्रमाणादेकतो मतः ॥ ८ ॥

108कदाचिद् 109देशादिसहभावावच्छिन्नस्वलक्षण 110रूपविशेष-त्यागेन 111एकसामान्यरूपाग्न्याभास 112मात्र एव 113त्रैलोक्यत्रैका-ल्यगामित्वेन निजप्रमाणादेकस्मादेव विशिष्टकार्य 114कारणतो-षणोदर्वी 114भागाग्नशब्दवाच्यता 116दिस्वभावसिद्धि: ॥ ८ ॥

101anekasyikatā cit. ibid. p. 114 102avibhāgenāvabhāsante : avibhāgā ābhāsante T; avibhägena cit. ibid. p. 117. 103ca only found in T, but cf. ibid. p. 117, where the two ca in the sentence are referred to. 104patatvādīo : patatvādio J 105avabhāsāḥ : oābhasāḥ J T. 106eva omitted in T N J. 107tad etat : tat E Gh Ch Jh Ñ T, etat Ñ J. 108kadācit cit. (perhaps from the tikā) ibid. p. 121. 109Cf. the paraphrase ibid. p. 121 vrttau deśakālābhāsāv eva tyajyatvena kasmād uktau. 110rūpao : svarūpao T. 111Cf. ibid. p. 124 avasthātṛrupam hi tatra ekam sāmānyam ity upapāditam / ekamekasāmānyasvarūpam upapādya .... 112mātra omitted in E (but found in Kh). 113Cf. ibid. p. 124 yenaiva prathamaprathamapravṛttena pramānena bhāsvarasanniveśarūpam auṣnyādi avyabhicāritayā gṛhitam; tenaiva trailokye traikālye ca bhāsvarapayogah. 114kāranatosnao : kāranordhvao T. 115ūrdhvabhāgāgni° : ūrdhvabhāgāgni° Ñ J. Cf. also vrtti ad II. 2. 20 ūrdhvabhāktvād agni°. 116ādi° cit. ibid. p. 125; svabhāvāo, omitted in E Gh Ñ C Ch Jh Ñ T, cit. ibid. p. 125 svabhāvaśabdena iti vrttigateña.

II 3.9

सा तु देशे'९९दिकाद्यक्षान्तरभिन्ने स्वलक्षणे ।

तात्कालिकी प्रवृत्तिः स्यादर्थिनोऽप्यनुमानतः ॥ ९ ॥

'१"कायप्रवृत्तिः पुनर्देशकालादिप्रत्यक्ष"६भेदसाहित्येन '२०स्वलक्षण एव तद'२'अर्थितया प्रमाणसमूहादेव । अनुमानादपि'२२धर्मिप्रत्यक्षविशिष्टादेव प्रवृत्तिः ॥ ९ ॥

दूरान्तिकतयार्थानां परोक्षाद्यक्षतात्मना ।

नाभान्तरतया दोषैर्ग्रहणकस्यान्यथापि वा ॥ १० ॥

भिन्नावभासच्छायानामपि मुख्यावभासतः ।

एकप्रत्यवमरर्शाख्यादेकत्वमनिवारितम् ॥ ११ ॥

'२३दूरान्तिकादिस्फुटास्फुटत्वादिना बहिरन्तश्'२४चरत्वेन वाभासस्य छायामात्रभेदेऽपि '२५तन्मुख्यस्वभावरूपप्रत्यवमरशै'२६क्यादर्थानाम् '२७एवेयम् '२८अबाधितम् ॥ १०-११ ॥

अर्थक्रियापि सहजा नार्थानामीश्वरेच्छया ।

नियता सा हि तेनास्य नाक्रियातोऽन्यता भवेत् ॥ १२ ॥

'२९उल्लेखघटादीनां बाह्यार्थक्रियाविरहे'डपि घटादितैव, '३०अस्वाभाविकत्वात् '३'तस्या ईश्वरेण प्रत्याभासं नियमिताया: ॥ १२ ॥

'१७७°आदिका° : °आधिका° E. '११८काया° : kārya° E Gh N C Ñ, kāya° corrected to kārya° J. '१९९भेदसाहित्येन : °भेद एव sāhityena T. '१२०स्वलक्षण एव tad° : svalakṣane eva tad° E Gh N C Ch Jh Ñ T. '१२'अर्थितया° : °arthatayā° N. '१२२T has hi. '१२३दूरान्तिकादिसु sphutatvāsphutatvādinā Gh Ñ Ch T, दूरान्तिकात्vādisu sphutāśphutatvādinā Jh. '१२४bahirantaścaratvena vābhāsa-sya Ñ, ca gatvarābhāsasya J. ābhāsasya cit. ibid. p. 150 (perhaps through the tīkā). '१२५tanmukhya° cit. ibid. p. 150. '१२६°aikyād : °ākhyād Ñ. '१२७aikyam omitted in Ñ. '१२८abādhitam : apabādhitam E C Jh Ñ T, upabādhitam Ch, api bādhitam corrected to apabādhitam J. '१२९Cf. the paraphrase ibid. p. 151 ullekha eva ghaṭādayas teṣām, bāhya-sya svalakṣanasya yā arthakriyā tayā virahe 'piṭi vrttyahah. '१३०asvābhāvikatvād : svābhāvikatvād E Gh Ñ C Ch Jh Ñ T. '१३'tasyā : asyā E (tasyā K).

रजतैः 132 कविमर्शेऽपि शुक्तौ न रजतिस्थितिः ।

उपाधिदेशसंवादाद् द्विचन्द्रेऽपि नभोद्यथा ॥ १३ ॥

133 रजते शुक्तौ च रजातं 134 वमर्शक्येन रजतत्वेऽपि पुनः

शुक्तिदेशसंगति 135 बाधेनोपाधिसंवादाभावात् तद्बुद्धिरस्थैर्याद् भान्ता ।

द्विचन्द्रेऽपि 136 नभोदेशसंवादानिमित्था ॥ १३ ॥

गुणैः शब्दादिभिर्भेदो जात्यादिभिरभिन्नता ।

भावानामित्थमेकत्र प्रमातर्युपपद्यते ॥ १४ ॥

भावानां शब्दरूप 137 संस्थानादिना भेदाक्षेपो जात्यादिना च

138 एकताक्षेपोऽनुभूयमानः प्रमात्रैक्ये घटते ।

अन्योन्य 139 भेदद्‌य-

वस्थाप्यनुसंधानायत्ता ॥ १४ ॥

विश्ववैचित्र्यचित्‌त्रस्य समभित्तितलोपमे ।

विरुद्धाभावसंस्पर्शे परमार्थसतीश्वरे ॥ १५ ॥

प्रमातरि पुराणे तु सर्वदाभातविग्रहे ।

किं प्रमाणं नवाभासः सर्वप्रमितिभागिनि ॥ १६ ॥

140 अपूर्वाभासस्प्यासिद्धवस्तुसाधनान् 141 नित्याभासिनः प्रमातुर्न

132 eka° : aikya° E Gh N C Ch J Jh Ñ T. A fluctuation between these two readings is also found in the editions of Abh.'s commentaries (°eka° in the two editions of the ĪPV, confirmed by the BH; °aikya° in the ĪPVV). The vrtti seems to presuppose °eka° and so also does the ĪPVV, though it begins precisely with rajatayā yadyapi śuktīrajatasya aikyaṃ vimrśyate. But these are simply two ways of formulating a concept that is basically the same. 133 Cf. ibid. p. 152 "rajate" iti satyarajate, "śuktau ca" iti rajatvena adhyavasitavyam, "dvicandre 'pi buddhir mithyā" iti vrttiyojanā. 134 °avamarśa° : °vimarśa° E Gh N C Ch J Jh Ñ T, bādhana nopādhi° corrected to bādhanenopādhi° J. 135 °bādhenopādhi° : °bādhanenopādhi° E Gh N C Ch J Jh Ñ T, bādhana nopādhi° corrected to bādhanenopādhi° J. 136 nabhodeśa° cit. ibid. p. 153. 137 °samsthāna° : °samsthā° T C Ch J Jh Ñ T; cf. ibid. p. 158 samsthānāsya gunair eva vidhīyate. 138 ekatākṣepo : ekatādikṣepo J. Cf. ibid. p. 159 nanu ākṣepo'nubhūyate. 139 Cf. ibid. p. 160 vrttau bhedavyavasthāiva upasamhrtā; api cit. ibid. p. 160. 140 apūrvābhāsasya : apūrvārthābhāvasya T. 141 nityābhāsinah : nityābhāsinah J; nityāvabhāsinah cit. ibid. p. 163.

II 3.15-16 प्रमाणोपयोगः । ततत्न्नवा¹⁴²वभासानां ¹⁴³तत्तत्प्रमाणत्वात् कामं स प्रमाणमन्यस्य स्यान्न तु स्वात्मनः सदा सिद्धस्य, केवलं ¹⁴⁴विभुः प्रमितौ स्वातन्त्र्यात्र् प्रमातैव । ते ते ¹⁴⁵विचित्रा विश्वाभासास् तस्मिन्स्थायिनि वैचित्र्येणोपपद्यन्ते । ¹⁴⁶तदिच्छातो ¹⁴⁷sपरावभास- मानार्थानां तत्सार¹⁴⁸स्वरूपाभ्रंशे ¹⁴⁹मायाशक्तितवशादुन्मज्जन- निमज्जने, तत्स्वतस् ¹⁵⁰तस्यैवापराधीनां सत्ता सदातनी¹⁵¹ विरुद्धप्राक्प्रध्वंसाभावादियोगात्, सोऽपि यावद् ¹⁵²उपदिश्यते भावाभावाभ्यां तावत् स एव प्रमातृतां ¹⁵³भजत ¹⁵⁴उपदि- दिक्षोरभावाद् ¹⁵⁵upadeśānupapetteḥ ॥ १५-१६ ॥

अपवर्तिंतपूर्वोऽत्र केवलं मूढतावशात् । शक्तिप्रकाशेनेशादिव्यवहारः प्रवर्त्यते ॥ १७ ॥

मायान्यामोहवशात् केवलमस्मिन्नात्मनि प्रमातृरूपे शिवेश्वरवादि- व्यवहारो यो न प्रवर्तितः स शुद्धस्वातन्त्र्यादिहेतुप्रदर्शनेन ¹⁵⁶भावनाद्युपदेशाय साध्यते ॥ १७ ॥

इति क्रियाधिकारे तृतीयमाह्निकम् ॥

¹⁴² °avabhāsānām : °ābhāsānām T, °ābhāsānām corrected to °avabhāsānām Ñ J. ¹⁴³ tattatpramānatvāt : tatpramānatvāt E Gh Ñ Ch Jh T, tanmayatvāt T, tanmayatvāt corrected to tattatpramānatvāt J. ¹⁴⁴ vibhuh : prabhuh Ñ. ¹⁴⁵ vicitrā viśvābhāsās tasmin : vicitrābhasā viśvā (then two illegible aksaras) smin T. ¹⁴⁶ tadicchāto cit. ibid. p. 163. ¹⁴⁷ °avabhāsamāna° : °ābhāsamāna° T. ¹⁴⁸ svarūpābhramśe : °svarūpabhramśe ca E Gh Ñ C Ch J Jh Ñ T. ¹⁴⁹ māyāśaktivasaśāt cit. ibid. p. 163. ¹⁵⁰ tasyaiiva : tasyā vā E Gh Ch Ñ J Jh Ñ T add purātanī, which is probably a gloss; sadā purātanī C. ¹⁵¹ E Gh Ch Ñ J Jh Ñ T add purātanī, which is probably a gloss; sadā purātanī C. ¹⁵² upadiśyate : upadiśyeta E, uddi- śyate G Kh. ¹⁵³ bhajata omitted in T. ¹⁵⁴ upadidikṣor : upadidr̥ṣor J. ¹⁵⁵ upadeśānu- papatteh, conjectural emendation for upadeśānupapattiḥ E Gh Ñ C Ch J Jh Ñ T and deśānupapatteḥ T. ¹⁵⁶ The phrase bhāvanādyupadeśāya found in all MSS. is apparently corrupt (unless one takes it as a rather unlikely, samāhāradvandva formed by bhāvanādi and upadeśa); sādhayate : bhāvayate T.

चतुर्थमाहिकम्

एष चानन्तशक्तितत्त्वादेवमाभासयत्यमून् ।

भावानिच्छावशादेषा क्रिया निर्मान्तृतास्य सा ॥ ९ ॥

प्रमाता चिदूпоsन्न्तशक्तिरेश्वरः स्वेच्छावशात् तथातनभावान् आाभासयेत् । सैव चेच्छाशक्तर्निर्मातृताख्या क्रिया तस्य ॥ ९ ॥

जहस्य तु न सा शक्तिः सत्ता यदसतः सतः ।

कर्तृकर्मत्वतत्त्वैव कार्यकारणता ततः ॥ २ ॥

जडं प्रधानपरमाणबीजादि तु न शक्तमस्तो निर्माणे,

कर्तृत्वमेव हि कारणत्वं कर्मतैव कार्यत्वं न त्वनयत् ॥ २ ॥

यदसतदघुकता नासतः सत्स्वरूपता ।

सतोडपि न पुनः सत्तालाभेनार्थोंsथ चोच्यते ॥ ३ ॥

कार्यकारणता लोके सान्तर्विपरिवर्तिनः ।

उभयोन्द्रियवेद्यत्वं तस्य कस्यापि शक्तितः ॥ ४ ॥

असतः सत्स्वभावता विरुद्धा सतश्र सिद्धा ।

सिद्धस्यैवान्तर्बाह्यान्तःकरणद्रयीवेद्यतापादनम् ईश्वरेणोत्पादनम्

॥ ३-४ ॥

157tathā tān bhāvān : tathā tadbhāvān T, tān ābhāsān E Gh N C Ch Jh Ñ T.

158ābhāsayet : bhāsayet J, avabhāsayet T.

159ca omitted in E Gh N C Ch Jh Ñ T.

160tatah : yatah N.

161jadam : jada° T.

162ca : hi T.

163siddhā cit. ibid. p. 188.

164īśvarena : īśvara° J.

II 4.5

एवमेक क्रिया सैषा सकृमान्तर्बहिःस्थितितः ।

एकस्यैवोभयाकारसहिष्णोरुपपादिता ॥ ५ ॥

सैषा क्रियन्तर्बहिःस्थितिरूपतया सक्रमाप्येकस्य स्वसंवित्स-

दृश्य कर्तृभावसच्छाया¹⁶⁵भेदेऽप्यैक्येन प्रत्यवमृश्यस्य कर्मणश्

¹⁶⁶चकस्य सम्बन्धेन तदभिन्नाश्रितत्वादेकत्वं साधितम् ॥ ५ ॥

नहिस्तरप्यैव तत्कार्यं यदनन्तर्यदपेक्षया ।

प्रमातृपेक्षया चोकता द्वयी बाध्यान्तरस्थितिः ॥ ६ ॥

अर्थस्य बाह्यतापादनं ¹⁶⁷कार्यत्वम्, ततो बाह्यता कार्यता

चैका¹⁶⁸पेक्षया प्रमातारम् ¹⁶⁹अपेक्ष्य चान्तर्बहिर्वहारः ।

¹⁷⁰ततस्तस्यैव कार्यम् ॥ ६ ॥

मातैव कारणं तेन स चाभासद्वयस्थितौ ।

कार्यस्य स्थित एवैकस्तदेकस्य क्रियो¹⁷¹चिता ॥ ७ ॥

एवं प्रमातैव कारणं ¹⁷²स च बाह्या¹⁷³न्तरकार्याभास¹⁷⁴क्रमे

स्प्येक एवेत्येमप्येकस्य कर्तुः सिद्धा क्रिया ॥ ७ ॥

¹⁶⁵ oभेदे 'py aikyaena : oभेदेनैकत्वेना E Gh Ñ C Ch Jh Ñ T, oभेदे 'py eka-

rvena J; cf. ibid. pp. 189-90 ābhāsabhedayoge 'pi ity arthah.

¹⁶⁷ kāryatvam omitted in E Gh Ñ C Ch J Jh Ñ T.

¹⁶⁹ apeksyā : apeksā Ch.

¹⁷⁰ T adds tatas tasyaiva kārya, which is very probably a gloss.

अत एवाडकुरेऽपीषो निमित्तं परमेश्वरः । तदन्यस्यापि बीजादेर्हेतुता नोपपद्यते ॥ ८ ॥

¹⁷⁵bāhyābhāsatāpadanmrtpadanm iti chidrupasyāiv kāraṇatā tato 'dukṛtādṛ kau nimittakāraṇatvenaiśvaraḥ kaśchidṛṣṭaḥ, na cāpi bījadeṛdasya kāraṇatā niranusandhānasyā yuktā ॥ ८ ॥

तथा हि कुम्भकारोऽसावैश्वर्यैव न्यवस्थया । तततन्मृदादिसंस्कारकमेण जनयेद्घटम् ॥ ९ ॥

कुम्भकाररूपे प्रमातरि कारणे स्थितेऽपि मृदादिसंस्का- ¹⁷⁶ra¹⁷⁷pekṣā ghaṭasyaiśvarakṛtaniyatisṅjamar yādaya na svabhāvena ॥ ९ ॥

योगिनामपि मृद्वीजेऽ विनैवेच्छावशेन तत् । घटादि जायते तत्तत् ¹⁷⁸sthirasvārthakriyākaram ॥ ९० ॥

मृदाद्यनपेक्ष्य योगिनामिच्छामात्रेण घटादेः स्थिरस्य घटाद्यर्थ- क्रियाकारित्वं ¹⁷⁸ca nirṛttiḥ ॥ ९० ॥

योगिनिर्माणताभावे प्रमाणान्तरनिश्चिते । कार्यं हेतुः स्वभावो वा तदुत्पत्तिमूलजः ॥ ९९ ॥

अत एव कार्यं स्वभावो वा तदुत्पत्तिगर्भो योगिनिर्मित- त्वाभावनिश्चयाभावे हेत्वाभासः । ¹⁷⁹tanniśchaye punarīśvar- नियत्य¹⁸⁰pekṣayā hetutā स्पात् ॥ ९९ ॥

¹⁷⁵ bāhyābhāsatā° : bāhyatā° (bāhyābhāsatā° Kh G). ¹⁷⁶ °apekṣā : apeksayā J T. ¹⁷⁷ sthirasva° : sthirasya T. ¹⁷⁸ ca nirṛtti : cittavṛttih corrected to ca nirṛttih J. ¹⁷⁹ tanníscaye : tanníscayena J. ¹⁸⁰ °apekṣayā : °apekṣā yā T.

II 4.12-13

भूयस्ततत्प्रमात्रेकवह्न्याभासादितो भवेत् ।

परोक्षादप्यधिपतेर्धूमाभासादि नूतनम् ॥ १२ ॥

कार्यमन्यभिचार्यस्य लिख्यन्यप्रमातृगात् ।

तदाभासस्तदाभासादेव त्वधिपते: पर: ॥ १३ ॥

"'अवह्न्याभासस्फुरणपूर्वकोऽpi जायमानो धूमाभासो दूरादौ

'²नियतिशक्त्या कृतसामर्थ्यादन्याभासादेव '³तद्वेशगतततत्'⁴प्र-

मात्रन्तरसाधारणात् तस्यैव सोऽन्याभिचारेण गमकः, जातरूढस्तु

धूमाभासस्तद्वह्न्य'⁴भासस्य '⁶अत्यन्तपरिक्षयात् '⁹तत्तत्प्रमात्र-

न्तरगाद् धूमाभासादेव परोक्षादपि '⁸पूर्ववत् कृताधिपत्यात्

॥ १२–१३ ॥

अस्मिन्सतीदमस्तीति कार्यकारणतापि या ।

साप्यपेक्षाविहीनानां जहानां नोपपद्यते ॥ १४ ॥

अस्मिन्सतीदं '⁹भवतीति नियतं पूर्वापर्य कृतिकारोहि-

न्युदय्योरकार्यकारणयोरप्यस्तीति पूर्वस्य '⁰सामर्थ्ये परस्य सत्तेति

स्यात् कार्यकारणभावः, '¹तच्चापेक्षा'²रहितानां जड़ानां न

181 avahni° : vahni° E Gh Ñ C Ch J Ñ T. Cf. ibid. p. 215 dhūmābhāso 'nyābhāsād eva jāyamānas tasyaiwa gamakah. jātarūḍhas tu dhūmābhāsād eva jāyamānas tasyaiwa

gamakah iti pūrvena sambandha iti vrttiyojanā. 182 niyati° : niyata° C. 183 tad° : tat-tad° J. 184 °pramātr° : °pramānā° Ñ. 185 °avabhāsasya : °ābhāsasya corrected to °avabhāsasya T. 186 °atyanta° : antara° E Gh Ñ C Ch J Ñ T. 187 tattatpramātr° : tatah

pramātr° Ñ. 188 pūrvavat : ca tatpūrvakrta° E Gh Ñ C Ch J Ñ T. 189 bhavati cit. ibid. p. 229 bhavatiśabdena ... vrttigatena. E Gh C Ñ, sāmarthye corrected to sāmarthyam T, sāmarthya Ñ. Cf. ibid. p. 218, where

the two meanings implicit in the Buddhist formula asmin sati idam asti are clearly distinguished: atraiva ca sūtre pūrvasāmarthyaparasyā satteti pūrvasāmarthyam parasya

satteti pakṣāv api sūcitāv eva. 191 tat cit. ibid. p. 227 nanu vrttau sa iti vaktavye katham "tat" iti nirdeśaḥ / ... nātra kāryakāranabhāvah pāramṛśyate, api tu tallakṣaṇarvena

sammato viśiṣṭaḥ sattabhūto vākhyārtha iti bhāvah. 192 °rahitānām : °hīnānām Ñ.

58

युक्तम् । एतावदेतस्यात् पूर्वस्य 193सामर्थ्यं परस्या सत्ता न चैवं II 4.14 किंचिदुक्तं स्याद् न च 194पूर्वस्य सामर्थ्यलक्षण: स्वभाव: परसत्तारूप: ॥ ९४ ॥

न हि स्वात्मैकनिष्ठानामनुसंधानवर्जनानाम् । सदसत्तापदेsप्येष सप्तम्यर्थ: प्रकलप्यते ॥ ९५ ॥

सदसद्वा कार्य कारणमप्यात्मपर्यवसितं यदमनुसंधानशून्यं नान्यापेक्षस्वभावं तत्स्च नात्र 195प्रधानापेक्षामयो गुणविभक्त्यर्थो घटते ॥ ९५ ॥

अत एव विभक्त्यर्थ: प्रमात्रेकसमाश्रय: । क्रियाकारकभावाख्यो युक्तो भावसमन्वय: ॥ ९६ ॥

एकप्रामातृसंलग्नस्तु क्रियाकारकभावाख्यो विभक्त्यर्थों भूमिभिर्जोदकादीनां समन्वयो युक्तो न तु शुष्कोडन्य: कार्यकारण96भाव: ॥ ९६ ॥

परस्परसवभावत्वे कार्यकारणयोरपि । एकत्वमेव भेदे हि नैवान्योन्य97स्वभावता ॥ ९७ ॥ कार्यकारणयोरन्योन्यरूपत्वेsप्येकतैव तद्भाव: ॥ ९७ ॥

193 sāmarthyam : sāmarthye T; cf. ibid. p. 226 etad vastu vṛttau vivarisyate / katham pūrvasya sāmarthyam parasya sattā. 194 Cf. ibid. p. 223 tatra hi pūrvasya sāmarthyam nāma svabhāvah sa parasya sattārūpo neti vaksyate. 195 pradhāna and gunavibhakti cit. ibid. p. 231. 196 obhavah : obhāvākhyah E. 197 svabhāvatā : osvarūpatā Kh G (this is also the reading of the IPVV and the two editions of the IPV).

II 4.18

एकात्मनो '९८विभेदश्य क्रिया कल'९९क्रमानुगा । 200 तथा स्पात् कर्तृतैवं तथापरिणमत्तया ॥ १८ ॥

201एकस्वभावस्य भेदेन स्थितिः परिणामः कालकलितः: क्रियैव, ततः परिणामे स्वतन्त्रस्य शक्तिमतः कर्तृतैव हेतुता ॥ १८ ॥

न च युक्तं जहस्प्यैवं भेदाभेदविरोधतः । आभासभेदादेकत्र चिदात्मनि तु युज्यते ॥ १९ ॥

202जडस्पाभिन्नात्मनो भेदेन 203अवस्थितेर्विरोधादयुक्तम्, स्वच्छे चिदात्मन्येकसिम्नेवमनेकप्रतिबिम्ब204धारणेनाविरोधाद्युज्यते ॥ १९ ॥

वास्तवेऽपि चिदेकत्वे न स्यादाभासभिन्नयोः । चिकीर्षालक्षणेकत्वपरामर्श विना क्रिया ॥ २० ॥

एकसिम्न्श्चितत्त्वेऽप्य् 204अकस्माद् 206आभासभेदो न घटते,

198 vibhedaś ca : vibhedena N Ch J T, vibhede 'pi corrected to vibhedena Gh. 199 °kramānugā : vibhāgini Gh N Ch T. 200 The readings of the third pāda fluctuate: hetah syāt kartrtaivai'vam E Ch T, hetuh syāt kartrtaivitam Gh, hetuh syāt kartrtaivevam N, hetuh (corrected to tathā) syāt kartrtaivaivam C, hetuh syāt kartrtevaivam J N, tato 'nyat kartrtaiveyam T, tathā syāt kartrtaivaivam IPV, tathā syāt kartrtā vaivam IPVV. Except for hetah, °tā vaivam and 'nyat, all the other readings might be admissible. As it is impossible to ascertain the original reading, I have decided to accept the reading of IPV, which was definitely the one Abh. had in his text (this is also confirmed by Bhāskara). 201 ekasvabhāvasya : ekasvabhāve sya J. 202 Cf. ibid. pp. 238-9 nanu evam pratibimba-grahane jadasya virodhāt cidātmana eva ekasya bhedena sthitir ity etat yujyate, jadasya bhadastitir hi viruddheti vrttau kim iti pratibimbadhāranam darśitam. 203 avasthiter : avasthite N. 204 °dhāranena : °dhārana° T. 205 akasmād, my correction: ekasmād E T, ekah sadā G Gh N C Ch J Jh N T. Cf. ibid. p. 248 vrttau "akasmāt" iti nimittasya alābhād iti bhāvah; cf. also ibid. p. 249 nirnimittatvāt in the paraphrase of the țikā. 206 ābhāsabhedo cit. ibid. p. 248, where the whole sentence is commented on: ābhāsabheda iry ādinā vedyagatam rūpam uktam / tad viṣayatvena yat kriyātvam nāma kartrtādhāyakam, tat na ghatate iti brahmagatam rūpam uktam.

न च तत्र क्रियात्वम्‌ । यदा तु स चिदात्मा तथाचिकीर्ष्या

परामृशन् बहिराभासयति तदा तदुपपद्यते । जडस्याप्यसित

भवतीत्यस्यामपि सततक्रियां बुभूषायोगेन स्वातन्त्र्याभावाद्

अकर्तृत्वम्‌, तेन प्रमाैतैव तं भावयति तेन तेन वा

निमित्तात्‌निमित्ते सा भवतीत्यत्र परमार्थः ॥ २० ॥

इत्थं तथाघटपटाद्याभासजगदात्मना ।

तिष्ठासोरेवमिच्छैव हेतुता कर्तृता क्रिया ॥ २१ ॥

चिद्‌घनः स्वतन्त्रस्य विश्वात्मना संस्थातुमिच्छैव जगत्प्रति

कारणता कर्तृतारूपा सैव क्रियाशक्तिः । एवं चिदूपस्यैवस्य

कर्तुरेव चिकीर्षण्या क्रिया मुख्य्या, नाकर्तृकं कर्मीसित

कर्मादीनां कर्तृमुखेनैव पच्यारतेः ॥ २२ ॥

इति क्रियाधिकारे चतुर्थमाहिकम्‌ ॥

इति क्रियाधिकारः ॥

207 cikīrṣā parāmrśan cit. ibid. p. 250, where the rest of the sentence is also commented on: vrttyartham vyākaroti "atra" hi vrttigranthe ābhāsanā ca kriyā tasyā ādya eva parāmarśo lakṣaṇam / sa eva bahīkaraṇātmakaābhāsanāparyantā / tena trayam etad upakrame bhāsamānām api dvayam eva, vastutas tu ekam apīti tātparyam. 208 tad cit. ibid. p. 248 "tat" ity ābhāsabhedasya ghatanam kriyātvam. 209 jadasyāpi : jadasyāsyāpi T (also possible); jadasya cit. ibid. pp. 248 and 250. 210 Cf. ibid. p. 252 tadābhāsaś ca iti svātantryābhāvasya nimittam ukta icchābhāvā iti saṅgatih / iyatā vrttir ubhayathā gamitā bubhūṣāyogena yat svātantryābhāvas tasyā svātantryābhāvas tata iti. 211 akartrtvam : akartṛkatvam E; akartṛtvam cit. ibid. p. 248. The reading of T is not clear (pramave?). 213 tam cit. ibid. p. 248 tam iti jadam. 214 tena tena : tena T Gh N J. 215 ādinā : "ādi" E Gh N C Ch J Jh N T. 216 sa cit. ibid. p. 248 sa iti pra-mātā. 217 tathā : tadā Ch. 218 sthātum : kartum E Gh N C Ch Jh N T; sthātum is also confirmed by the quotation of this passage (cidvapusah ... rūpā) in MMP p. 39. 219 kriyā mukhyā : mukhyā kriyā N. 220 nākarṛkam : nākartrtvam N Ch Jh, nākarṛtvam corrected to nākartrkam J. 221 upacāratah : "upacārāt E C N, upacāratah corrected to upa-cārāt J.

III 1.1

आगमाधिकारः

प्रथममाहिकम्

एवमन्तर्बहिर्वृत्तिः किया कालकमानुगा ।

मातुरेव तदन्योन्यावियुक्ते ज्ञानकर्मणी ॥ १ ॥

'प्रमातुरन्तःस्थितस्य 'आत्मनो बहिष्कार एव क्रमानुगता

क्रियेति 'परस्पराविरहिते 'तस्य ज्ञानक्रिये ॥ १ ॥

किंतु आनतरदशोद्रेकात् साधार्यं तत्त्वमादितः ।

नहिर्भावपरतवे तु परतः परमेश्वरम् ॥ २ ॥

'ईशितुरन्तर्बहिःस्थितावन्तर्भावप्रधान्ये पुनः साधार्यं तत्त्वम्,

अपरं बहिर्भावोद्रेकादेश्वरम् ॥ २ ॥

ईश्वरो बहिरुन्मेषो निमेषोऽन्तः सदाशिवः ।

सामानाधिकरण्यं च सदृद्रियाहमिदंधियोः ॥ ३ ॥

उन्मेषनिमेषौ बहिरन्तःस्थिती 'एकेश्वर सदाशिवौ, बाह्यान्तरयोः

'वेद्यवेदकयोरेकचिन्मात्रविश्रान्तेरभेदात् सामानाधिकरण्येनैदं विश्व-

महमिति विश्वात्मनो 'मतिः 'शुद्धविद्या ॥ ३ ॥

1 pramātuh and kriyā cit. in ĪPVV III p. 260 vrttau pramātuh kriyā iti sam-bandhasaṣṭhī, na tu bahiṣkaranāpekṣayā kartari. 2 ātmano, omitted in T, cit. ibid. p. 258. 3 °anugatā : °anugatah T. 4 Cf. the quotation from the ṭikā ibid. p. 259 tad etad darśaya-ti “parasparāviraha” ity ādinā. 5 tasya omitted in C N. 6 īśitur cit. ibid. p. 265. 7 eva omitted in E Gh. 8 °āntarāyor : °abhyantarāyor E Gh J. 9 vedyavedakayor omitted in T. 10 matin cit. ibid. p. 273. 11 śuddha° cit. ibid. p. 273.

इदम्भावोपपन्नानां वेद्यभूमिमुपेयुषाम्‌ ।

भावानां नोधसारत्वादथावस्तुव लोकनात्‌ ।। ४ ।।

भिन्नवेद्यभूमाविदन્ત्या हृश्यतामापादितानाम‌¹²अपि भावानां

चिन्मात्रसारत्वादहमिति तत्त्व¹³प्रतिपत्तिः शुद्धताज्ञान्तिः ।। ४ ।।

अन्नापरत्वं भावानमनात्मत्वेन भाषणात्‌ ।

परताहन्तयान्च्छादात्‌ परापरदशा हि सा ।। ५ ।।

अत्रेदन्तामतेर‌¹⁴अपरत्वम हन्तया सर्वस्य विद्यस्याच्छादनात्‌

¹⁵परतेति परापरावस्थेषा ।। ५ ।।

भेदधीरेव भावेषु कर्तृबो धातमनोऽपि या ।

मायाशक्त्येव सा विद्येत्यन्थे विद्येश्वर यथा ।। ६ ।।

बोधकर्तृतामयस्यापि भेदेन विश्वेक्षणं विद्येति केचित्‌ ।

मायाशकितरप्येषा विद्या ¹⁶एव । संसारोत्तीर्णित्वात्‌ तत्रस्था

मन्त्रेश्वरविद्येश्वरा: ।। ६ ।।

तस्यैश्वर्यस्वभावस्य पशुभावे प्रकाशिका ।

विद्याशकितरिरोधानकरी मायाभिधा पुनः ।। ७ ।।

संहारिणा ¹⁷एकैवशस्य ¹⁸रसानां ¹⁹प्रत्यभिज्ञानं विद्यया,

पशु²⁰भावो ²²मायया ।। ८ ।।

12 api only found in T (cf. ibid. p. 274 vedyadaśām angīkrtavatām api; anyhow, not strictly necessary). 13 opratipattih : opratipattio T (but the omission of visarga before a sibilant is frequent in Malayalam script). 14 aparatvam cit. ibid. p. 275. 15 paratā cit. ibid. p. 275. 16 eva ommitted in Ch Jh. Gh Ñ C Ch J Ñ T put a danda before tatrāsthā. 17 aiśvaryasya : aiśvaryena T. 18 svātmanah : ātmanah E Gh Ñ C Ch J Jh Ñ T. 19 pratyabhijñānam : pratyabhijñāne E Gh Ñ C Ch J Jh Ñ T. 20 vidyayā : vidyā E C Ñ, vidyayā G Th T. 21 obhāvo : obhāve E Gh Ñ C Ch J Jh Ñ T. 22 māyayā : māyā E C Ñ, māyayā Gh Jh T.

III 1.8

भेदे त्वेकरसे भातेsहन्तयानातमनीक्शिते ।

शून्ये बुद्धौ शरीरे वा मायाशक्तिर्विजृम्भते ॥ ८ ॥

यदा भावो भिद्यनेदान्तया ²³एव भासन्तेsह्मिति प्रमातृत्वेन ²⁴च देहादि:, तदा ²⁵विपर्ययदयहेतुर्मायाशक्तिर्विमोहिनी नाम ²⁶वि-

भोरिविजृम्भते ॥ ८ ॥

यश्च ²⁷प्रातिरक्तप्रमेयजाते प्रमाता शून्यादि: स इदमिति

वेद्य एव वस्तुत: कालाद्²⁸ै: पञ्चभिश्²⁹च परतन्त्रीकृत: ।

यश्च प्रमाता शून्यादि: प्रमेये न्यतिरेकिणि ।

माता स मेय: सन् कालादिपञ्चकवेष्टित: ॥ ९ ॥

तत्र कालाद्वर्तमाना³⁰भासेन तस्य भूतभविष्यत³¹संभव:, नियते:³²कार्यकर्मफलानियम:, रागाद्रोहाभिष्वङ्ग:, विद्याकलाभ्यामचिदा-

त्मनोsस्वतन्त्रस्यापीषज्ज्ञानक्रिये चित्स्वातन्त्र्यस्³³रूपे ॥ ९ ॥

²³ eva : omitted in T, evam Ch Jh T, evam corrected to eva J. ²⁴ ca omitted in T; ca dehādih cit. and commented on ibid. p. 285 ekavacanāntam iti vrttau dehādir ity etadupekṣayā cakāreṇa sphutatvena vākyabhedo 'tra ānita ity āśayena evam uktam. ²⁵ vi-paryayadvayahetur : vibhur yaddvayahetur Ch Ñ, vibhur yadavayahetur (corrected to vi-paryayahetur) Jh, vibhur yadvaya (corrected to yaddvaya)hetur T; viparyayadvayahetur cit. ibid. p. 280. ²⁶ vibhor vijrmbhate : vijrmbhā E Ñ C Ch Jh Ñ T, vibhor °ādibhih J. ²⁷ vyatiriktaprameyajate : vyatirikta° T. ²⁸ °ādyaih : °ādyaḥ : °ādyaih J. ²⁹ ca omitted in T. ³⁰ Cf. ibid. p. 288 ... tat kim vrttāv ābhāsapadena. ³¹ °sambhavah : °bhāvah T (equally good). ³² After kārya° T adds °kāraṇa°, as, indeed, meaning would require. However, according to what Abh. seems to say (ibid. p. 290), kāraṇa is to be considered as understood in kārya and is not directly mentioned: niyatim vyākartur yā vrttih, tāṃ vivrnoti "kāryakāraṇabhāvas ca" iti ! kāryalakṣaṇaśabdena tadbhāva ucyate. Cf. ibid. p. 290 karmatatphalaniyamavayākhyāne siddhe ... and pp. 290-91 vrttau karmaśabdena cet kriyā, tadā phalam tatkrta ātmavartī sam̧skārah, atha sa eva karmaśabdena tadā phalam svargādīni. ³³ J adds iti.

64

2

त्रयोविंशतिधा मेयं यत् कार्यं करणात्मकं । तस्याविभागरूप्येकं प्रधानं मूलकारणम् ॥ ९० ॥

त्रयोविंशविधा चात्र वाच्यावाच्यन्तःकरणावली । कार्यवर्गश्च दशधा स्थूलसूक्ष्मत्वभेदतः ॥ ९९ ॥

कार्याणि द्विधा । शब्दस्पर्शरूपरसगन्धानि सूक्ष्ममतवेन तन्मात्रसंज्ञानि पञ्च, अन्योन्यनव्यूहेन स्थूलानि पृथिव्यादि-संशानि तान्येव पञ्च । श्रोत्रत्वगाक्षिजिह्वानासिकेति पञ्च बुद्धीन्द्रियाणि , वाक्पाणिपादपायूपस्थानीति पञ्च कर्मेन्द्रियाणि, मनोबुद्ध्यहंकारा इति त्रिधान्तःकरणमिति त्रयोविंशतिभेदस्य कार्यं करणातमनः प्रमेयस्य मूलभूतैकविभागदशा प्रधाना रुः इत्यागमाधिकारेऽपि प्रथममाहिकम् ॥

॥ ९९ ॥

इत्यागमाधिकारेऽपि प्रथममाहिकम् ॥

द्वितीयमाहिकम्

तत्रैतान्मात्रतामात्रस्थितौ रुद्रोऽधिदैवतम् । भिन्नप्रमेयप्रसरे ब्रह्मविष्णू व्यवस्थितौ ॥ ९ ॥

शून्ये पुरुषशक्त्यौ त्मकेदन्तःसूक्ष्ममेदह एैव वो प्रमातृमात्रशून्ये पुरुषशक्त्यौ त्मकेदन्तःसूक्ष्ममेदह एैव वो प्रमातृमात्र(...)

34 °karana° : °kāraṇa° Ñ J Ñ. 35 °cātra : cātha E Gh Ñ C T. 36 °dvidhā : dvidhāni daśadhā T. 37 °gandhāni : °gandhākhyāni E, °gandhātmakāni T. 38 T adds ca. 39 anyo-nyavyūhena, my conjectural emendation : anyonyasampatvyūho na T, anyonyavyūhasam-bandhena E Gh Ñ C Ch J Jh Ñ T. 40 °samjñāni tāny eva : °samjñātany eva T; tāny eva J Jh Ñ T. 41 E adds °ghrāṇa°. 42 °nāsikā° : °nāsikā° ca T, °nāsiketi Gh Ñ C Ch J Jh Ñ T. 43 °upasthākhyāni : °upasthāni Ñ C Ch J Jh Ñ T. 44 manobuddhyahamkāra : mano buddhir ahamkāra T, manobuddhyahamkāra Gh C Ch T. 45 °karana° : °kāraṇa° Ñ T, °ātmake 'tyantasūkṣmagrahanam (...) : °ādyarthe sūkṣma° E Gh C J, °ādyarthasūkṣma° Ñ Ch Jh Ñ T, °ātmake sūkṣma° T.

III 2.1 स्थितावन्यप्रमेयोपसंहारे स्रष्टोद्रोधिष्ठतात्, भिन्नप्रमेया⁴⁹भासे सर्ग-स्थितिहेतू ब्रह्मविष्णू ॥ १ ॥

एष प्रमाता मायान्धः संसारी कर्मबन्धनः । विद्याभिज्ञापितैश्वर्यश्चिद्‌धनो मक्त उच्चते ॥ २ ॥

मायीयशून्यादिप्रमाता⁴⁸नियत्या कर्माधीनः संसारी, विद्यावशाद्⁴eआत्मतत्त्वाभिज्ञया मुक्तः ॥ २ ॥

स्वाध्रूपेषु भावेषु प्रमाता कथ्यते पतिः । मायातो भेदिषु क्लेशकर्मादिकलुषः पशुः ॥ ३ ॥

ऐश्वर्यदशायां प्रमाता विश्वं शरीरतया पश्यन् पतिः,⁴⁰पशुपाशवशायां तु रागादि⁴¹क्लेशकर्मविपाकजैः परीतः पशुः ॥ ३ ॥

स्वातन्त्र्यहानिर्‌बोधस्य स्वातन्त्र्यस्याप्यबोधता । द्विधाणवं मलमिदं स्वस्वरूपापहायिनत॥ ४ ॥

⁴²स्वतन्त्रो बोधः⁴³परमार्थः, तथारूपत्वादेव पूर्णः । तस्य स्वातन्त्र्यादेव तथेच्छया स्वातन्त्र्यं बिना बोध⁴⁴मात्रनिर्माणे बोधतां विना स्वतन्त्र्यमात्रनिर्माणि वा, पूर्णत्वाभावेन परिमितत्वाद्‌द्विधाणवत्वं तत्त्वकस्वरूपविपर्य्यासान्मलत्वम् ॥ ८ ॥

⁴⁷ °ābhāse : °ādhāre T; cf. ibid. p. 308 sūtre ya evamarthah prasaraśabdah, sa ābhāsapadena vrttau vivrtah. ⁴⁸ niyatā karmādhīnāḥ : niyatiniyamitaḥ E Gh N C Ch J Jh Ñ T (also possible). ⁴⁹ ātmatattvābhijñayā : ātmatvābhijñatā T, ātmatattvābhijñatā Ñ ; cf. ibid. p. 311 manu ātmatattvam pratyuta heyam ... iti śankitvā āha “ātmanā” iti saṁsthīsamāso 'yam vrttāv ity arthah. ⁵⁰ pumstvāvasthāyām cit. ibid. p. 312. ⁵¹ Cf. ibid. p. 312 kleśakarmavipākaśayair yukto 'stu pumān. ⁵² svatantro bodhaḥ ity etat tattvam. ⁵³ paramārthaḥ : paramārthataḥ T. ⁵⁴ °mātā° cit. and commented on ibid. p. 314.

भिन्नवेद्यपथात्रैव मायाख्यं जन्मभोगदम् ।

कर्तृबोधे कर्म तु मायाशक्त्यैव तत्रयम् ॥ ५ ॥

अत्रैव द्विधाणवे वेद्यमभिन्नमपि भेदेन यदा भाति

तदातोऽपि विपर्य्यासाद नाम्ना मायीयं मलगम् ।

अहेतूनामपि कर्मणां जन्मादिहेतुभावविषयाविपर्य्योसादबोधात्मककर्तृंगतं

कारम् ।

तन्मलत्रयनिर्माणे प्रभोर इच्छा मायाशक्तिरुच्यते ॥ ५ ॥

शुद्धबोधात्मकत्वेsपि येषां नोत्तमकर्तृता ।

निर्मिता: स्वात्मनो भिन्ना भर्त्रा ते कर्तृतत्ययात् ॥ ६ ॥

वेद्यानुपरक्तबोधानामपि पूर्णकर्तृत्वशून्यत्वात्

स्वरूपान्यत्वेन प्रभुणा निर्माणम् ॥ ६ ॥

नोधादिलक्षणैक्येsपि तेषामन्योन्यभिन्नता ।

तथेश्वरेच्छाभेदेन ते च विज्ञानकेवलाः ॥ ७ ॥

शुद्धबोधानां बोधत्वनित्यत्वाद्यभेदेऽप्यन्योन्यभेदस्तथैवेश्वरेण

55 atraiva : atraivam G N C Ch J Jh Ñ T, tad atro 'pi C Jh Ñ T; cf. ibid. p. 318 "ato 'pi" [from the țikā] iti na kevalam pūrvo viparyāyo ya ānavamalātmā uktah. yāvad ayam apīty arthah. 56 viparyāsād nāmnā : viparyāsanamnā E Gh Ñ Ch Jh Ñ T, viṣayavāsanānāmnā C J. 58 māyīyam : māyīyaº E Gh Ñ C Ch J Jh Ñ T. 59 hetubhāviṣayaº : hetubhāve C Ch J Jh T (also possible), sya kartur dehāder bhinnavedyaprathāyām satyām ahetusu api jadeṣu karmasu nimittāva-grahātmā viparyayah / janmanas tu niyatāvadhyākṣiptasya āyuṣo bhogasya ca sukh-duḥkhādeḥ kāraṇabhūtaṃ karmam malam.

60 ichchā omitted in T. 61 yeṣām : teṣām C Ch J Jh Ñ T. 62 svātmano : svātmanā T. 63 vedyānuparaktaº cit. ibid. p. 319 "anena" : svarūpād E Gh Ñ C Ch J Jh Ñ T (tathā is also the reading of ĪPV and ĪPVV). 66 teṣām : yeṣām Gh Ñ. 67 tathā : tatra E N C Ch J Jh T (tathā is also the reading of ĪPV and ĪPVV). 68 bodhatvaº : bodhaº T. 69 tathai-va : tathā T.

III 2.7 निर्माणान्नान्यथा । ते च सांख्यपुरुषप्रापा विज्ञानकेवलैत्युच्यन्ते ॥ ७ ॥

शून्याध्वनोधरूपास्तु कर्तारः प्रलयाकलाः । तेषां कर्ममलाद्यास्ते मायीयस्तु विकल्पितः ॥ ८ ॥

शून्यप्राणाद्वै बोधरूपतातिक्रमेणान्तया स्थिताः प्रलयकेवलिनः । तेषामनूनां कर्मसंस्कारोऽप्यस्ति, मायामलस्तु वेद्ययोगायोगाभ्यां विकलिप्तः ॥ ८ ॥

बोधानामपि कर्तृत्वजुषां कार्ममलक्षतौ । भिन्नवेद्यजुषां मायामलं विदेश्वराश्च ते ॥ ९ ॥

कर्तृतायोगेऽपि बोधानां कर्मोत्तीर्णानां विदेश्वरत्वेऽपि भिन्नवेद्ययोगान्मायामलमस्त्येव । पूर्णकर्तृत्वाभावादेश्वराद् भिन्ना अन्योन्यं च पूर्ववदत एवैषामणुत्वमपि स्यात् ॥ ९ ॥

70 sāṃkhyapuruşaprāyā, though not directly quoted, is diffusely commented on ibid. p. 320. 71 E Gh C J Jh N have vijñānakala (vijñānakala N Ch T), but Abh. glöss (ibid. p. 322) refers explicitly to the reading vijñānakevalā. 72 iti only found in T. 73 śūnya0 : sūnye E Gh N C Ch J Jh N T. 74 bodha0 : deha0 E Gh N C Ch J Jh N T; bodha0 is confirmed by the quotation from the tīkā, ibid. p. 323 "bodhatātiraskārena", which pre-supposes bodharūpatātikramena. 75 api, omitted in Gh N C Ch J Jh N T, is put after tesāṃ in T. 76 °jusāṃ : °yujāṃ Gh T. 77 kartrtā0 : kartr° Gh N C Ch J Jh N T, kar-trtva0 E. 78 karma0 : karma E; cf. ibid. p. 324 karmasamskāravikalāś ca samsārottīr-na tvāt. 79 vidyeśvaratve 'pi : vidyeśvarākhyānāṃ E Gh N C Ch J Jh N T, but cf. ibid. p. 325 bodhayogena vidyāśparśah, kartrtayā ca īśvarateti "vidyeśvaratve 'pi" ity atra vrtti-vākye darśitam. 80 °yogān : °bhāktvena E Gh N C Ch J Jh N T (also possible). 81 The edited text of the vrtti stops at this point, where all the available śāradā MSS break off. From here onwards the only source (besides ĪPVV) is T. 82 īśvarābhinnā T. 83 pūrvavad ata cit ibid. p. 325. 84 Cf. ibid. p. 325 apiśabdasya bhinnakramatām darśayann asambhāvyam eṣāṃ tāvad anu rūpam, tat tu nyāyabalāyātam bhaved api ity etat dyotayati.

देवादीनां च सर्वेषां भविनां त्रिविधं मलम् ।

तत्रापि कर्ममेवैकं मुख्यं संसारकारणम् ॥ ९० ॥

देवादीनां स्थावरान्तानां संसारिणां त्रयोऽपि 85मलाः, कर्म एव संसारप्रयोक्तृकः ॥ ९० ॥

कलोड्डालितमेतच्च चित्तत्वं कर্তृतामयम् ।

अचिद्रूपस्य शून्यादेर्मितं गुणतया स्थितम् ॥ ९१ ॥

संसारिणामेतच्च चित्तत्वं कर्तृकारूपं शून्यपाणादौ जडे देवस्य कलाशक्त्योपोद्धृत्यमानमुपसर्जनत्वेनावस्थानात् परिमितम् । अतश्चा-

तस्यैव मुख्यमणुत्वम् ॥ ९१ ॥

मुख्यत्वं कर्तृतायाश्च बोधस्य च विचित्मनः ।

शून्यादौ तद्गुणे ज्ञानं तत्समावेशलक्षणम् ॥ ९२ ॥

एतदेव पुनः 86शून्यादि 87कर्तृतात्मनो बोधस्य यदो-

पसर्जनत्वेनास्ते तदास्य प्रमातुरेतद्बोधमयतामपन्नस्य ज्ञानं तच्छक्तिसमावेशलक्षणमुच्यते ॥ ९२ ॥

शून्ये बुद्ध्याद्यभावात्मन्यहान्ताकर्तृतापदे ।

अस्फुटाल्पसंस्कारमात्रेण क्रियाशून्यतां ॥ ९३ ॥

बुद्धिप्राणादिनिषेधमात्रे यदाहन्ताभिमानेन प्रमातृता तदा सौषुप्ते

85 The gender of mala fluctuates between masculine and neuter; see also above II.2.8 kārmamalo (karmo Gh ĪPV ĪPVV) and II.2.9 māyāmalam (°malo Gh ĪPV ĪPVV). 86 Cf. ibid. p. 328 tat vrttau prathamānirdeśena nirdiṣṭasya api śūnyādeḥ paścād gunatāpradarśanena vivrtam. 87 My conjectural emendation for kartrtātmano bodho yad upasarjanatvāste T, evidently corrupt. The expected meaning of the sentence is sufficiently clear. 88 bodhamayatām āpannasya cit. ibid. p. 331.

69

III 2.13 तावन्मात्रावस्थितौ वेद्याभावः संस्कारस्य संभवेsप्य् "अरूपत्वेन अनुपलक्ष्यत्वात् । "तावतैव सांख्यपुरुषाद्विरेषः ॥ १३ ॥

साक्षाणामन्तरी वृत्तिः प्राणादिपेริका मता । जीवनाभासधर्मा प्राणोधनता पुरुषेष्टकान्तिफका ॥ १४ ॥

एष एव शून्यप्रमाता "'सर्वेन्द्रियशक्तिसाधारणजीवनारण्य- प्राणादिपेरणान्तर्वृत्तीर्जीवाख्यः , तासामिन्द्रियशक्तीनामहन्तया सह पूर्वष्टकत्वमथवा प्राणाहन्तया प्रमातृत्वेन पूर्वष्टकता ॥ १४ ॥

तावन्मात्रस्थितौ प्रोक्तं सौषुप्तं प्रलयोपमम् । सवेद्यमपवेद्यं च मायामलयुतायुतम् ॥ १५ ॥

शून्यतायां "स्थिततौ वेद्याभावान्मायामलाभावः , "प्राणादौ सुख- स्पर्शादिवेद्ययोगान्मायामलवत् प्रलय इव सौषुप्तं पदम् ॥ १५ ॥

मनोमात्रप्रथेsप्यक्षविषयत्वेन विभ्रमात् । स्पष्टावभासा भावानां सृष्टिः स्वप्नपदं मतम् ॥ १६ ॥

चक्षुरादीन्द्रियशक्त्यपवृत्तावपि मनःशक्त्यैव हृष्टरूपाद्याभासेश्व-

89 My conjectural emendation for arūpatve vā T. 90 tāvataiva cit. ibid. p. 333, where the rest of the sentence also seems to be confirmed by the quotation from the țīkā : "tāvataiva iti vrttigatam evakāram vyākhyātum darśayati "sāṃkhyapuruṣāṇāṃ ca" iti / viśeṣam tu āha "kevalam" iti. 91 The compound is paraphrased and interpreted, following the țīkā, in two different ways (ibid. p. 335) : pūrvavyākhyāyine vrttau evam vigrahah - sarvendriyaśaktisādharaṇam jīvanākhyam yat pranaṇādikriyāpañcakam prānādivibhāga- kāraṇabhūtam tasyā vyāpapacanīkasya preraṇātmikā aham iti samrambhalakṣaṇā vrttir yasya śūnyasya, sa tathā uktaḥ / idānīṃ tu evam vigrahaḥ - sarvendriyaśaktīnāṃ sādhāraṇaṃ jīvanāśabdavācyam tat yasya sambandhi tasyā eva kartrtvāt, na tu tasyāṃ kāraṇatvād iti. 92 sthitau cit. ibid. p. 338. 93 prāṇādau cit. ibid. p. 338 "prāṇādau" iti vrttau "sthi- tau" iti pūrvakam yad vaiyādhikaraṇyena tad apeksaṇīyam.

रसृष्टिटरणोः स्वप्नपद्म् । अन्यप्रमातृसाधारणरूपाद्याभासनुवृत्तितः 94कालान्तराननुवृत्तेभिन्नरेषा ॥ ९६ ॥

सर्वाश्रगोचरत्वेन या तु नाङ्गतया स्थिरा । सृष्टिः साधारणी सर्वेप्रमातॄणां स जागरः ॥ ९७ ॥

सर्वान्तर्बहिष्करणशक्त्या सृष्टिटर्जागरा, तत्रापि 95पूर्ववद् द्वि- चन्द्रादिभ्रान्तिः ॥ ९७ ॥

हेया तृतीयं प्राणादेः प्राधान्यात् कर्तृतागुणे । तद्वानोपचयपायसुखदुःखादियोगतः ॥ ९८ ॥

एतানি जागरस्वप्नसुषुप्तानि प्राणादेरात्मत्वेनाभिमन्यमानस्यो- द्रेकात् स्वातन्त्र्यस्य्यापाहासाद्देयानि, स्वातन्त्र्यस्य मात्रया 96उपचयः सुखहर्षादियस्तथैव न्यूमता दुःखद्वेषादयः । सर्वभोगात्मा संसारो बन्धः ॥ ९८ ॥

प्राणापानमयः प्राणः प्रत्येकं सुप्तजाग्रतोः । तच्छेदात्मा समानाख्यः सौषुप्ते विषुवत्स्वव ॥ ९९ ॥

97स च 98प्राणात्मा 99[स्वप्ने] प्राणापानरूपश्वासप्रश्वासरूपो जागरेsथ सुप्ते 100[त्र] तयोरपि विषुवतीव रात्रिदिनयोः साम्ये 101तन्न्यूनताधिकयरोधात् समानसंज्ञः ॥ ९९ ॥

94 kālāntara° cit. ibid. p. 339. 95 pūrvavad cit. ibid. p. 340. 96 My correction, required by meaning. for °upacaya° T. 97 sa ca cit. ibid. p. 345. 98 prānātmā cit. and commented on ibid. p. 343 prānātmā iti vrttau karmadhārayo 'yam iti yāvat. 99 T adds sva-pne, apparently to be expunged. 100 T adds tra ('tra), apparently to be expunged. 101 ta-ra T, meaningless, is probably to be corrected to tan° (or expunged).

III 2.20

मध्योध्वगाम्युदानाख्यस्तुर्यगो हृतभुङ्मयः ।

विज्ञानकलमान्तेशो ग्यायो विश्वात्मकः परः ॥ २० ॥

102 प्राणापानयोः पार्श्वद्वयतिर्यक्प्रवाहविच्छेदेनैकतामापाद्योधर्व-

रूपमध्यमार्गप्रवाहेणोद्घमनाद् उदानः । 103 मध्यनेत्रमिव

104 ऊर्ध्वभाक्त्वादघिनधर्मा तुर्गदशायाम् । 104 तुर्यातीते दिक्काला105 नवच्छेदे

पूर्णे पवहातां हित्वांतिनिर्भरावस्थो ग्यानसंज्ञः । एतच्च दश-

द्वयमुपादेयं 109 प्राणशक्ते: परमेश्वर108 कल्पत्चनावस्थानात् ॥ २० ॥

इत्यागमाधिकारेऽ्वितीयमाहिकम्‌॥

इत्यागमाधिकारः ॥

102 Cf. ibid. p. 350 katham prānāpānau madhye ūrdhvvena ekatāpannau pravahata iti vrttau bhanitam ... prānāpānayoś ca katham vyatirekena tiryakpravāha uktaḥ ; pārśvad-vaya° cit. (from the ṭīkā) ibid. p. 350-51. 104 Cf. ibid. p. 350 turyādilakṣaṇam ūrdhvabhāktvādīti vibhāgah. 105 Cf. the paraphrase of the parallel passage of the ṭīkā ibid. p. 352-53 “turyātītapade tu” iti pravahanavyāpārasamparśanāhetuh, deśakālabhedāpekṣam gamanam pravahanam, deśa-kālāv api ca atra ahantaikamātrasārajātau, na tu kathamcid bhinnāv iti. 106 °anavacchede (or °anavacchedana), my correction for °anavaccheda° T. 107 Cf. ibid. p. 353 “prāṇaśaktiḥ” ity ādi parameśvaronmeṣaṃsātrūpā cety arthah. 108 °kalpa° cit. (Through the ṭīkā) ibid. p. 353 “kalpaśabdena” iti vrttigatena / kalpaśabdaḥ svabhāvavācī.

तत्वसंग्रहाधिकारः

स्वात्मैव सर्वजन्तुनामेक एव महेश्वरः ।

विश्वरूपोऽहिमेदमित्यखण्डमार्शवृंहितः ॥ १ ॥

एकः 'प्रथमोपादेयतुर्यदशायाम् अखण्डितग्राहकाखण्डितग्राह्य-

तन्मेलनाचमत्कारोपबृंहितः साधारण एव सर्वप्राणिनामात्मा विश्व-

रूपो महेश्वरः ॥ १ ॥

तत्र स्वसृष्टटेवम्भागे बुद्ध्यादि ग्राहकात्मना ।

अहंकारपरामर्शपदं नीतमनेन तत् ॥ २ ॥

महेश्वरस्य यदृम्भायेsसिमन्निर्गते तस्मिन्नदन्तापरामर्शो ग्राह्यं

1 Cf. ibid. p. 357 yad āha "ādibhūmikāyām" iti prathamopādeyatā tataḥ prabhrti ma-lavigalanārambhāt.

2 My conjectural emendation for °grāhyohamkāra° T. Abh. ibid. p. 357 refers to the 'long compound' (mahāpadam) in the vrtti, quoting very concisely passages from the commentary devoted to it by the tīkā, integrated with his own glosses.

It appears that the last word is upabrmhitah and that therefore the reading T is corrupt; in T the compound ends with grāhyo, which might be theoretically admissible if the following word 'hamkāra did not give rise to considerable perplexity.

The words that Abh. expressly quotes are: akhanditagrahaka, akhanditagrahya, melanā, upabrmhita.

It is worth quoting the whole passage: akhanditagrahaka ity ādikam vrttigatam mahāpadam vyācaṣṭe "yat tat" iti / anena grāhakapadam vyākhyātam, akhanditapadam vyācaṣṭe "yatna" ity ādi-nā "akhandita" ity antena / "tatra" ity ādinā grāhyapadam vyācaṣṭe, "tatra" ity ādinā tadviśeṣaṇam "akhandita" iti . "tayoh" ity ādinā melanāpadam / tayor iti grāhaka-grāhyayoh / "paraspara" iti akhanditapadena grāhakāsya grāhakāntarāt, grāhyasya grāhyāntarāt na viccheda iti darśitam.

amunā tu parasparato 'pīti / upabrmhitapadam vi-vṛṇoti "nirākānksā" ity antena.

This permits us to establish the general sense with sufficient precision; there remains the doubt as to what exactly the word preceding melanā must be.

My hypothesis is tan (on the basis of tayoh, which seems referred to it in the tīkā), but it might also be something like ubhaya, or ahamidam (which would be closer to the corrupt 'hamkāra T).

3 ĪPV (KSTS) ĪPVV and E read buddhyādi°.

IV 2 यन्निर्मितं बुद्धि: पाणोऽथ शून्यं तद्वैैकदेशरूपमहंकारा-

वमृश्यतापादनेन परिच्छिन्नग्राहकीकृतम् ॥ २ ॥

स्वरूपापरिज्ञानमयोऽनेक: प्रमाता ।

तत्र सृष्टौ क्रियानन्दौ भोगो दु:खसुखात्मक: ॥ ३ ॥

एतदेव विश्वात्मन: परिमितत्वकरणमप्रत्यभिज्ञानमुच्यते । एवं चानेकबुद्धिप्राणादि'खण्डगतापराहंकारपरामर्श: परापरिज्ञानसंज्ञ: । प्रत्यगात्मानो बहव:, तेषु प्रमातृरूपेषु महेश्वरेण स्वानन्द: स्वक्रियैककर्तृतानुसारिणी निर्मिता । स एव भोग आनन्दनले शाख्य: सुखसंज्ञित: क्रियालवात्मा दु:खरूप: क्रिया दु:खं च वक्ष्यते ॥ ३ ॥

स्वाकारूपेषु भावेषु पत्युज्ञानं क्रिया च या ।

मायातृतীয়े ते एव पशो: सत्त्वं रजस्तम: ॥ ४ ॥

6 ईश्वरस्य ज्ञानक्रिये ते मायया सहिते पशो: सत्त्व-

रजस्तमांसि ॥ ४ ॥

4 The same expression cit. (from the țikā) ibid. p 359 in the introduction to the fol-

lowing verse: vedyaikadeśe 'pi ity ādi. 5 T has okhandanatāpara°; okhanda° and oapara°

(aparāparaśabdena is most likely a mistake for aparaśabdena) are quoted ibid. p. 360,

°gata° is my tentative emendation for °nata°, evidently corrupt. In the same passage

°aneka° and °parijñāna° are also quoted: "anena" iti vrttigranthena / khandaśabdena

atidīrghagonāsaśarīragatābhinnajīvarve khandīkarane khandakhandānām prthagjīvatvam

punar ekībhāvāpattāv ekajīvateti dhvanatā anutvam aparāpara(read apara°)śabdena ca

parimitatvam darśitam / anekatvāpurijñāne tu svaśabdenaiva ukte. 6 The lack of com-

mentary on the particularly important first pāda might be considered as already

of the vrtti. However, the content of the first pāda might be explained above (cf. III.2.3).

भेदसिथते: शक्तिमत: शक्तित्वं नापदिशयते । एषां गुणानां करणकार्यत्वपरिणामिनाम् ॥ ५ ॥

सर्ववरजस्तमसां च भेदेनावभासनान्न शक्तिन्यपदेश: शक्तितन्मतोर्बेदात् । वस्तुत: शक्तिविकासो विश्वम् ॥ ५ ॥

सत्तानन्द: क्रिया पत्युसतदभावोऽपि स पशो: । द्वयात्मा तद्रजो दुःखं श्लेषिसत्त्वतमोमयम् ॥ ६ ॥

ईश्वरस्य आनन्तकर्तृता चमत्काररूपा क्रियोकता परमप्रकाशानन्दमयी, प्रत्यगात्मनस्तु तौ प्रकाशानन्दौ ग्राहाभूतौ सत्त्वमुच्यते । तदभावश्च तम:, सत्त्वतमसी प्रकाशानन्दतदभावरूपे द्वे अपि श्लेषातमनी रज: । पशो रज:संज्ञा क्रिया च दुःखं चोच्यते ॥ ६ ॥

ये ऽप्यासामयिकेदन्तापरामर्शभुव: प्रभो: । ते विमिश्रा विभिन्नाश्च तथा चित्राभासिन: ॥ ७ ॥

ये चैते महेश्वरस्येदंशब्दसंकेतानुसंधानं विनापि प्रकाशस्य परामर्शसारत्वाद् बालदशायामिवेदमर्थनिर्देश्या भावाः, ते स्ने कातु सामान्यलakṣaṇarūpā nirdiṣṭāḥ.

7 Cf. ibid. p. 363 "āgame" iti śrīmangalāśāstre; the tīkā is probably referring to the well-known ardhaśloka of the Mangalāśāstra (or Sarvamangalā) śaktayo 'sya jagat kṛtsnam, echoed by the vṛtti. 8 ślesi°, which in E and in the two editions of ĪPV alone, is to be taken as compounded with the following sattvatamno°, as we can gather from the vṛtti and the other commentaries (only Bh., II p. 291, refers it to rajaḥ); tattyamano° corrected to sattvatamo° T. 9 Cf. ibid. p. 364 atra hetuḥ pārimityam, tatrāpi grāhyatāpādanam. 10 The context seems to require the addition of rajaḥ. 11 Cf. ibid. p. 367 na kevalam sadyojātābālāder gocarikāryaḥ ...; cf. also ibid. p. 368. 12 Cf. the quotation (from the tīkā) ibid. p. 368 "aneka" ity ādipadena svalaksanātmānaḥ, "prthak" ity ādinā tu sāmānyalakṣaṇarūpā nirdiṣṭāḥ.

IV 7 भाससामानाधिकरण्येन १३स्वलक्षणात्मान: पृथक्सामान्यरूपतया१४नानाकाराश्च तथा तन्निर्माणात् प्रथन्ते ॥ ७ ॥

ते तु विभिन्नावभासार्थैकल्प्या: प्रत्यगात्मन: । तत्तद्विभिन्नसंज्ञाभि: स्मृत्युत्प्रेक्षादिगोचरे ॥ ८ ॥

ते विभिन्नावभासा: १५सामान्यात्मानोऽथासन् १६तदनुभवसंस्कृतै: कृशोदहं दु:खी सुखी वाहमिति विचित्र्यपदेशविषयीकि- माणात्मभि: १७क्षेत्रज्ञैर् १९विकल्पनशक्त्या तत्तद्रटजतशुक्लपटश- कटादिनाम्नान्त:प्रत्यवमर्शनीयत्वेन २०प्रदर्श्यन्ते स्मृतौ पूर्वानुभवा- प्रमोषे, विचित्रोत्प्रेक्षादिषु तु स्वातन्त्र्येण । अयमेव ग्राह्यग्राह- भेदावभास: २१शब्दमय: २२संस्कारबन्ध: ॥ ९ ॥

13 My conjectural emendation for svalakṣanātmānam T, meaningless in this context (see also fn. 12). 14 After some perplexity (nānākāráś might seem connected with citra° and in this case the text should be modified at several points to yield a satisfactory meaning) I have fully accepted T, taking nānākāráś as a gloss of vibhinnāś and referring to the sāmānyas. 15 My conjectural emendation for ye T. 16 Following Abh.'s commentary one would expect here something like viśesasāmānyātmano, but Utp.'s intention is certainly different and sāmānyātmāno continues the line of interpretation begun in the preceding verse, where vibhinnā refers to the universals. The sense of sāmānyātmano, which occurs also in the țikā, is clearly explained ibid. p. 370 "sāmānyātmanaḥ iti sva-lakṣanarūpā api śabdagocarikāryatvena sāmānyātmakaḥ eva iti prāg eva uktam asakrt.

17 Abh. points out (ibid. p. 370) that according to the vrtti the expression tattadvibhin-nasamjnābhiḥ in the kārikā refers to both the objects and the subjects: "tattadvibhinna" ity ādinā sautram vrttāv api ubhayathā yojitam. 18 kṣetra° cit. and commented on ibid. p. 379 vrttim vyācaṣṭe "atra iti kṣetrasabdena śarīram vadatā tacchūnyānām nāsti vikal-panaśaktir ity uktam. 19 vikalpana° (or vikalpa°), my correction (vikalpanāśakti occurs many times in the parallel passage ibid. p. 379): vikalpya T. 20 In T the last aksara is not clearly legible, fluctuating between te (as in the devānāgarī transcript ) and nte. But meaning requires pradarśyante. 21 śabdamayaḥ cit. and commented on ibid. p. 380 etac ca vaksyate "samsāro viśrāntaḥ" samsaranasya heyatām āha "bandha-rūpa" iti.

तस्यासाधारणी सृष्टिटरीशसृष्टयुपजीविनी ।

सैषाप्यअज्ञातया सत्यवेशशक्त्या तदात्मनः ॥ ९ ॥

स्वविश्रान्त्युपरोधाय चलया प्राणरूपया ।

विकल्पक्रियया तत्दर्शनवैचित्र्यचित्रया ॥ १० ॥

क्षेत्रज्ञशेषवरूप एव ततत्तत्सामान्यरूपानर्थानिशक्त्यैव अपरिज्ञातया सर्वसाधारणार्थदर्शनसंस्कृतानन्यवेद्यान् निर्मिमीते ।

सा चेश्वरशक्तिः स्वातममात्रविश्रान्तिविरोधाय मायावयपदेश्या तत्ककारादिवर्णभेदशतानन्तगणना प्राणरूपेण चलचलत्मापन्ना विकल्पारण्यापारा तथान्तः सृजत्यर्थीन् ॥ ९-१० ॥

साधारणोsन्यथा चैशः सर्गः स्पष्टावभासनात् ।

विकल्पहाननेनैवाग्र्याद् क्रमेंश्वरतापदम् ॥ ११ ॥

ईश्वरस्य तु सृष्टिः सर्वप्रमातृणां साधारणी तेषां तन्मध्य

tadeśakālā eva svatantrāvikalpatadeśakālā eva svatantrāvikalpa-

23 upajīvinī : upajīvanī T. 24 ajñātayā : ajñatayā Ñ (also found in E, ĪPV and ĪPVV). 25 sva : sa T. 26 All the editions of the kārikās have rūpayā (also found in C Ch Ñ T) instead of citrayā T Jh Ñ, but all the commentaries presuppose the latter (ĪPV kakārādivarnavaicitryair vicitrayā, Bh tattadvarnavaicitryacitrayā, ĪPVV aṣṭāṣṭakabheda-sahasravaicitryavicitritā). 27 sāmānyarūpān cit. and commented on ibid. p. 382 nanu niya-tadeśakālā eva svatantrāvikalpe 'pi ābhāsante arthāḥ, tat katham vrttau “sāmānyarūpān” ity uktam / etat pariharati “sāmānyarūpatvena” iti smrtiviṣayo vikalpyamāno 'pi na svotthāpitadeśakālayogī, kintu nirūdhadeśakālayogī eveti svalaksanātmā, tena svasṛṣṭa-deśakālābhāsamelanā eva arthāḥ svātantrya[svatantrā?]vikalpaktau, na tu nirūdhāv eṣām kaucanadeśakālābhāsāv ity etat tātparyam. 28 aparijñātayā, my correction: aparijñatatayā T. 29 Cf. the quotation (from the ṭīkā) ibid. p. 382 “tattat” ity ādinā “samskrta” ity antemā samāsapadena “īśasṛṣṭyupajīvinī” iti vyākhyātam. 30 sā, my correction : sa T. 31 Cit. ibid. p. 384 “varnabhedaśatānantaganana iti vrttigatena bahuvrīhinā arthāḥ. pakriyayā” iti bahuvrīhir ayam śaktim viśiṇaṣṭi, ity āha. 32 cañcalatām, my conjectural emendation : cañcalataram T (ā and ra are very similar in Malayalam script). 33 Cit. ibid. p. 386 “vikalpākhyavyāpāra” iti vrttyā sūtre “vikal-pakriyayā” iti bahuvrīhir ayam śaktim viśiṇaṣṭi, ity āha. 34 caiśaḥ sargah : ceśaḥ sarga

o avabhasanāt : o bhāsavan T (which would be acceptable and even give a better meaning, but all the MSS and the editions of the commentaries agree on avabhasanāt).

36 Cf. ibid. p. 387 sargasya pramātṛsādhāryārānyādicintā kva upayujyate iti cet pramātṛsāhityena prameyasya srṣṭir na anyathā bhedāvabhāsabhājānam vinā tad anupapatter iti. 37 srṣṭih, my correction : srṣṭeh T.

एवोत्पादात् , एकप्रमातृनियता चैकप्रमात्रवेशेन स्वप्नभान्त्यादौ सृष्टिः । ³⁸सा चाहमिदमित्येतावत्परामर्शमयी ³⁹भेदानुदयादि-कल्पनिर्हासपरिशीलनेन संसारिणां कमेणैकरसैश्वर्योन्मातृ क्षेत्रज्ञताविमुक्तिः ॥ ११ ॥

⁴¹sarvo 'ham ity evaṃ parijānataḥ | viśvātmāno vikalpānāṃ prasarē'pi mahēśatā ॥ १२ ॥

kṣetrajñasya piśvarasaktyai vaikalpāram-bham iti tadṛśyayām api

³⁸ sā cāham idam ity etāvar cit. ibid. p. 387. ³⁹ bhedānudayāt cit. (from the ṭikā) ibid. p. 388. ⁴⁰ The text is doubtful and my restoration is only tentative. Here the second part of the kārikā is glossed (cf. ibid. p. 388 evam ślokārdham vyākhyāyatam, which concludes the gloss of the first half). T reads tantrāntarāntarod yat kṣetrajñāvyāpāravikalpa-nirhrāsaparaśilpanena. I correct tantrā to tatrā on the basis of the quotation ibid. p. 388 tadvṛttim "tatrāntara" ity ādikam; antarāntarā is confirmed by its mention in the ṭikā "yady api" iti / "sahabhāve ca antarāntarā" iti yuktam iti bhāvah / "tathāpi" iti na atra vikalpasṛṣṭar antarāntarā sambhava uktaḥ, kintu tannirhrāsasya. In the same compound, after kṣetrajña, T has some akṣaras deleted and corrected, which are read in the transcript as vāpārakalpa. Actually, T has vikalpāpāra, where the part ikalpā is deleted and va is corrected to vya, thus forming the word vyāpāra followed not by kalpa but by vikal-panirhrāsa occurs many times ibid. p. 389. Then I read pariśilanena instead of paraśilpanena T (my conjecture is supported ibid. p. 389 tam eva ca vikalpanirhrāsam pariśīlayati .... ⁴¹ sarvo : so 'ham E. According to Bh (II p. 305), sarvo, found in the text of ĪP commented on by him, would be the corruption of the original reading sargo, graphically rather similar in śāradā script. In his opinion sargo would fit the context better (ayam sargo, that is, the creation of the vikalpas of which it is question), but the decision is left to the reader's sensitivity (sahrdayā atra pramānam). However, a close examination of the vrtti induces me to think that the original reading was precisely sar-vo (which is not directly glossed, but which lends itself to not being glossed). I would even add that Abh.'s commentaries may allow one to suppose that his text had precisely the corrupt reading sargo (he would gloss it, in fact, with vikalpasṛṣṭiḥ), then returned, by a further corruption or correction, to the original one which was in front of Bhāskara. Accordingly, Abh.'s interpretation of vibhava is also different from the vrtti's. But another hypothesis may also be put forward. Bhāskara records and immediately discards a further reading so 'ham (kutrāpy ādarśe "so 'ham" iti pāṭho 'sti, sa tu na yuktataḥ). This reading, found in E, is accepted by Śivopādhyāya in the Vijñānabhairavivṛti (p. 95) and fiercely defended against the upholders of the other two (ibid. p. 96) ye tu "sarvo mamāyam, sargo mamāyam" ity ādipāṭhāntaraṃ kalpayaṃti te pāṃḍitamanyāḥ ... mūḍhāḥ). The

परिज्ञातेश्वरभावस्य ममायं संसारमयो विभव इत्यभेदेन IV 12 विश्वमाविशतः परामर्शमात्रान्वितकाल्पानुसंपादयतो महेश्वरतैव ॥ १२ ॥

मेयं साधारणां मुक्तः स्वात्माभेदेन मन्यते । महेश्वरौ यथा बद्धः पुनरत्नन्तभेदवत् ॥ १३ ॥ बद्धमुक्तयोर्‌वेद्यमेकम्, किंतु बद्धो‌ऽत्नन्तविभेदेन तद्‌द्वैति विमुक्तः स्वात्मदेहत्वेन ॥ १३ ॥

सर्वथा त्वन्तरालीनानन्ततर्‌वौघनिर्भरः । शिवश्‍चिदानन्दघनः परमाक्षरविग्रहः ॥ १४ ॥ सर्वथा त्वन्तर्लीने प्रमेये ४२शहंमतौ पूर्णायां शिवतैव ॥ १४ ॥

एवमात्मानमेतस्य सम्पक्‍ज्ञानक्रिये तथा । ४३पश्‍यन् यथेप्सितार्थान् जानाति च करोति च ॥ १५ ॥ इत्थमाविष्कृतशक्‍त्यभिज्ञानम् आत्मानमनन्तज्ञानक्रियाशक्‍ति४४निर्भतमेश्वरं प्रत्यभिज्ञाय यथेच्छं सर्व पश्‍यति निर्मिमीते ॥ १५ ॥

first pāda of the kārikā, in this form, would refer to the two aspects of 'recognition' (pratyabhijñāvaidham), connected by him with the vv. 109-10 of the Vijñānabhairava. According to Śivopādhyāya, this would be the reading commented on by Abh. (pratyabhijñāvimarśinīkarena); and indeed ĪPV II p. 305 (and almost in the same terms ĪPVV III p.393) says: na hi sah pratyagātmā nāma paśuh kaścid anyo 'ham, api tu parigrahyāgrahākaprakāśaikaghanah paro yah sa evāham sa cāham eva .... But this may also be a mere digression of Abh., not directly dependent on the text of the kārikā. Lastly, sahrdayā atra pramāṇam. 42 Cf. ibid. p. 395 "cidānandaghana" iti sautram ahampratītau pūrṇāyām" itīyatā vyākhyātam. 43 paśyan yathepsītān arthān : jānan yathe- psītān paśyan C Ñ (this is also the reading of ĪPV, ĪPVV and Bh), paśyan yathepsītān ātmā Ch Jh Ṭ (ātmā then corrected to arthān in Ṭ). The text of E Ṭ Ñ is the one presupposed by all the commentaries. 44 This word in Ṭ is difficult to read. Among the various options °nibhrtam (or °nirbhrtam) seems to me the only one that may furnish an acceptable meaning (the transcript has °nivṛtam).

इति प्रकटितो मया सुघट एष मार्गो नवो

महागुरुबिरुच्यते स्म शिवदृष्टिशास्त्रे यथा ।

तदत्र निर्द्धृतपदं भुवनकर्तृतामातनो

विभाव्य शिवतामयीमनिशमविशंसद्यति ॥ १६ ॥

यत्र यथावस्थित एव न्यवहारे प्रत्यभिज्ञामात्रात् शिवतालाभः ।

सोऽयमवक् एवाभिनवो मार्गः ⁴⁵साक्षात्कृतपरमेश्वरभट्टारकाक-

रैर्भट्टश्रीसोमानन्दपादैः शिवदृष्टिनाम्नि प्रकरणे निर्दिष्टो मया

युक्तिनिबन्धनेन हृदयंगमीकृतः। एतत्परिशीलनेन शिवतावेशाज्

जीवन् ⁴⁶एव मुक्तो भवति ॥ १६ ॥

तैस्तैरुपायैश्चितैरुपनतस्तत्न्या: सिथतोऽप्यनितके

कान्तो लोकसमान एवमपरिज्ञातो न रन्तुं यथा ।

लोकस्पैष तथानवेक्षितगुणः स्वात्मापि विश्वेश्वरో

नैवालं निजवैभवाय तदियं

तत्पत्यभिज्ञोदिता ॥ १७ ॥

कान्तदृष्टान्तेन ईश्वरप्रत्यभिज्ञोपपादन्या चापूर्वफला च

॥ १७ ॥

45 Cf. ĪPVV I p. 38 tīkāro 'pi "sākṣātkṛtaparameśvarakārair" iti vaksyati. 46 eva cit. in ĪPVV III p. 403 "jīvanmukto na saṃśayah" ity ādiṣu āgamāsthānesu niyamārtha-vyākhyānāya vṛttau ya evakārah ... / adbhūtatvam api ca evakāreṇa sūcita asya upadeśasya.

जनस्यायतनसिद्ध्यर्थमुदयाकरसूनुना ।

ईश्वरप्रत्यभिज्ञोन्मेषलेनोपपादिता ॥ १८ ॥

सुलभोऽयं सिद्धिमार्गः प्रदर्शितः ॥ १८ ॥

इति तत्त्वसंग्रहाधिकारः ॥

॥ समाप्तेयमीश्वरप्रत्यभिज्ञावृत्तिः ॥

TRANSLATION

SECTION I. KNOWLEDGE

Chapter I

1. Having in some way1 attained the state of servant of Maheśvara and wishing to offer assistance also to the whole of mankind, I shall - by giving it logical justification - make possible the awakening2 of the recognition of the Lord, which brings about the achievement of all success3.

• I who have obtained, thanks to the benevolence of the Supreme Lord4, the benefits that derive from being His servant5- a state it is very difficult to achieve - being ashamed of my solitary success, shall, by the method that will here be described, enable the whole of mankind to recognize their Lord, in order to gain my complete fulfillment through the attainment also by them of the Supreme Reality. - 1 -

2. What intelligent being could ever deny or establish the cognizer and agent, the Self, Maheśvara, established from the beginning (ādisiddhe)?

1 I.e. in a way that eludes all description.

2 I have thus tried to render the various meanings that Utp. gives to the one verb upapādayāmi (cf. ĪPVV I pp. 31-32; cf. also the closing verse of the ĪPK). The main meaning of the causative here is samarthācaraṇa, i.e., to act in such a way as to bring about the production of a certain action (in this case the awakening of recognition).

3 Thus, following the order that appears most obvious, I refer the compound sama-stasampatsamavāptihetum to pratyabhijñām, but, as Abh. points out on the basis of the tīkā and the vrtti, it not only qualifies tatpratyabhijñām but also dāsyam. The compound, which is implicitly understood as tatpurusa in the vrtti, is the object of intricate interpretations, sometimes as tatpurusa and sometimes as bahuvrīhi, in the ĪPV.

4 Or also, as Abh. suggests (see text note 1), «thanks to the benevolence of the master».

5 Cf. the vigraha of the compound (see text note 3).

85

I 1.2

• The Self of all beings, the substratum of the establishment of all objects6, who embraces the establishment of himself7 - since otherwise it would be impossible to establish all the various objects - self-luminous, whose nature is uniquely that of cognizer, 'formerly8 established, 'ancient', possesses knowledge and action. Sovereignty (aiśvaryam) is established through inner awareness. Therefore only the foolish strive to establish or deny the Lord. - 2 -

3. However, since He, though being directly perceived (drṣṭe 'pi), is not discerned for what He is because of delusion, precisely for this reason, by bringing His powers to light, the recognition (pratyabhijñā) of Him is shown.

• However, since the Lord, though established through inner awareness9 (svasamvedanasiddham), does not enter the sphere of full and definition (viśrāntisthāna) of the cognitive act, at which

6According to Abh. (see text note 11), this is to be understood as meaning that the cognizer constitutes the final stage (viśrāntisthāna) of the cognitive act, at which the revelation of the object (prakāśa) becomes reflective awareness (vimarśa).

7The Cognizer becomes implicitly conscious of himself, of his being 'light', precisely through perception (the 'illumination' of the object); indeed, the illumination of the object - its perception - presupposes a light into which it enters, a light that by definition is not its own but pertains to the subject.

8The being established of the subject always 'precedes', as an a priori condition of knowledge, and for this reason eludes any objectification that attempts to capture it, so to speak, from behind. That would be like trying to step ahead of one's own shadow, as a well-known verse of the Trikahrdaya (quoted in SSV p. 4) puts it. Abh. (ĪPVV I p. 51) gives the example of inferring fire from smoke, which, indeed, implies that it is always the smoke that is already established first: the difference being that the priority of the self is ad infinitum (yadā yadā upakramah tadā tadā pūrvasiddhatvam ity eso 'tra paramārthah). This is precisely what the nearness of pūrvasiddha to purāṇa signifies, which taken singly may apply to various things (see text note 15); the same theme will be taken up again later.

9The expression svasamvedanasiddha interprets and orientates, by limiting its meaning, drṣṭe in the kārikā, which, insofar as it is passive, seems to degrade to the status of object He who is the subject par excellence, the Lord. But, Abh. notes (see text note 19), this expression, too, is, in the strict sense of the word, inadequate, as the tīkā itself has pointed out (nāpi svasamvedanasiddhatvam ātmanah); it can be accepted only if understood in a metaphorical sense, as expressing the undeniability of the experience of the I. The theme of the absolute impossibility of objectifying the I. Śiva. recurs with particular insistence in the work of Utp. Cf. ĪPVV III p. 162 yat prameyīkrto 'smīti sarvo 'py ātmani lajjate / katham prameyīkaraṇaṃ sahatāṃ tan maheśvaraḥ // 'Everyone feels ashamed in himself at seeing that he is transformed into an object of cognition; then, how might the Great Lord stand this?'

86

nite knowledge (ahrdayangamatvāt)10 because of the delusion caused by māyā, His mere 'recognition'11 is here shown - in the form of the acquisition of unswerving certainty - through the illustration of that sign of recognition that is represented by the faculties peculiar to Him12. - 3 -

4. Indeed, the foundation of insentient realities rests on the living being; knowledge and action are considered the life of the living being.

• There are two kinds of reality: insentient and sentient. The establishment of an insentient nature rests on the living being; the being such of the living, i.e. life, is represented precisely by knowledge and action. - 4 -

5. Knowledge is self-established (svatah siddham); action, when it manifests itself through a body, becomes cognizable also by others. Thanks to it, knowledge in others can be guessed.

• In living beings action, when it reaches the final stage of bodily movement, also becomes directly perceptible in other subjects13; knowledge, which is in itself capable of self-perception, becomes evident (prasidhyati) also in others precisely through action14. Therefore, the Lord,

10 Lit.: 'does not reach the heart'. i.e. the plane of vimarsa which alone makes a cognition fully accomplished and effective (cf. ĪPVV I p. 80).

11 The reality of the Self is not therefore 'known' - as happens with any object in phenomenal reality, which, not shining itself, needs to be illuminated by consciousness - but is simply 'recognized'. Moreover, this act of recognition is not something that was not formerly and must be brought into existence, but eternally present and merely concealed, it is only 'shown' (darśyate) as predominant, by placing it near (upa), in contact with the heart (ĪPVV I p. 87 see text note 26). Thus Abh. intends to remove the act of recognition from the sphere of the vyavahārasādhana as well (cf. p. 173 n. 3). In fact, there may not be anything asiddha in the Lord, not even as regards any aspect of vyavahāra (prakhyā-upākhyā, jñāna-abhidhāna) concerning him, for in this case even the establishment of the smallest portion of the knowable would become impossible (ĪPVV I p. 87 tadasiddhau prameyaleśasyāpi hi na kācit siddhir ity uktam).

12 That is, above all, the powers of knowledge and action.

13 This means that action, as an inner reality (vimarsa), inseparable from knowledge, is also self-established. The possibility of objectification only regards its extreme phenomenal form of bodily movement.

14 If action is examined first in the vrtti, thus reversing the order of the kārikā where knowledge is mentioned first, this is because action is the means by which knowledge (from which it is inseparable) can be inferred in others (ĪPVV I pp. 104-105). That bodily action, and the use of the word, presuppose knowledge is a general rule.

I 1.5 the Self perceived as 'I' in oneself and others, is established insofar as it is directly experienced through inner awareness. Because of the obscuring of his true nature caused by the power of māyā, the Self is thus15 erroneously conceived. - 5 -

rally recognized fact; cf. for instance Dharmakīrti's Santānāntarasiddhi 1-2 (Kitagawa 1955: 58; Stcherbatsky 1969: 64). But to infer knowledge means to objectify it and it is for this reason that Utp. in the kārikā and in the țikā prefers to resort to other expressions, such as to conjecture, suppose, guess etc. (cf. ūhanam tarkanam sambhāvanam ibid. p. 101). The problem, however, arises again and is given an elaborate solution in IPVV I pp. 105-108. Abh. begins by stating that since it is a question of an inference founded on the svabhāvahētu (see below p. 179 n. 17) - that does not aim to make known an object not formerly known, but only to rid it of erroneous conceptions - the real basic self-luminosity of knowledge and of the subject is not contradicted. After a series of objections and replies, following the line of argument in the țikā, the conclusion is reached: a certain level of objectification in inferring knowledge in another body is undeniable, but this applies only to the initial and medial stages, whereas in the end (paryante) knowledge appears in all its luminosity, at unity with the subject making the inference, as happens at the conclusion of every cognitive act. In saying prasiddhyati the vrtti intends to express the emergence of the natural self-luminosity of knowledge and to exclude the possibility of its being a luminosity induced by the subject, as is the case for the common object of knowledge (siddham bhavati, na tu sādhyaṃ).

15 I.e. in the way that will be described in the following chapter. In understanding tasya both as objective and subjective genitive (it is Abh. himself who underlines this, see text note 43), the Self becomes both the object and the subject of the error: the Self in its freedom is mistaken about itself. Indeed, every reality, even error, has the Lord, the Self, as its ultimate source.

88

I 2.1-2

Chapter II

1-2. [Objection]1 There is one type of cognition2 in which the particular reality (svalaksana3)3 appears and another type of cognition,

1 In the whole of āhnika II a Buddhist pūrvapakṣin is speaking; his arguments are broadly speaking those of the Buddhist logicians, that gradually become more specifically vijñānavādin, sautrāntika etc. The target of Buddhist criticism here are some ātmavādins whose (various) positions are only partially shared by the Śaiva ātmavādin. One might almost say that Utp. sometimes uses the Buddhists' own weapons to demolish doctrines which, though apparently closer to the Śaiva positions, remain, however, extraneous to what is their core and essential tenor. In the text the word 'Objection' refers to the Buddhists and the word 'Reply' refers to the ātmavādins.

2 I have understood ekam ... param here in the plainest sense, which is also the one explained in the vrtti. Abh. (IPVV I p. 114) lists three possible interpretations, and the one accepted here is the third. The second, which Utp. adopts in the vīkā, takes ekam as 'single, undifferentiated', in this sense referring to direct experience (anubhava) as opposed to the vikalpa, which assumes different forms (memory, imagination, doubt etc.). The only differentiation in direct experience is the one deriving from the variety of its objects. Thus param, apart from 'another' correlated to 'one', is also to be understood as 'subsequent', in the sense that the vikalpa always depends on a previous direct experience.

3 The svalaksana - which Dharmottara understands as a karmadhāraya 'unique character, unique reality' (NBT p. 70 svam asādhāranam laksanam tattvam svalaksanam) - is, for the Buddhist logicians, the object of direct perception. Dharmakīrti (no definition of svalaksana is found in the extant works of Dignāga) describes it as follows: capable of causal efficiency, dissimilar from everything else, not being the object of words and not being known in the presence of signs that are different from it (PV III.1-2); and that which, owing to its being far or near, determines a difference in the form that appears in cognition (NB I.13). The svalaksana - says Mokṣākaragupta (TBh p. 11) - is something real, unique, determined by a space, a time and a form which are peculiar to it; he gives the example of a particular jar, capable of containing water, with a definite place, time and form, directly present to knowledge, devoid of differentiation into many properties, different from what belongs to the same or a different class. The svalaksana strictu sensu is an absolutely undivided reality; it is an instant (NBT p. 71 pratyaksasya hi ksana eko grāhyah). The term svalaksana is borrowed by the śaiva authors to denote "an entity having a character which is only its own, consisting in a determined place, time and form which 'contract' its own nature" (ĪPV I p. 86 svam anyānuyāyi svarūpasamkocabhāji laksanam deśakālākārarūpam yasya ...). The basic difference

89

I 2.1-2 called mental elaboration (vikalpa°), inseparably connected with discourse (sābhilāpam), which appears in manifold forms. For neither of the two is there any necessity to posit any stable perceiving subject, since he does not appear4 in them. Also the notion of ‘I’ (ahampratītiḥ) has in reality as referent the body etc...

• One type of cognition, consisting in the direct perception of the clearly manifested (sphuṭāvabhasā°) particular reality, is called ‘exempt from mental elaborations’ (nirvikalpakam); the other type of cognition, on the other hand, which, permeated by the word, appears in the various forms of memory, doubt, fantasy etc., is called ‘representation, mental elaboration’ (vikalpā°). Neither the one nor the other are admissible as depending on another entity distinct from knowledge itself in the form of consciousness, since this other entity is not perceived (tasyānupalabdheḥ). Who, then, is this permanent Self? Even on the basis of the notion of ‘I’, which is indissolubly connected with discourse, the existence of a cognizer, who transcends what are simply cognizable realities, namely the body and so on, cannot be ascertained5. - 1-2 -

with the Buddhists lies precisely in conceiving of place-time-form not as the intrinsic nature of the thing but as the contingent contraction of an entity whose real nature is, on the contrary, expanded (vikāsitavabhāva eva, vītātmā; cf. ŚD V.6c), ‘open’ (cf. ĪPVV II p. 27). Though appearing as a unitary entity, the svalakṣaṇa, in Utp.'s view, is in fact formed by a number of ābhāsas, i.e. universals, which have become a particular owing to their mutual delimitation and their connection with space and time (Torella 1987: 168-69). In the process of particularization space and time are considered the most essential elements (antaranga). If, already in perception itself, the svalakṣaṇa appears as a unitary reality (the world of practical experience is, indeed, made of svalakṣaṇas), this is due to the unifying power of the mind (anu-samdhāna) which links together a group of ābhāsas by making them subordinated to a predominant one (sāmānādhikaraṇya; see below pp. 166-167) Thus the śaiva position may be defined ābhāsanikurumbā̄timakavāda as opposed to the niramśalaksanavāda of the Buddhists (on svalaksana in the Buddhist pramāṇa tradition see Stcherbatsky 1930³: 181-198, Katsura 1991: 135-137).

4 The Buddhist logicians acknowledge the validity of non-perception as the criterion for establishing the non-existence of something that could (if present) be perceived (drśyānupalabdhi); this subject will be extensively taken up again later.

5 The Buddhist knows he is laying himself open to an obvious objection: how is it possible to say that no ‘I’ is manifested in knowledge when, on the contrary, we all usually say «I know, I am happy or unhappy, I am thin or fat». But he has his answer ready: this notion of ‘I’ does not reveal a permanent subject but refers to the series of distinct moments of cognition (jñānasantāna) and of body (śarīrasantāna) on which apparent personal identity is based. And in any case this experience

3. [Reply] How could we explain memory, which conforms to direct perception when the latter is no longer present, if there were not a permanent self, who is the subject of the perception?

• Since the former direct perception has disappeared at the moment of the memory, the memory, whose essential quality is precisely its dependence on that former perception of the object, could not arise, unless one admits the persistence6 of the awareness of this perception also at the moment of the memory. And this lasting awareness at different times is precisely the self, the perceiving subject. - 3 -

4. [Objection] Even if we do acknowledge the existence of a self, memory is still not explained, given that the perception no longer exists and that only through it [the perception] does memory have access to the objects formerly perceived. [Reply]. But memory acts on those very things that were the object of the perception...

• Once the direct perception (manifestation) of the object has ceased, the object no longer exists even for memory, since it is assumed precisely through direct perception. Therefore, even if one acknowledges a self consisting in a unitary consciousness, memory finds itself without an object and thus all worldly activity collapses. If, on the other hand, you claim that memory has as its object that of a perception that no longer exists... - 4 -

5. ...insofar as the occurrence of memory is due to the latent impressions left by direct7 perception. [Objection] If that is how things stand, what need is there for this useless burden of a permanent self?

would come into the category of the vikalpa, since it takes place through verbal mediation (the word 'I'); and therefore, even if an 'I' existed, it would become transformed into object.

6 Or, if we accept the reading āvartate or āvarteta, 'the arising again'.

7 This is the reason for the ātmavādin's reply; the ātmavādin and the Buddhist pūrvapakṣin completely agree on this point. However, while this admission allows the ātmavādin to elude the criticism levelled at him in the previous verse, he lays himself open to further, even more radical criticism: once the fundamental role that the saṃskāra plays in memory is acknowledged, resorting to a permanent self becomes superfluous. The grounds of the Buddhist argumentation are clearly outlined by Abh. (ĪPVV I pp. 126-7). Every cognitive act has two levels: an outer and an inner one. The first (bahirnukha) consists in the illumination or perception of the

I 2.5 From direct perception there derives a latent impression (samskārah);

the memory arising from this conforms to that former perception and

makes that perception - in which the object is immersed - manifest. If

things are explained in these terms, why is it necessary to assume a use-

less permanent subject, since even the supporter of the existence of the

self acknowledges the existence of the latent impression and this alone

is enough to account for memory? - 5 -

6. If the qualities are separate [extrinsic to the self], then, since the

nature of the self remains unaltered, the latent impressions are sufficient

to explain the phenomenon of memory. The subject of memory is there-

fore only a mental construct (kalpitah), as was the subject of the percep-

tion.

• The self, even if it is invoked as a substratum of qualities that are

distinct from it, such as pleasure, pain, knowledge etc. serves no useful

purpose. Indeed, it does not perform any function in the phenomenon of

memory8, since it does not undergo any modification, seeing that it does

not combine with the above-mentioned qualities, which are conceived as

separate. Therefore, as in the case of the subject of perception, to say

that the self is the subject of memory is purely a mental construct. - 6 -

object, the second (antarmukha) is constituted by the self-consciousness (sva-

samvedana) of this perception. According to the Buddhist, the ātmavādin stakes

everything on this continuity of self-consciousness (to which he gives the name of

ātman) to explain the phenomenon of memory, and in so doing leaves memory

without an object which can only derive from the bahirmukha level of the former

cognition. Now, if this has completely ceased there cannot be a memory (which cor-

responds to saying 'that') and, if it is fully present there, there is only a new direct

perception. What is required is 'the non-cessation of something that has however

ceased (naṣṭasyāpy anāśah)' and this is precisely what the samskāra is. Even the

ātmavādin cannot refute this either and thus he gives renewed vigour to the Bud-

dhist's attack, which is expressed in the second part of the kārikā.

8 The kārikā is parrying an objection that the ātmavādin might raise, this time

on behalf of a Vaiśeṣika: the samskāra is a quality (dharma) of the self, on a par

with knowledge, pleasure and pain, and as such presupposes a substratum (dharmin)

where it inheres (i.e. the self); the dharmin, on the other hand, cannot be represented

by a santāna, because it must be a unitary reality (eka) (cf. ĪPVV I p. 130); but if

one wishes to preserve the permanence of the ātman one must consider these

qualities as separate from him and not able to modify him (this is explained in ĪPV

I pp. 92-3 and ĪPVV I p. 130). In this case, however, neither a samskāra which does

not modify the self, nor a self which is not modified by the samskāra, can explain

the phenomenon of memory.

7. If9 cognition were conscious (citsvarūpam) then it ought to be permanent like the self10; if, on the contrary, it were not sentient, how could it illuminate objects?

• If cognition were by nature conscious, then, unable to be associated with time and place - which are qualities pertaining to objects -, it would become permanent etc., like the self11. If, on the other hand, it were not sentient, how could it illuminate the object? - 7 -

8. [Reply]. Just as the intellect assumes the form of the object, so it assumes the sentience (caitanyam) of the self12. [Objection]. In that case it is not insentient, for if it were so, it could not illuminate the object.

9 After confuting those who maintain that the existence of the self is directly perceptible or inferable, the pūrvapakṣin turns to attacking the 'sovereignty' (aiśvarya) of the self, i.e. its being endowed with the powers of cognition and action. Cognition as a reality or function distinct from the self is primarily criticized; this is a conception that may be ascribed to the ātmavādins in general but - as we shall see - rejected by the Śaivas.

10 This is, of course, the opinion held by the majority of ātmavādins. The reasoning is as follows: the self is nitya, because it is citsvarūpa svaprakāśa, and time can qualify only objects. Abh. (ĪPVV I pp. 135-6) mentions the positions of the followers of the Sāṃkhya (kāpilānām tāvat nitya eva asau), of the Mīmāṃsakas (jaiminīyāḥ puruṣam samvidrūpam eva sukhādyavasthābhinnaṃ manyante ... cidrūpamśena nityah, avasthāṃśena tu anityo 'sti, this is a doctrine which stems from the Bhāṭṭas, cf. ŚV, Ātmavāda, vv. 26 ff.) etc. The positions of the ātmavādins differ greatly as regards whether the self is conscious or not.

11 If cognition, as distinct from the self, is conceived as being conscious by nature - and such, for instance, is the position of the Vaiśeṣikas and the Prābhākaras - it follows that it, too, is permanent. The contradiction that arises (in the eyes of the Buddhist) and is not made clear in the ṛtti nor, it seems, in the ṭīkā, is pointed out by Abh. (ĪPVV I p. 103): if both jñāna and ātman are nitya, there can be no relation between them, because, according to the Buddhists, the only relation is that of cause and effect, which, indeed, cannot exist between two permanent realities. Abh. (ĪPVV I p. 134) points out a further consequence: cognition thus having the same properties as the self, would only be another self, and thus the hetu that it represents in the inference of the self would become asiddha.

12 This is the Sāṃkhya theory, according to which the buddhi, though itself insentient, cognizes by means of the light that is reflected in it from the puruṣa. Thus the pūrvapakṣin recognizes as untenable (in order for it to be able to fulfil its function in the final analysis it must be sentient and therefore the criticism levelled earlier is valid). The view of the Sāṃkhya is inadmissible - it is said, among other, in the ĪPV and ĪPVV - because only a clearer thing can receive the reflection of another and the buddhi is certainly no 'clearer' than the self.

I 2.8 The intellect (buddhih) is cognition13. Though it is itself insentient, just as it assumes the reflection of the form of the object, so, in the same way, it also assumes the reflection of the sentience pertaining to the self. Thus [in assuming them both14], it can illuminate the object. However, [it is replied] it follows that it must be sentient by nature. Therefore15 cognition, though it exists, is not related to anything else [i.e. to a subject], because this has been recognized as untenable. Action, on the other hand, neither exists in itself, nor as related to anything else. - 8 -

9. Also action [as a separate reality is logically inadmissible as it] consists in the coming into being of bodies etc. in different places etc. (tattaddesādijātatā)16 and nothing more, since nothing more is perceived; nor is it tenable that it - being one and also characterized by succession - may be related to a unitary reality.

• Action, too, conceived as one and constituted by various preceding and succeeding parts, consisting in the activity of factors (kāraka°)17, is not tenable, as it is not possible to attribute unity to something characterized by succession, i.e. that exists in a multiplicity of moments18. Nor is it admissible that it has a substratum that is both characterized by tem-kārikā from which the conclusion is reached that cognition is neither permanent nor related to a subject.

13 The various reasons for this apparently inappropriate identification (of the organ with the function etc.), are extensively analyzed in ĪPVV I pp. 149-50.

14 Cf. text note 87.

15 That is, on the grounds of the arguments put forward in this and the previous kārikā from which the conclusion is reached that cognition is neither permanent nor related to a subject.

16 Cf. e.g. TS 706 desāntaropalabdhes tu nairantaryena janmanah / samānapa-ravastūnām gatibhrāntīḥ pradīpavat //.

17 This is an allusion to Patañjali's famous definition, kārakānām pravrttiviśeṣah kriyā (Mahābhāṣya, vol. I p. 258), understood in various ways by generations of commentators (e.g. What is the meaning of kārakānām? The subject, the subject and object, all the kārakas without distinction? etc.). This is the background to the whole of the Kriyāsamuddeśa of the VP, which is, in turn, the reference point for the Śaiva conception of action (cf. the extensive and repeated quotations from this fundamental text in the ĪPVV's comment on this verse).

18 This is a reply to an implicit objection, formulated in ĪPVV I pp. 186-7: there are, indeed, collective units such as the army etc., but here the various antecedent and consequent parts, linked by a reciprocal relation, exist simultaneously; this is not so in the case of action.

poral succession and is unitary in nature19. Action is only 'to go' 'to I 2.9 change' and so on, which is but the assumption on the part of bodies etc. of new states of existence at various and diverse times and places, since nothing else distinct from this is perceived. - 9 -

10. The various things come into being in concomitance with the presence of certain other things: this is what is experienced and nothing more. There is no relation (sambandhah) other than that of cause and effect20.

• What is directly perceived is simply that, when a certain preceding thing is present, a subsequent thing comes into being. As was argued in the case of action, no relation of action and factors (kriyākārakasambandhah) exists, since such a 'relation' is not perceived as a distinct entity. There is no connection between things other than that of cause and effect. - 10 -

11. [A relation (sambandha)21, however conceived, is inadmiss-

19 The difference in nature inherent in the various antecedent and consequent parts determines a difference in nature in the substratum they share (cf. IPVV I p. 193).

20 IPV I p. 117 summarizes the series of criticisms the purvapaksin levelled against the Saiva statement «the I is sovereign (īśvara) because it is in union with knowledge and action (jñanakriyāyogāt)». Jñāna is produced (kārya) by its own group of causes, such as the faculty of sight etc.; kriyā does not exist: therefore they cannot be kāryas of the self (nor can the self, by definition, be a kārya of them). And therefore, given that the only relation acknowledged by the Buddhists is that of kārya-kārana, it follows that there cannot be any relation between jñāna and kriyā on the one hand and the self on the other.

21 In presenting the Buddhist criticism of the concept of relation (sambandha) Utp. refers to Dharmakīrti's Sambandhaparikṣā. The complete text of this short work, and the vrtti with which Dharmakīrti accompanied it, has only survived in Tibetan translation (the commentaries Sambandhaparīkṣātīkā and *anusāra, respectively by Vinitadeva and Śankaranandana, are also extant only in Tibetan translation). The Sanskrit text of twenty-two of the twenty-five kārikās that compose it can be obtained from Devasūri's Syādvādaratnākara (see Frauwallner 1934: 261-300) and Prabhācandra's Prameyakamalamārtānda (see Jha 1990). Like Dharmakīrti, after dealing with the concept of relation in various parts of his magnum opus, the PV, felt the need to return to it both organically and concisely in the SP, similarly - and this does not seem merely coincidental - after the IPK, Utp. also composed the Sambandhasiddhi (with a vrtti), a short work in which he outlines the Saiva conception of relation. Indeed, the critical reference point for the latter is the SP, an ardhaśloka

I 2.11 ible]22 since, as it rests on the two related terms23, it cannot be unitary in nature; since a thing that is [already] accomplished (siddhasya) can- not 'require' (°apekṣanāt) another and dependence (pāratantrya°) etc. are not logically tenable24. Thus the agent, too, is merely a mental construct.

of which is even included in the text (4cd ity amiśrāḥ svayam bhāvās tān yojayati kalpanā, corresponding to v. 5cd of the SP). Some verses quoted by Abh. in ĪPVV I, without any reference to their source, also come from the SP. It is worth quoting them in full, also because they differ in some places from the form in which the same verses are quoted by Devasūri and Prabhācandra (the latter are in square brack- ets): bhāvabhedapratītyartham samyojyante 'bhidhāyakāḥ (v. 6b, cit. p. 195), paśyan- ekam adrṣṭasya darśane tadadarśane | apaśyan [Prabh. apaśyat] kāryam anveti vināpy ākhyātrbhir [Prabh. vinā vyākhyātrbhir] janaḥ || darśanādarśane muktvā kāryabuddher na sambhavāt [asambhavā] | kāryādiśrutir apy atra lāghavārtham niveśitā // (vv. 13-14, citt. p. 199), ity amiśrāḥ svayam bhāvās tān yojayati [miśrayati] kalpanā // tāṃ eva cānurundhānaih kriyākārakavācinah | bhāva- bhedapratītyartham samyojyante 'bhidhāyakāḥ // (vv. 5cd-6 citt. p. 200), saṃś ca sar- vanirāśamso bhāvah katham apeksate | apeksā yadi [parāpekṣā hi; Prabh. inverts the two hemistichs] sambandhah so 'san katham apeksate (v. 3, cit. p. 203), ... tān miśrayati kalpanā (v. 5cd, cit. p. 204), dvitve sa]i [Dev. dvisthe sa ca, Prabh. dvitve sa] katham bhavet (v. 2a, cit. p. 205), siddhe kā paratantratā (v. 1a, cit. p. 204).

It should be noted that the readings given by Devasūri and Prabhācandra are generally more in line with the Tibetan translations.

22 In Dharmakīrti's opinion the connection between cause and effect is not a relation (as conceived by the realists), since the essential requisite of a relation is the simultancous presence of the two related terms on which it rests equally - whereas cause and effect must necessarily belong to two different moments in time (cf. PV III.246 asatah prāg asāmarthyāt paścāc cānupayogatah | prāgbhāvah sarvahetūnām ...//) - (SP v. 7 kāryakāranabhāvo 'pi tayor asahabhāvatah | prasi- dhyati katham dviṣṭho 'dviṣthe sambandhatā katham //). Thus, the cause-effect relation is, in the final analysis, unreal precisely qua relation, being reduced to a concatenation of presences and absences (anyavya-vyārtireka, bhāva-abhāva); cf. SP v. 17 etāvanmātra[i.e. bhāvābhhāva]rattvārthāḥ kāryakāranagocarāḥ | vikalpā darśayan- ty arthān mithyārthā ghaṭitān iva //.

23 The fact that it rests on the two related elements is, as we have seen, the essential feature of a relation, as it is commonly understood by the realists: in the opinion of the Buddhist opponent this is a contradiction in terms, but precisely be- cause this is the essential requisite it cannot be abandoned without rejecting the very concept of relation (see SP v.11ab dviṣṭho hi kaścit sambandho nāto 'nyat tasyā laksanam).

24 Here, too, a direct reference is made to the SP the first ślokas of which are devoted to confuting one by one the various forms in which a relation may be represented. Utp. expressly mentions and criticizes only apeksā and pāratantrya; the kārikā, however, adds °ādi° and the ṭīkā shows that this is intended to refer to rūpaślesa, which in the SP is actually examined in v. 2, after pāratantrya and before apeksā.

– A relation is based on two terms (dviṣṭhah) and it is not logically tenable for it to rest on both and preserve its unitary nature. Neither is a relation conceivable in the form of a reciprocal requirement (anyonyapekṣāo) between two things which are·already accomplished25 nor in the form of a dependence (pāratantrya°) of two self-contained things26. On the grounds of what has been said, just as the state of cognizer is a mental construct, the same holds good for the agent27.

25 I.e. they already have their own independent existence. Primarily one must distinguish, as Utp. does here, between apeksā and pāratantrya, often generically translated as 'dependence'. In apeksā the sense of 'need, requirement, expectation' predominates: the shoot 'requires' the seed (SSVṛ p. 2 ankura eva bījam apeksate); the genititve rājñah 'requires' ('depends' on) puruṣah (apeksā, ākānksā ; see ĪPVV I pp. 203-204), without which it would remain incomplete. Dharmakīrti's criticism (SP v. 3 parāpeksā hi sambandhah so 'san katham apeksate / samś ca sarvanirāśamse bhāvah katham apeksate //) is taken up again in the ṭīkā and then in ĪPVV (I pp. 203-204). If a cause is existent (sat) it cannot 'require' anything else; if it is non-existent (asat) it cannot 'require' at all; nor can it be said that a thing 'requires' its effect in order to be qualified as a cause, because this is not anything other than its very nature, but is identified with it. If one objects that its nature, though being fully present, may not be realized as such in practical reality (vyavahāra), one may reply that, if the vyavahāra is cognition and verbalization (jñāna-abhidhāna), then it is the subject (i.e. the mind), not the thing, that is directly involved, and therefore he is given the responsibility for establishing such unreal relations (cf. SP v. 5cd ity amiśrāḥ svayaṃ bhāvās tāṃ miśrayati kalpanā). In the case, then, of rājñah (which is sat) with respect to puruṣah (also sat) there is no reciprocal apeksā, but only of the former towards the latter, (ĪPVV I p. 204 rājñah satah sati eva pumsi apekseti, na tu anyonyāpeksā atra vivakṣitā); therefore it cannot be represented as sambandha, which – we know – must be dviṣṭha.

26 The criticism of pāratantrya (more precisely 'dependence' as subordination, heteronomy) is outlined in the first verse of the SP (pāratantryaṃ hi sambandhasiddhe kā paratantratā). Abh. notes, following in the steps of the ṭīkā (IPVV I p. 204), that a reciprocal 'requiring' (anyonyapeksā) may well come within the sphere of pāratantrya; but this is still not enough to fulfil the prime requisite of sambandha, which is that it must be equally present in the two related terms (dviṣṭha). In effect, in the relation of pāratantrya anyonyāpekṣarūpa (glossed in the ṭīkā with parasparonmukhatā and anyonyam aunmukhya), like that of wife and sacrificer in the sacrifice or of two lovers making love, one cannot say that the 'expectation' (un-mukhatā) of the one coincides with that of the other. Thus unmukhatā does exist, but it is not a proof of sambandha.

27 It is the use of language (Dharmakīrti notes) – serving as a means of differentiating things – that suggest the existence of relations like that of factors and action. This relation has no correspondence in reality, but is merely the result of the synthesis of discursive thought (kalpanā), which, in fact, comes about through language.

I 2.11 Thus28 how is it possible to claim that the Self is the Lord of all?

• 11 -

guage (tām [kalpanām] eva cānurundhānaih kriyākārakavācinah / bhāvabhedapratītyartham samyojyante 'bhidhāyakāḥ). The valuation of kalpanā according to Dharmakīrti can obviously not be shared by Utp. (cf. SSVṛ p. 5).

28 In ĪPVV I p. 206 it is debated whether the two terms of the simile are on the same plane. In fact, from the previous arguments, in the case of jñātṛtva, jñāna is acknowledged and only the existence of a subject is denied (and should therefore exist, the possibility of a relation between the two is denied), whereas in the case of kartrtva even kriyā itself is denied. It is replied that the example tends to show their equality as regards being kalpita; and it is added that also jñāna is essentially kriyā - because of the verbal root it contains - and that therefore the confutation of kriyā also implies the confutation of that particular kriyā that is jñāna, which places them on the same plane.

98

CHAPTER III

1. Agreed.1 However, that form of cognition which is memory, though arising from the latent impression deposited by the former direct perception, is restricted to itself (ātmaniṣṭham) and does not know the original perception2.

• Memory, though arising from the reawakening of the latent impression deposited by the former perception, because it is restricted to itself exclusively knows only its own form. It cannot be claimed that memory determines the object formerly perceived, since it [memory] does not penetrate the former direct perception. - 1 -

2. A cognition is self-revealing (svābhāsaiva)3 and cannot be the object of another cognition, just as the cognition of taste is not known by that of shape. The fact that [memory] arises from latent impressions implies its similarity to the former perception, but not its cognition of that (tadgatih)4.

1 The author shows that he partly shares the view that the pūrvapakṣin puts forward regarding memory in the previous āhnika, especially with respect to the importance of the function of the saṃskāras. He departs from this view, however, when he examines their role, taking into account the impossibility (also recognized by the Buddhists) of one cognition becoming the object of another.

2 Every cognition - and memory is a form of cognition - is svasamvedana and it is in this sense that the expression ātmaniṣṭham, which generally refers to insentient realities (jada), must be understood. It is 'confined, restricted to itself' in the sense that it is only aware of itself and not of other cognitions (cf. ĪPVV I pp. 210-11).

3 The expression svābhāsa (cf. below p. 161 n. 2, p. 162 n. 4) may be interpreted in various ways and my translation of it has been left deliberately vague. The vrtti understands it as svasamvedanaikarūpa. Abh. glosses it in the ĪPV in two different ways (ābhāsah prakāśamānātā sā svam rūpam avyabhicāri yasyāh; svasya ca ābhāsanam rūpam yasyāh, which coincides with the vrtti); in the ĪPVV he even gives four different interpretations.

4 The saṃskāra of the former perception is reawakened by a present perception - similar to the other - which gives rise to the memory. The saṃskāra, therefore, ensures this 'similarity' in the memory, but the memory itself has no direct access

I 3.2 - Every cognitive act is by nature only aware of itself (svamvedanaikarūpā) and does not become the object of another cognition: if the cognition of shape could grasp the cognition of taste and vice versa, then the one would perceive the object of the other and in this way every restriction on the activity of the individual senses would cease to exist. Since memory arises from the latent impression left by the former perception it only bears a similarity to that perception but does not have direct cognition of the latter; and, moreover, as there is no cognition of this former perception (tadabhāvāt) not even the similarity to it can be maintained. - 2 -

3. [Some might object that]5 memory appears erroneously (bhrāntyā) as having as its object something directly perceived (dr̥ṭālambanatā), being reduced, instead, to the only determinative activity directed at them6 (tadavasāyatah), despite the fact that they have never directly entered its cognitive sphere (atadviṣayatve 'pi). But this objection, too, is inconsistent.

• And it cannot even be claimed7 that memory has these as its object only erroneously, in the sense that it makes the object of its own determinative activity (adhyavasyati) the former direct perception and its object, which in reality are not experienced, as happens when one states that one sees silver when faced with mother-of-pearl. - 3 -

4. How8 is it possible to reduce the true nature of memory to this? (smrtitai'va katham tāvad). And how is it possible that the establishment to (cannot 'know') the former perception and therefore nor can it, strictly speaking, establish the similarity between the latter and the present perception which has reawakened the samskāra. This, as will be shown later, is an operation that presupposes the 1. As one can gather from the vritti, and as Abh. (IPVV I p. 214) explains, tad in tadgatih has two different meanings; the first of which refers to anubhava and the second to sādṛśya; the first is to be considered the cause of the second: 'na tadgatih' atra dvāv arthau tacchabdena anubhavasyasyā ca parāmarśāt ! tatra prathamārtho dviṭīyasmin hetutvena mantavyah. IPV I pp. 129-30 defines the question with great clarity (and in this the Sanskrit language is an unrivalled instrument): saṃskārait param saviṣayāmātram smr̥teḥ siddham, natu anubhaviṣayatvam, nāpy asya viṣayasya pūrvānubhavaviṣayīkṛtatvam iti niścaya eṣaḥ.

5 This is once again a Buddhist speaking.

6 The former perception and its object.

7 As IPVV I p. 236 (cf. text note 130) explains na negates the whole sentence.

8 The author states the reasons that induced him to consider (in the last pāda of the previous kārikā) the opposing thesis untenable.

of objects (arthasthitih) should come about thanks to error9? And, if such I 3.4

is the case, what sense is there in claiming dependence on the latent im-pressions left by the former direct perception?

• Memory cannot be identified with something that derives only from determinative knowledge (adhyavasāyamātrāt), the object of the former perception being absent (tadviṣayasampramoṣe) because it [the former perception] is not made manifest [in the memory, according to your conception]. Neither can the definite establishment of an object formerly perceived be taken to be due to error. And moreover, [if it is a question of 'error'] why insist so much on the fact that memory - arises from the latent impressions, when, on the contrary, it is different from the former direct perception from every point of view (tadbhin-navogakṣemāyāḥ)10, not coming into contact with it in any way? - 4 -

5. If11 the determinative knowledge (avasāyasya) is error, how can it then, being insentient, establish objects? If, on the contrary, it is

9 The essential core of memory is the former perception recalled to mind. There is therefore no new cognition and it is only a new cognition that can be imputed for error (cf. ĪPVV I p. 239 and text note 136). The vrtti and, as can be gathered from the ĪPVV, also the fīkā, leave no doubt as to the meaning that Utp. attributes to the second pāda of the kārikā (bhrānteś cārthasthitih katham). It must be pointed out, however, that the interpretation of the ĪPV (I p. 132) is completely different. It essentially says that obviously no establisḥment of the nature of the object can derive from error, since the former is not manifest in error (if one understands error, as the opponent does, as asat° or ātmākhyāti), and that on the contrary everyone can see that the organization and functioning of the world of experience depend on memory. Therefore memory cannot be an illusion. This interpretation of the kārikā (misunderstood in Bh.'s commentary) is, in fact, what everyone would give on a first reading, and it anticipates the argumentation that Utp. actually presents further on.

10 In the expression yogakṣema (according to ĪPVV I p. 240) yoga (which has the general meaning of the attainment of something not formerly possessed) stands for the object's own nature characterized by light-knowledge (prakāśa), and kṣema (the preservation of what has been attained) refers to the part-object. Thus these features differentiate smṛti-bhrānti from anubhava, since in the former there is no new cognition (I read anapūrvātvat instead of apūrvatvāt) and the object is not really present.

11 The argumentation of this verse presupposes the opponent's objection, clarified in ĪPV I p. 133: the determination (adhyavasāya) of the former direct perception and its object, which is brought about by memory, establishes a partial similarity of the latter to the perception; and for this one needs to resort to the saṃskāras. The discourse, therefore, turns to the nature of this adhyavasāya (and

I 3.5 conceived of as being sentient, how can it, restricted as it is to itself and to its own ideation (nijollekhaniṣṭhāt), establish objective reality?

– It is the determinative knowledge (adhyavasāyah) alone that – erroneously – establishes objects, and not its self-awareness12 (svasamvit).

If such is the case, it is sentient and – being insentient – how can it cause the establishment of objects?

Thus, even if one were to acknowledge it was insentient, [this should be understood in a limited way, that is, in the sense that] it would only be able to make manifest objects pertaining to the past, or illuminate only non-external objects – itself and its own ideation (abāhyasvātmollekhamātraprakāśah) – and, therefore, it could not equally be accepted as the cause of that establishment of objects we are dealing with13. – 5 –

6. Thus, the functioning of the human world – which stems precisely from the unification (anusamdhānā°) of cognitions, in themselves separate from one another and incapable of knowing one another – would be destroyed…

this is what the present kārikā is about) and as ĪPVV I p. 243 underlines (adhyavasāyamātrasya rūpam dūṣitam, taddvārena tu smṛtyadhyavasāyasyeti) through the confutation of this conception of the adhyavasāya in general, the form in which it operates in the ambit of memory is also confuted.

If the object is illuminated by it, then one cannot define it error; if it is not illuminated, then this means that memory does not have access to the object and therefore it does not make sense to speak of 'similarity'.

12 The adhyavasāya is on the one hand insentient (as regards its actual not coming into contact with the object, its being bhrānti), and on the other sentient as regards its self-awareness.

But only the first aspect is significant in this case.

13 This point is made clear in a passage from the ĪPVV (I p. 243), which due to its importance and clarity I have quoted in full (see text note 141).

The adhyavasāya is insentient insofar as it does not illuminate the external object (otherwise it would be no different from direct experience) and neither can it be said that it does not illuminate at all (otherwise it would simply be jada), nonetheless, it is acknowledged that at least it illuminates itself and its own ideation.

But even this is not sufficient for it to cause the determination of external objects, because precisely as regards the latter (and it is this that counts here) it is insentient.

All this criticism of the adhyavasāya aims at reducing its significance in the phenomenon of memory, with the ultimate purpose of making way for the admission of another principle that may consistently explain the point, i.e. the self, as we shall see later.

• Cognitions are restricted to themselves only (svātmamātraparini- I 3.6 sthitāni) and cannot be the object of other cognitions (parasamvedyāni) being by nature [exclusively] conscious of themselves. But then how would the dimension of human activity and behaviour (lokavyavahārah) - culminating in the teaching of the absolute reality - be possible, since this consists precisely in the interconnection between the objects of knowledge? (anyonyaviṣayasamghaṭṭanāmayah)14. - 6 -

7. ...15 if there were no Maheśvara who contains within himself all the infinite forms, who is one, whose essence is consciousness, possessing the powers of knowledge, memory and exclusion16.

• The mutual unification of all cognitions of things is [constituted by] the consciousness principle (cittattvam) whose form is all, since nothing distinct from it is admissible. The powers of knowledge etc. only pertain to this consciousness principle. It has been said17: «From me derive memory, knowledge, exclusion». - 7 -

14 See Torella 1988: 160-66.

15 It continues from the preceding kārikā.

16 Cf. ĪPVV I p. 276. The implication of the opponents' theses is recognized as untenable (naśyej janashitih) and it is this untenability that leads to the confirmation of what they sought to deny (prasaṅgaviparyaya), i.e. the Self as the subject of knowledge and action, unifier and coordinator of the discontinuity of reality. Consciousness is the unification of cognitions both in the sense of 'becoming one' (ekībhavana) and of 'making one' (ekīkaraṇa), 'merging together' (miśrīkaraṇa). In the vrtti anusamdhāna and cittattvam are in a relation of sāmānādhikaranya since action and the subject of the action are in reality the same (yuktam eva abhedena upacāranam kriyākartroḥ paramārthata aikyāt). See Torella 1988: 166-67.

17 Bhagavadgītā XV, 15.

I 4.1

Chapter IV

1. The Free One (svairī), the perceiver of the object formerly perceived, continuing also to exist later, has the reflective awareness: 'that'. This is what is called 'remembering'.

• The light-perception of the object formerly perceived is not extinguished [at the moment of memory1], since he [the knowing subject, the Lord] also continues to exist later as the perceiver of the object formerly perceived. Having here (atra)2 a reflective awareness of an object in terms of 'that' - as formerly perceived - on the part of he who is one, the Lord (vibhuh), agent, is the function called 'memory'. - 1 -

2. [He who remembers] must necessarily, having a reflective awareness of (āmrśan) the particular entity (svalakṣaṇam)3 formerly manifest, make it manifest4 at the actual moment of the memory

1 Cf. the classical definition of memory in Yogasūtra I. 11 anubhūtaviṣayāsampramoṣah smṛtih.

2 I.e. in the present limited condition, at least according to Abh.'s interpretation which is, in turn, based on the ṭīkā. The subject of memory is thus Śiva himself as had already been suggested in the previous kārikā. This is underlined in ĪPVV II p. 11 sūtravṛttyor bhagavatah smartrtvam uktam, tadanusāreṇa ṭīkāyām api and p. 13, where it is excluded that vibhu may be simply taken to mean 'pervader', vidham kṛtvā bhavati vibhutvam, na tu iha vyāpakatvam... The phenomenon of memory, though having as its ultimate subject Śiva, the Supreme Consciousness, must necessarily occur on the individual plane, otherwise it would not make sense to speak of a before and after of a perceived object, etc... And this is what the vṛtti refers to with atra, which is glossed in ĪPVV (see text note 157) «... function of the Lord contracted in the breath of life, the eightfold body, the physical body etc.» and at greater length ibid, p. 11, where the expression samāropitasvātantrya, which glosses atra in the ṭīkā, is cited and explained.

3 The object of memory is the svalakṣaṇa, or the individual entity in space and time. In memory, in fact, there is necessarily an association with a former time and place, otherwise there would be no memory but a general vikalpa (cf. ĪPVV II p. 19 smratau hi prāgdeśakālayogo 'vaśyaṃ sphurati / anyathā vikalpamātram tat bhavet).

4 Without prakāśa, in fact, there would be no pramāṛśa, but - as ĪPVV II p. 18 remarks - the latter should also immediately follow prakāśa, being inseparable

104

(svakāle), either as a single manifestation 'jar' or as the totality of its I 4.2

components (akhilātmanā).

• Thanks to the power of memory (smṛtiśaktyā) the subject, when

having a reflective awareness (parāmarśan) as 'that' of the particular en-

tity formerly perceived, must make it manifest [in the present]

(ābhāsayaty eva)5, for there could be no reflective awareness of an ob-

ject made manifest [only in the past] (prākāsitāsyā)6; and this manifesta-

tion occurs at the very moment of the act of remembering (svasattākāla

eva). Therefore it is not erroneous to state that there is the manifestation,

at the moment of the memory, of an object, which, however, is no longer

present at that moment. At certain times the object appears in the form

of a single manifestation7, i.e. limited to one of the many manifestations

that constitute it, such as - in the case of a jar - 'jar', 'golden', 'individual

substance', 'existence' etc., depending on the subject's intentions8. In these

cases its manifestation in memory is distinct and vivid (sphuta°)9. At

from it (parāmarśanam ca prakāśanāntarīyakam). The question this kārikā is meant

to answer is, precisely how to bridge the gap between the prakāśa of the object -

which occurred in the past, at the time of the former perception - and its parāmarśa,

being effected now by memory.

5

In the sense that making the object manifest in the present is the necessary

condition for the memory to be such. ābhāsayaty eva glosses ābhāsayet; therefore

this optative is to be understood in the sense of niyoga (cf. ĪPV I p. 161).

6

This interpretation is supported by Abh.'s argument (see text note 160). Cfr. n. 4.

7

The single ābhāsa in the memory is only single relatively speaking, i.e. it is

not combined with other ābhāsas such as 'golden' etc. In actual fact, due to the very

nature of memory, the connection with the ābhāsa 'time' is inevitable, and this is

sufficient in itself to make it svalakṣaṇa and in this sense sphuṭa (cf. ĪPV I p. 159

ābhāsāntaravyāmiśraṇābhāve 'pi tu kālābhāsasambhedenaiva svalakṣaṇyam tāsya

ābhāsasya karoti kālaśakter eva bhedakatvāt iti vaksyate / evam tāvat svalakṣaṇībhāvah

prāktanadehābhāsasācivyād yuditaikālābhāsayojanayā ghatābhāsasya iti).

8

This theme is extensively developed later (II 2. 5-7).

9

The readings of the MSS oscillate between sphuṭa and asphuṭa and, at first

sight, there seem to be arguments in defence of both. In fact, in the ṭīkā the

manifestation of an object in the form of sāmānya is said to be asphuṭa. But this is

so as regards the so-called independent vikalpas, whereas we have seen that for

memory the case is different. The ṭīkā goes on to say that in general many sāmānyas,

in combining with one another, mutually delimit each other and thus may be said

to become sphuṭa, and, once they have been further delimited by the ābhāsas of

space etc., they finally give rise to the svalakṣaṇas. It is true that the case in ques-

tion concerns a single ābhāsa, but its status in memory is a particular one, as has

I 4.2 other times, on the contrary (anyadā tu), the object appears in its totality, since this is the subject's intention: its manifestation is equally distinct and vivid (tathāiva), as in the previous case10. And finally, the subject whose mind is intensely concentrated without interruption even directly visualizes the object formerly perceived (drṣṭārtha°)11. - 2 -

3. It would not be possible to speak of the manifestation of the object being remembered if it appeared as separate from the memory12;

just been said. Moreover, the tīkā adds, the sāmānya even when taken singly may be sphuṭa (compared to others), when it implicitly contains other broader and more generic sāmānyas (e.g. the sāmānya of the dhava tree, compared to the sāmānya of tree in general). See Torella 1988: 168-70. The concept of a hierarchy of sāmānyas as regards their extension probably stems from Vaiśeṣika and grammatical speculation.

10 In ĪPV I p. 160 the degree of sphuṭatva in this case (where there is a combination of different ābhāsas) is considered greater than the former: i.e. it is atisphuṭa.

11 It is the case that ibid. p. 160 is defined atyantasphuṭa. ĪPVV II p. 29, following the tīkā, considers it implicit in the expression akhilātmanā in the kārikā, that would therefore be understood as referring in the broad sense to a multiplicity of ābhāsas (bahvābhāsopalakṣanam) and with a further differentiation within it (samagratva and asamagratva). Thus the object would appear in the memory as a combination of many ābhāsas: when the latter do not cover the whole (asamagratva) of the object this is the intermediate level (introduced in the vrtti by anyadā tu); when, on the other hand, the object is recalled in the totality of its aspects, the memory is so vivid that it closely borders on the direct perception from which it differs only in the fact that its object is a thing 'already' seen. The expression drṣṭārtha distinguishes this from other forms in which the object visualized had not been previously seen but experienced in various ways through the revelation of the scriptures, as in the case of the visualization of deities etc.

12 In other words, if the object appeared as external to and distinct from the memory (ĪPV I p. 163 bhinnatvena bahīrūpatayā), then the requirements for direct perception and not for memory would exist. The object is present in the memory only through the manifestation of the former perception, but it is only in it, i.e. in the past (tadā), that the object bahir avabhāsate. What is manifested in the present is only the reflective awareness (vimarśana). See Abh.'s lucid formulation of the question in ĪPVV, which may serve as a general definition of memory according to the Pratyabhijñā: pūrvadrṣṭataiva savimarśā satī idānīntanavimarśabhūmipatitā sa iti vimarśa ucyate (II p. 30). This stanza actually shows how the apparent inconsistency between a (present) vimarśana and a (past) anubhava is only resolved by the link which ensures the possibility of unifying the various cognitions occurring at different times. A further clarification is provided by ĪPVV II p. 32. The prakāśa concerning the part-object (arthāmśe) in the former perception belongs to the past; but the prakāśa as grasped by the vimarśa, concerning the part-self (svātmāmśe), is not

therefore the unity of cognitions that occur at different times is necessary, and this unity is precisely the knowing subject.

• And the object formerly perceived, manifesting itself together with the perception in the present 'light' of the memory (tātkālikasmrtiprakāśe) (directed at that past time), is necessarily not separate from the memory, since that which is separate from the light cannot shine. There is therefore a unity of the distinct cognitions such as direct perception, memory etc. and this unity is constituted by the self, the knowing subject. To elaborate: - 3 -

4. In fact, in memory the former perception is not manifested separately - like the object - since it appears as resting on the self, as the expression 'I perceived in the past' indicates.

• In memory the former perception - unlike what happens to the perceived object that is remembered13 - is not manifested as separate (differentiated), since it is the self itself that is manifested - the object of the notion 'I' - whose essence is informed by this perception. And it is precisely that reality present at many different times, known as 'I', that is the self.

limited by time (kālānavacchinna eva). Thus the vimarśa in the memory can connect itself with the vimarśa in the perception and, through it, with the former light of the object - in this way meeting both requirements, that is, taking place in the present and not being divorced from prakāśa (tannantarīyakavimarśasahitah punar vimarśas tu idānīntanah).

13 In the perception the object appeared - due to the power of māyā - as if it were separate, external to consciousness. This may not happen in the memory for otherwise it would not be memory (ĪPVV II p. 34 smaranatīva na bhavet); in fact, the light of memory embraces, so-to-speak, the object much more strongly (ibid. artham atītamām kroḍīkaroti). However, this is true only to a certain extent: the assumption of externality - in the form of residual traces, as a shadow (ibid. chāyāmātrena) - continues to exist (p. 32 prthagbhāvaprakāśābhimāno 'nuvartate), remaining in the object remembered in the background. In underlining this, Utp. also implicitly draws a sharp distinction between the 'external' thing, on the one hand, and perception, on the other: unlike the thing, the latter can never be objectified (see next kārikā). Thus those who think they can do without the I as the unifier of different cognitions, by claiming that memory has access to its object simply by embodying (i.e. objectifying) the former perception of it, are mistaken. In ĪPV I p. 167 Abh. gives an interpretation of the example concerning the status of the object (arthavat in the kārikā) both as sādharmya° and vaidharmyadrṣṭānta ; neither of them reflects the interpretation found in the vrtti.

I 4.5

5. The cognitions [of others] are not manifested [as an object] even in that particular cognition belonging to the yogins. The only means of knowing them is self-awareness. What has just been stated also applies if they are acknowledged as an object of cognition (meyapade 'pi vā).

– For the omniscients, too, the cognitions belonging to other subjects - which can be known only through self-awareness14 - must always be manifested as resting on their own self (svātmārūdhā eva); and, therefore, what in reality occurs (tattvam) is the identification on the part of the yogins with the self of others15 (parāmatmāpattir eva). Should one attribute [to the cognitions of others] the nature of cognizable object (prameyakakṣyāyām api) such as the jar etc., they would have to be manifested each in its own form of pure awareness (śuddhabodhātmanā rūpeṇa), whereas that is impossible16. - 5 -

14 I.e. that introspective awareness that one has of every cognition or that every cognition has of itself. This position had also been accepted by Bhartrhari; see the last four verses of the Jātisamuddeśa, in particular v. 106 yathā jyotih prakāśena nānyenābhiprakāśyate / jñānākāras tathānyena na jñānenopagrhyate (Abh. quotes this verse in another section of the ĪPVV giving a slightly different reading of the second hemistich: jñānarūpaṃ tathā jñāne nānyatrābhiprakāśyate) 'Just as light is not illuminated by another light, in the same way, the form of cognition is not determined by another cognition' (transl. Iyer 1971: 62) and v. 110 yato visarūpeṇa jñānarūpam na grhyate / artharūpaviviktam ca svarūpam nāvadhāryate // 'As the form of the knowledge is not cognized as an object, its form as distinct from that of the object is not grasped' (transl. Iyer 1971: 63). See also Helarāja's lucid commentary. For a penetrating analysis of the various theories about 'knowledge of knowledge' I refer to Matilal 1986:141-179.

15 This is the most obvious meaning of the text of the vrtti. According to Abh.'s interpretation (ĪPVV II p. 47, see text note 189), who in turn takes this from the țīkā, tattvam is 'the being such of the omniscient' (mentioned at the beginning of the vrtti), meaning the condition of subject identified with the Supreme I (parātmavedakatvam); thus the meaning of the sentence would be the following: 'the condition of subject identified with the Supreme I, pertaining to these yogins, means in point of fact the attainment of identification with the self of others.' On the yogin's cognition of other minds see Dharmakīrti's Santānātarasiddhi vv. 89-93 (Stcherbatsky 1969: 89-92, Kitagawa 1955: 108-10).

16 The nature of every cognition is in the final analysis śuddhabodha and it should appear in this form to the yogin's cognition: but precisely this is impossible, because descending to the condition of vedya necessarily entails the flawing of śud-dhatva (see ĪPVV II p. 52 svasmin svasmin darśane yādṛśaṃ bodhasya śuddhaṃ rūpaṃ, ... na tādrśaṃ niruparāgaṃ vedyatām paricchināne 'bhyeti; cf. also ĪPV I p. 183). A similar remark is made by Helarāja, commenting on VP III.1.109: cf.

108

6. The memory17 expressed as 'thus that cognition occurred in me' is I 4.6 none other than an analysis (vyākaraṇam) in terms of differentiation of the memory expressed as '[that thing] was seen by me'18.

• The memory presented as 'thus that perception occurred in me', in which the cognition is mentioned separately, is nothing more than the analytic exposition of the memory understood as '[that thing] was seen by me', which is characterized precisely by having a reflective awareness of the visual perception as residing in the knowing subject. - 6 -

7. Determinate cognition (avasā) too, in the form of 'I see this' or 'this is a jar' regards (manyate) the visual perception as being indissolubly inherent in the subject of the determinate cognition (avasātari)19.

• Also in the determinate cognition of the object in front of one - whether it appears in the form 'I see this jar' or 'this is a jar' - the visual

Prakīrṇakaprakāśa I pp. 103-4 ghatājñānagatam iti tu jñānam na jñānarūpānupāti, yato ghatajñānagatam, viṣayam nirūpya jñānarūpam kimapi tad ity etāvad yadi param nirūpayitum śaknoti, na tu sā tadīyā bodharūpatā tasyā pratibhāsate parasamvedana iva «But the cognition '[This is] the cognition of a jar' does not come after the form of the cognition ['This is a jar'], for, having brought to the intellect the object in the cognition of the jar, it can at the most bring to the intellect 'that is some form of cognition'; the awareness character [bodharūpatā] of that [cognition, namely 'This is a jar'] is not reflected in it as in the case of somebody else's cognition.» (transl. Aklujkar 1970: 179).

17 This kārikā aims at confuting anyone who might still wish to claim, in presenting the memory as 'I remember having had this perception', that the perception appears as the object in the memory (cf. ĪPV I p. 177).

18 Abh. (ĪPV I p. 182) also points out a different interpretation of the kārikā, supported by 'others' (anye tu), but he keeps to that of the vrtti.

19 This kārikā is in reply to the hypothetical objector who might seek to invalidate the principle of the non-objectifiability of cognition (applied here to memory in order to show the need for the I as the fundamental reality), by claiming that the determinate cognition (adhyavasāya, vikalpa) that follows the undifferentiated perception (nirvikalpa) knows the latter as object (cf. ĪPV I p. 186, ĪPVV II p. 54). It must be remembered that the adhyavasāya plays an essential role in memory, because memory is not linked directly to the anubhava itself, but to the anubhava as filtered and determined by the adhyavasāya; indeed there is no memory of an anubhava not followed by adhyavasāya (ĪPVV II p. 54 avikalpite smaraṇāsambhavāt). The criticism levelled at the adhyavasāya therefore constitutes both an analogical and a direct criticism of memory.

I 4.7 perception. one becomes aware of (pratyavamrśyate) has the knowing subject as its constitutive reality20. - 7 -

8. Therefore, when there is the reflective awareness 'that is seen by me', 'this', 'that', the two elements though divided into perceiving subject and perceived object are manifested within the [true] cognizer (pramātari).

• Therefore in memory or in generic mental elaboration (vikalpamātre)21 - which both presuppose the reflective awareness of a former perception - only the 'object' part in the form of 'this' or 'that' may be expressly mentioned, even without being intermixed with the word 'perception' (drkśabdanuvedham vināpi). Nonetheless, in all cases, the perceiving subject and the perceived object, though22 they appear to be separate due to the influence of māyā, are manifested as absorbed within the one cognizer. - 8 -

20 The 'I' that is the subject of the determinate cognition is that same 'I' that constitutes the centre of the anubhava, even if the latter may be such that the I does not appear in it ('this is a jar' as opposed to 'I see this jar'). But the absence of the I is only apparent and does not concern the anubhava in itself but a particular aspect of the following adhyavasāya, which, depending on the circumstances, may stress the objective (i.e. when one is intent on bringing about some practical activity connected with it) or the subjective element (cf. ĪPV I pp. 187-8).

21 I.e. in the determinate cognition (adhyavasāya) that follows the direct perception. In the construction I have followed Abh.'s indications (see text note 202).

22 Utp., unlike Abh. (see text note 207), assigns api a different place from the one it occupies in the kārikā. The separateness of the perceiver and the perceived - Abh. (says (ĪPVV II p. 58) developing Utp.'s remark in the tīkā - is only apparent as the very use of the dual form shows. On the basis of the sahavivaksā principle, in grāhyagrāhakatābhinnāv arthau each term refers both to itself and the other; they are interdependent (anyonyāpekṣau). For the analysis of a similar use of this grammatical argument see Torella 1987:154-157.

Chapter V

1. The objects that are manifested in the present can be manifested as external only if they reside within.

• Even in direct perception1, however2, the manifestation of objects as separate3 is admissible only if they are absorbed in the cognizer (pramātrantarlīnānām)4. - 1-

2. If it were not essentially light, the object would remain non-light as before5; and the light is not differentiated [from the object]: being light constitutes the very essence of the object.

1 In this the manifestation of the object is present, unlike in memory etc., in which the direct presence of the object has vanished.

2 The reference is to an implicit objection to which this kārikā is intended as a reply (cf. ĪPV I p. 195, ĪPVV II pp. 67-8). If it has been said that in direct perception the object appears as external, differentiated from the knowing subject, and that in memory (and also in vikāpa etc.) it appears in union with it, how is it possible to claim the dependence of the latter on the former? Indeed, this kārikā establishes the sense in which the 'externality' of the object in direct perception is to be understood.

3 Separate from the māyic cognizer, identified with the mind, the body etc.

4 The only true cognizer in the absolute sense (paramārthapramātr, ĪPV I p. 197), i.e. the Supreme Consciousness.

5 An object that is external to consciousness does not exist at all, since it would be jada and what is jada cannot shine in knowledge; cf. ŚD IV.29-31, V.12 etc. This principle, which often recurs in Vijñānavāda texts, has a lucid formulation in a verse of Prajñākaragupta's Pramāṇavārttikālamkāra (quoted in TBh p. 35) yadi samvedyate nīlam katham bāhyam tad ucyate / na cet samvedyate nīlam katham bāhyam tad ucyate «If blue is perceived, then how can it be called 'external'? And, if it is not perceived, how can it be called 'external'?»; see also TS 1998 ff., etc. Furthermore, the Vijñānavādin says, blue and cognition of blue are never perceived separately (sahopalambha); therefore, they are non-different. Cf. the well-known passage from the PVin (I.55ab; Steinkellner 1972: 206) sahopalam-bhaniyamād abhedo nilataddhiyoh, also quoted by Abh. (ĪPVV II p. 78). For a comprehensive treatment of the inference based on sahopalam-bhaniyama, see Iwata 1991. Among the many examples in monistic śaiva texts cf. the verse from the lost Ucchuṣmabhairava (cit. ŚSV p. 4) yāvan na vedakā ete tāvad vedyāḥ katham priye

I 5.2

– If it did not have light, i.e. the cognizer, as its own nature, the jar, just as it was not manifest [to the cognizer] at first, would continue not to be so even at the moment of cognition. What constitutes the object's being manifest (prakāśamānatā) is light insofar as it is its own form (svarūpabhūtaḥ) and not a separate reality6. – 2 –

3. If light were undifferentiated [in itself] and differentiated [from objects], then objective reality would be confused. The object that is illuminated must itself be light; that which is not light cannot be established.

– Light, conceived as absolutely undifferentiated [as regards itself] (prakāśamātram)7 and as differentiated from the object would be common to all objects8. In this case the delimitation of objects ('this is the light-cognition of a jar; this other, of a piece of material') would have no plausible grounds. Therefore, the establishment (siddhiḥ) of the object depends on its essence being light (prakāśātmatāyattā). – 3 –

4. [Objection]9 Since consciousness-light (bodhasya) being undifferentiated cannot be the cause of a multiform manifestation, all this

/ vedakam vedyam ekam tu tattvam nāsty aśucis tataḥ //; See also SPr I.12 tvadātmakatvam bhāvānām vivadanti na kecana / yat prakāśyaśāyāto nāprakāśaḥ prakāśate //; etc.

6 In formulating the Śaiva position, Utp. is implicitly critical particularly of the Buddhist and Kumārila theses. According to the Buddhists (kṣanikavādin) the manifestation (perception) of the object is understood as a 'luminous' moment produced in a chain of moments, e.g. of a jar – a luminous moment determined by a concurrence of causes, such as the operation of the sense of sight, light etc. For Kumārila, on the contrary, an added quality, 'being manifest' (prakatatā) occurs in the object, from whose presence a former cognitive act is inferred. In both cases the light is not separate – as Utp. also maintains – but it is found, so to speak, completely resolved within the object (IPVV II p. 69 arthasarīramagnaḥ) and one is unable to understand how a thing can be cognized only by some and not by others. According to Utp. there is an illumination of the thing (whose essence is light) by the light of the subject.

7 Cf. text note 216.

8 I.e. it would illuminate all objects indiscriminately.

9 The preceding argumentations may lead to a Vijñānavādin-like conception. In order to undermine this, Utp. temporarily opposes it with the Sautrāntika criticism (which he agrees with up to a point, at least with the motives for it) and in this context presents their doctrine known, at least in later sources, as anumeyabāhyārtha-

112

various manifestation lacking in an apparent cause (ākasmika°) leads to I 5.4

the inference of an external object [as its only possible cause].

• The establishment of the being of insentient realities is merely their becoming manifest (ābhāsamānataiva), and this is precisely having manifestation as their essence (ābhāsātmataiva). Their true reality, therefore, lies in the one pure consciousness. The gradually changing manifestation of objects which occurs though there is no diversity in the latter induces one to hypothesize an external object, different from consciousness (tato 'nyam), not shining (aprakāśamānam), as the cause. This reasoning is analogous to that which establishes the existence of the senses10. - 4 -

5. Not even a varied reawakening of the karmic residual traces can be taken to be the cause [of the multiform manifestations], for in that

vāda (Sarvasiddhāntasamgraha V.2ab sautrāntikamate nityo [°am?] bāhyārthas tv anumīyate, 7ab viṣayatvam hi hetutvam jñānākārārpanakṣami; TBh p. 36 arthasya svarūpeṇa nāsti vedanām 'bhāktam svād arthavedanam' iti vacanāt; SDS p. 94 anumeyam satyam ity āsthitāin vijñeyam anumeyam iti). The bāhyārthānumeyavāda as upheld by Badhanta Śubhagupta is mentioned and criticised in TS vv. 2050-55. If consciousness is abhinna, as the Vijñānavādins claim, and no other cause can convincingly explain the emergence from it of the various differentiated images that constitute phenomenal reality, all that remains is to hypothesize the existence of external objects as the source (bimba) of the reflection (pratibimba) present in consciousness. The existence of the external object is strictly speaking only inferable (nityānumeyah; on the proof by the method of vyatireka see TBh p. 35), not directly perceivable, as it is in Vaibhāsika conception: in cognition only ākāras figure.

Moksākaragupta summarizes the various positions as regards the reality of external object as follows (TBh p. 36): ye 'pi tadārambhakāḥ paramānavo vaiśeṣikānām, sākṣādadhyakṣagocarā vaibhāṣikānām darśane, svākārasamarpanapravanāḥ sautrāntikānām mate 'pi yogācārānām darśane na sambhavanti; see also SDS p. 35 te ca mādhāmika-yogācāra-sautrāntika-vaibhāṣikasamjñābhiḥ prasiddhā buddhāḥ yathākramam sarvaśūnyatva-bāhyārthaśūnyatva-bāhyārthānumeyatva-bāhyārthapratyakṣatva-vādān ātiṣṭhante. (In Sautrāntika epistemology this is to be understood in the sense that the thing, the absolute particular that is the object and cause of the direct perception, is (imperfectly) cognizable only through the filtered image of discursive thought - anumāna in the broadest sense -: if, on the one hand, there is absolute otherness between the thing and its mental image - which is sāmānya in nature - there is, nonetheless, an undeniable coordination, too: the one is the cause of the other).

Sautrāntika criticism does not succeed in undermining the Vijñānavādin position, but serves to show the need for a more coherent and comprehensive conception, which is represented precisely by the definitive conclusion (siddhānta) of the Śaiva doctrine.

10. Indeed we do not perceive the senses directly but we infer their existence from the existence of cognition, which only occurs through them (cf. below I. 5. 8.).

I 5.5 case a new question would arise: what is the cause of the variety of such a reawakening?

– The varied reawakening of the karmic residual traces does not occur outside consciousness; also what might be the cause of this reawakening is not seen. Consequently the manifestations cannot be anything but the external object11. – 5 –

6. That may be (syād etad)12. [But] seeing that ordinary worldly activity can be accomplished on the basis of such 'manifestations' alone, what sense is there in wanting to resort to an external reality [than consciousness], which is not supported by reason?

– Ordinary human activity takes place through objects precisely insofar as they are manifested (ābhāsamānair evārthaih). So what harm could there be in considering them essentially 'manifestations' (ābhāsa°) and nothing more13? In which case there is no longer any need to resort to an external object, since they alone are sufficient to ensure the complete functioning of practical reality (tāvatā lokayātrāsamāpteh). Furthermore14, the external object is contradicted by the criteria of right

11 The argumentation of those who claim the existence of external objects continues with the reply to the expected Vijñānavādin objection. A detailed account of the arguments set forth by the two opponents (Vijñanavādin's vāsanāvaicitrya and Sautrāntika's pratyayavaicitrya) is to be found in SDS pp. 80–84; Vācaspatimiśra also refers to them in the Bhāmatī and Tātparyadīpikā (La Vallée Poussin 1901: 190, n. 114).

12 At this point the author takes up the discussion in the first person (but the same thesis with the same arguments might also be put forward by a Vijñānavādin), after having voiced the opposing theses of the bāhyārthavādins. Abh. sees two possible interpretations of the expression syād etad: the partial acknowledgement of the opposing theses and then the exposition of his own, introduced by a kimtu that is to be understood; or the simple addition to the adversary's thesis of another even more convincing one, his own (the vrtti gives the latter interpretation). But, as can be seen, the general meaning remains the same.

13 It is understood by this that the external objects themselves 'inferred' precisely as such – i.e. illuminated, brought to consciousness through inference – are inseparable from the light of consciousness which is their essence; for if they were outside it, there would be no knowledge of them (cf. ĪP I. 5. 9.). This implication is taken up and developed by Abh. (ĪPV I, p. 222).

14 After the fundamental negative reason given in the previous lines, additional reasons (abhyuccayabādhaka) are put forward. Utp.'s remarks are the same as those a Vijñānavādin might make; cf. Viṁśatikā 11–15 and vrtti, Sthiramati's bhāṣya on

114

cognition (pramāṇabādhitah): this is so if it is considered as having parts, I 5.6 since this would result in attributing to it contrasting qualities etc.; if it is considered as devoid of parts it is still contradicted in various ways (bahuśah), because it is simultaneously in contact with the six directions of space, etc.15 - 6 -

Trimśikā 1, Ālambanaparīkṣā 1-5 etc. Particularly exemplary is the argumentation in Vimśatikā, often taken up again by later authors (cf. Mokṣākaragupta's TBh, SDS etc.; see also Lévi 1932: p. 52 note 2) and, implicitly, also referred to by Utp. Vimśatikā 11 reads: na tad ekam na cānekam viṣayah paramānuśah / na ca te samhatā yasmāt paramānur na sidhyati // «The object is not one; it is not multiple as composed of many atoms; and it does not even consist in these atoms agglomerated, since the atom is not established.» The first thesis belongs to the Vaiśeṣikas who consider the object an individual substance (dravya), a totality (avayavin) formed by a complex of parts. The Buddhist tradition is unanimous in rejecting the avayavin as a mere mental construct (see Milindapañha pp.26-28 (PTS Ed.), Samyuttanikāya I p.135 (PTS Ed.), TS 555-620 and Pañjikā, Paṇḍita Aśoka's Avayavinirakaraṇa etc.). The second thesis may also be ascribed to the Vaiśeṣikas who consider the atoms, imperceptible if taken singly, as being the parts that constitute the avayavin (cf. TBh p. 36 ye 'pi tadāmbhakāḥ paramānavo vaiśeṣi-kānām ...). The third thesis is upheld by the Vaibhāṣikas (see, for example, AK I.35d sañcitā daśa rūpinaḥ - Bhāṣya: pañcendriyadhātavaḥ, pañcaviṣayāḥ sañcitāḥ paramānusaṅghātatvāt) and, in a modified form, by the Sautrāntikas; the Sautrāntika criticism of the Vaibhāsika conception is expressed in various places in the AKBh. The Vijñānavāda rejects the theory of the agglomeration of atoms, however formulated (sañcaya, saṅghāta, samudāya etc.; see bhāṣya on Trimśikā 1, Vimśatikā 12-15 and vrtti, Madhyāntavibhāgaṭīkā p.21 (Stcherbatsky 19362: 93-95). One of the Abhidharma passages which lends support to this thesis is sañcitālambanāḥ pañca vijñānakāyāḥ. This is also quoted by Dignāga (Vrtti on PS I.4ab; see Hattori 1968: 88-89) who endeavours to show how it is possible for direct perception to have a group of entities as object while remaining avikalpaka. The same passage is quoted by Manorathanandin in his commentary on PV III.194, where Dharmakīrti's defence of Dignāga's controversial statement begins. According to Dharmakīrti, arthāntarā-bhisambandhāj jāyante ye 'nvo 'pare / uktās te sañcitās te hi nimittam jñānajan-manah // «Those other atoms which come into being in close relation [Manorathanandin glosses: 'in close proximity'] to other things [i.e. atoms] are called 'agglomerated'. They are the cause of the arising of knowledge». In this way Dharmakīrti tries to show that the agglomeration is not a different thing from the atoms themselves. This position, too, expressed from a Sautrāntika standpoint, is, however, altogether unacceptable to a Vijñānavādin like Sthiramati, who concludes his criticism (Bhāṣya on Trimśikā 1) by pointing out the internal contradiction: na hi asañcitāvastātah sañcitāvastāyāṃ paramāṇūnāṃ kaścid ātmātiśayah / tasmād asañcitavat sañcitā api paramānavo naivālambanam.

15 A critical allusion to the Vaibhāṣikas's anusamcayavāda. If the ultimate reality, the atom (paramāṇu), has no parts, it is not understood how this can aggregate to

I 5.7 7. Indeed, the Conscious Being, God, like the yogin, independently of material causes, in virtue of His volition alone, renders externally manifest the multitude of objects that reside within Him16.

• It is therefore the consciousness-principle alone that, in his sovereignty, renders manifest as external the multitude of things such as the jar, cloth etc., which are spontaneously manifested as being informed by his very self17. Since his power is infinite, this external manifestation of objective reality takes place through his volition alone without resorting to causes such as clay etc. - 7 -

8. Neither can one speak of inference if the thing that is the object of this has not been formerly directly perceived (anābhātapūrve). This is also the case as regards the senses, whose direct perception may be said to have occurred through the perception of a reality that has the characteristics of a cause (hetuvastunah), such as the seed etc.18

• Inference is an act of reflected thought (vimarśanam) which has as its object a thing that was formerly manifest to consciousness and [as such] is internal to it (purvabhātāntaḥsthita eva); it occurs on the basis of the direct perception of an object invariably concomitant with this entity and involves the linking [of this entity] with this or that determinate

others, aggregation only being possible through a contact between parts which by definition it does not possess (see e.g. Vimśatikā 11 ff. and vrtti). See also Dignāga's Ālambanaparīkṣā vv. 1-5. In the text bahuśah might equally well refer to the previous pramānabāditah.

16 Cf. ŚD I. 44-45ab yoginām icchayā yadvan nārūpopapattitā / na cāsti sādhanam kimcin mrdādicchām vinā prabhoh // tathā bhagavadicchaiva tathātvena prajāyate; cf. also III.35-37.

17 svatmrupāyayopapannābhasanām Upapanna", according to Abh. (IPVV II p. 146), is intended to underline that this manifestation as Self is not something that has to be proved, but is itself taken for granted and self-evident. This expression in the text may equally well refer to cittattvam: «which is spontaneously manifest as Self» (see text note 236).

18 This kārikā and the following one examine inference, since this is used by the bāhyārthavādin to establish the existence of the external object. In stating the inevitable necessity that it should operate on an object which has already been directly present to consciousness (ābhāta) and that therefore, as was shown earlier, is internal and coessential to it, Utp.'s aim is to exclude the possibility that inference might establish the existence of something that is absolutely other than consciousness, i.e. the bāhyārtha. To achieve this he is compelled, as will be seen, to include the concept of sāmānyatodrṣṭa within the pratyakṣatodṛṣṭa.

time and place19. The senses, too, are inferred only partially and generically (kimcinmātram), as cause, and this was, in fact, formerly directly present to consciousness, through the perception of the seed etc.20

– 8 –

9. But the object that is outside the light [completely extraneous to it] has not been manifested to consciousness at all; thus its existence cannot be established even through inference.

– There has been no manifestation at all of the object that is outside the manifestation as jar etc., since this would be logically untenable; therefore its existence cannot even be established by inference. – 9 –

19 This definition of inference does not only refer to the mode of the cognitive act, but - as is clarified in this last sentence - it is careful to add that it aims at proving the existence of an object that is able to propose itself as the possible object of a purposeful activity (i.e. specified by a definite time and place; cf. text note 243). It is true that the sphere of inference is constituted by the general (cf. NB I. 16), but its ultimate application is to the particular: its 'usefulness' consists in permitting one, via an argumentation in which there are only conceptual abstractions, to establish the existence, for instance, of a 'determinate' fire, which at this moment cannot be directly perceived, and to act consequently (cf. NBT p. 21 anumānam ca lingasambaddham niyatam artham darśayati). As Hetubinduṭīkā p. 34 puts it, yat mahānasādāv anagnivyāvṛttam vastumātram prāg anubhūtaṃ na tat taddeśādisambandhitayaivānumāvikalpena smaryate kintu yatra pradeśe prāg ananubhūtaṃ tatsambandhitayā (cf. Shah 1967; 278, 280; Katsura 1984: 227).

20 According to the bāhyārthavādin the existence of the external object is inferable from that of the ābhāsa, just as from the occurrence of perception the existence of the sensory faculty is proved, even though the latter is never directly perceived. We have, therefore, an inference based on the general correlation (sāmānyatodṛṣṭa) as understood, for instance, by Jayanta (NM I p. 120 sāmānyato-dṛṣṭam tu yatra sambandhakāle 'pi linga[read lingi]svarūpam apratyakṣaṃ nityapa-roksaṃ eva sāmānyato vyāptigrahaṇād anumīyate yathā śabdādyupalabdhyā śrotrādi karanam, which Bh. I p. 231 follows very closely), where the object was and continues to be inaccessible to direct cognition. Utp. replies that in this case it is not the sensory faculty as such, in its particular individuality, that is inferred, but its generic nature of cause (therefore only one of the many ābhāsas of which it is composed), which it shares with an infinite number of other objects, such as the seed, and in which it had been perceived several times before. Thus, here too, we are faced with a case of pratyaksatodṛṣṭa, in which there is perception limited to a genera! aspect (sāmānyatodṛṣṭa, but in the Śaiva sense of the term), with the aforementioned consequences. It must be kept in mind that the terms pratyakṣato and sāmānyatodṛṣṭa have been, since Vātsyayana, understood in various ways (see, e.g. Śābarabhāṣya on Mīmāṃsāsūtra I. 1.5, NM I pp. 120-21, TS 1441-1454; cf. also Matilal 1985: 29ff; cp. Foreword).

117

I 5.10

10. The multitude of things cannot but shine resting on the self of the Lord, otherwise that act of reflective awareness which is volition (icchāmarśah) could not be produced.

– As in the self of the Lord, whose essence is consciousness, so also in the objective realities themselves there is a shining without differentiation21, for otherwise reflective awareness, in the form of volition, which necessarily concerns objects that are already shining [to the subject], consisting in the state of creator, could not exist22. – 10 –

11. The essential nature of light is reflective awareness (vimarśam); otherwise light, though 'coloured' by objects, would be similar to an insentient reality, such as crystal and so on.

– Reflective awareness (pratyavamarśah) constitutes the primary essence (mukhya ātmā) of light. In the absence of this reflective awareness, light, though objects make it assume different forms, would merely be 'limpid', but not sentient, since there is no 'savouring' (camatkṛteḥ)23. – 11 –

21 The autonomous shining beyond every duality and scission, which is peculiar to the self, is a state that also extends to things when they are immersed in the self (cf. ĪPVV p. 167 ātmani hi nirbhāsamāne tadabhedasthitām katham na nirbhāseta) in such a way that the condition of idantā does not develop.

22 The will to produce something – at every level – presupposes an object that is in some way already present, 'shining', in consciousness (see below p. 126 n. 43). Volition, in order to be such, must be connected with a desired object that 'limits' it from within, that gives it a content. The will of the Lord, therefore, presupposes the shining of the whole of objective reality, which, not being separate from the subject, shines as the I shines (cf. ĪPVV II pp. 166-167). Furthermore, volition, being a particular form of reflective awareness (cf. ŚD II.84-5), is already in itself the proof of the existence of the 'light' with which we know it is inseparably connected (see ibid. p. 167 tena avaśyambhavino vimarśāt tatprakāśo 'numīyate; p. 169 icchādi parāmarśaviśeṣa eva parāmarśaś ca prakāśasya svabhāvah).

23 Camatkṛti, camatkāra, is one of the key-words of this school. Abh. (ĪPVV II p. 177) says: «Camatkṛti means the act of a person savouring (bhuñjānāsya), that is, the bliss constituted by the full achievement of fruition». Bhuñjāna, Abh. explains later, is he who, engaged in savouring a taste such as sweetness etc., unlike a voracious glutton, lets the experience rest on the cognizing subject, that is, makes the 'subject' part predominate over the 'object' part. Camatkāra, in its highest form, is therefore both the act of savouring and fullness, absolute independence, consciousness devoid of obstacles (avighnā samvit) a savouring which is in the final analysis a savouring of one's own self and of one's own bliss, which eludes every

118

12. Precisely for this reason the self has been defined as ‘sentience’ I 5.12 (caitanyam) meaning by this the activity of consciousness in the sense of being the subject of this activity. It is thanks to sentiency, in fact, that the self differs from insentient reality.

• Sentience - that is, the power of consciousness, perceiving - though an abstract (bhāva9)24, has been said to be the primary nature - the reflective awareness (vimarśa°) - of the substance self, being that which distinguishes it from insentient reality. This activity of being conscious means precisely being the subject of such activity25. - 12 -

screen or intermediation of the world of objects, and even goes beyond that relative and momentary transcendence that one has in the aesthetic experience of poetry and the theatre. The terms with which it is glossed or with which it is closely related may be grouped according to its principal components: cognition, bliss, wonder. The first group includes, for example, vimarśa (pratyavamarśa etc.), carvanā, āsvāda, rasanā, prīti; the second includes ānanda, nirvṛti, vísrānti, laya; the third vismaya, vikāsa. This last sense (prevalent in the sphere of rhetoric and aesthetics, and in non-technical usage) refers to the wonder and astonishment that accompanies the return of reality to its original ‘virgin’ state, the transfiguration that this type of experience operates on reality in freeing it from the veils of the I and of the mine. Utp. seems to have been the first to use this word pregnantly. However, already vismaya in ŚS (I. 12 vismayo yogabhūmikāh), glossed by Kṣemarāja with āścaryamānānatā (see also the verse from the Kulayuki, cit. ŚSV p. 13, ātmā caivātmanā jñāto yadā bhavati //) and āmoda in ŚD (c.g. I. 7b yadā tu tasyā cidharamavibhavāmodajrmbhayā) glossed by Utp. precisely with camatkāra, seem to hint at very similar concepts. On camatkāra see Gnoli, 1968: XLV-XLVII, 59–62; Gnoli, 1985: 32 and n. 90; see also Hulin 1978: 343–58, Masson-Patwardhan 1969: 46; id. 1970: 17–8 and notes; Larson 1976.

24 Bhāva can also be understood here in the sense of ‘quality (of the substance self)’; the general meaning is basically the same (cf. n. 25). According to Kātyāyana’s vārttika on P. V.1.119 tasya bhāvas tvatalau (see n. 25), the abstract nominal suffixes (bhāvapratyaya), such as -syañ in caitanyam, denote an essential quality (bhāva being thus interpreted as guna); cf. the remark of Helārāja (Prakīrna-kaprakāśa I p. 194) ittham atra viśesanam svarūpena param uparañjayati sūtre bhāvaśabdena pratipāditam. bhavaty asmāt tena prakāreṇa dravyam iti bhāvah.

25 Being ‘conscious’ is the attribute (dharma) of the substance (dravya, dharmin) ‘self’, that is in fact the dravya par excellence, because everything without distinction rests on him (cf. ĪPV I p. 248). Saying ‘the self is sentience’ (ŚS. I. 1), instead of ‘the self is sentient’ or ‘in the self there is sentience’, means underlining the absolute pre-eminence of this quality over all others (cf. ĪPVV II p. 33 bhāva-pratyaye taddharmapradhānyam darśitam; see also PV II.61-2). In fact, permanence, incorporeity etc. may also pertain to other entities, whereas consciousness only pertains to the self and it alone suffices to characterize it (cf. ĪPV I p. 248,

I 5.13

13. Consciousness has as its essential nature reflective awareness (pratyavamarśa°); it is the supreme Word (parāvāk) that arises freely. It is freedom in the absolute sense, the sovereignty (aiśvaryam) of the supreme Self.

• This is the First Word (ādyā vāc)26, in which the expressible is

ŚSV p. 2). «The suffixes of the abstract -tā and -tvam - says Kātyāyana (vārttika V on Pān. V I 119) - indicate that quality whose presence in the dravya determines dhāne tātvau bhavatah; one can guess the quotation in a corrupted passage of ĪPVV II p. 186, yad āha madhyamo muniḥ - yasya gunasyā hi ...). In the particular case of caitanya the abstract is particularly significant: cetana, as Abh. says (ĪPV I p. 247), is a kṛdanta and as such its abstract denotes a 'relation' (sambandha) and, through the relation, the two related elements, i.e. the subject and the action of perceiving, of being conscious (eit-). Abh.'s argumentation is implicitly based on the grammatical maxim (see Torella 1987: 158-59) samāsakṛttaddhiteṣu sambandhābhi-dhānam anyatra rūdhyabhinnaṛūpāvyabhicāritasambandhebhyah (Kielhorn 1874: 537;

cf. Siradeva's Brhatparibhāṣāvṛtti 130 and Haribhāskarāgnihotri's Paribhāṣābhāskarāḥ 132 in Paribhāṣāsaṃgraha pp. 272 and 373-74) «(the abstract nominal suffixes) when suffixed to compounds, to words formed with primary derivational suffixes and to words formed with secondary derivational suffixes, name relations, except (when words of the aforementioned types) are idiomatic or have the same phonic form (as a word expressing a general property or universal), or name an invariable relation» (transl. Hayes 1983: 710). This paribhāṣā is also quoted in Helārāja's Prakīrakaprakāśa on VP III.5.1 (I p. 194) and, in an abridged form (samāsakṛttad-dhiteṣu sambandhābhidhānam) ibid. p. 55 on VP III.1.47). Kauṇḍa Bhaṭṭa (who also quotes the same maxim in a slightly modified form) in Vyākaraṇabhūṣaṇasāra p. 344 refers to the treatment of the matter by Bhartrhari himself in the MBh ṭīkā (not found in the fragmentary text that has come down to us). Tattvabodhinī on Siddhāntakaumudī 1781 (cf. Hattori 1968:85) even seems to ascribe the authorship of the maxim to Bhartrhari himself. Cf. also, from a different point of view, VP III.1.34.

26 The vrtti does not dwell on the levels of the Word; it mentions only the supreme level (ādyā vāc). On the contrary, this subject is developed in the ṭīkā - as is seen from ĪPVV - where the polemic with the Vaiyākaraṇas, which Somānanda had started in Chapter II of the ŚD (Gnoli 1959; Gaurinath Shastri 1959: 69; Ruegg 1959: 11), is also reproposed. Grammarians (particularly Bhartrhari), as is known, are criticized for having conceived a division of the Word into three levels, instead of four, and for having consequently identified the supreme level with Paśyantī. For a description of the levels of vāc see PTV pp. 102-159 (Gnoli 1985: 60-97); Ruegg 1959: 79-81, Gnoli 1959: 55-63, Iyer 1969: 142-145, Aklujkar 1970: 67-75, Padoux 1990: 166-222, Torella forthcoming. Though Bhartrhari is unanimously thought (e.g. in the Śaiva tradition; but cfr. Intr. p. XXVI) to be the upholder of the threefold division, the VP seems to hint at a further level, which may be seen as the supreme form of Paśyantī (VP I vrtti

undifferentiated27, without beginning or end in that it is constituted by

I 5.13 perpetual consciousness, autonomous. This is pure freedom, independent

of any other reality, which has the name of ‘sovereignty’. - 13 -

14. It is the luminous vibrating (sphurattā)28, the absolute being

(mahāsattā)29, unmodified by space and time; it is that which is said to

be the heart (hrdayam) of the supreme Lord, insofar as it is his essence.

p. 216, but this interpretation seems more open to doubt) or as parā prakrti (VP III.

2. 15 etc.); cf. Iyer 1968: 143-5, Aklujkar 1970: 69-70 and n. 6, Aklujkar 1990.

Among the Vaiyākaraṇas only Nāgeśa (MBhUddyota pp. 40-41; Paramalaghumanjuṣā

p. 68 caturvidhā hi vāg asti ...; see also his commentary on the Yogasūtra) explicitly

maintains the division into four levels. It is to be noted that, four centuries before

Nāgeśa, Sāyana in the Rgvedabhāṣya proposed, among others, the same interpretation

of the verse catvāri vākparimitā padāni ..., quoted in the Paśpaśā. It would have

been very interesting to see how Bhartṛhari, whose ṭīkā on that portion of the MBh

is extant (Bronkhorst 1987), interpreted it, but unfortunately he is completely silent

on that point. The verse, however, is quoted in the Vrtti on VP I.159 and interpreted

there in terms of tripartition (p. 220).

27 Abh. refers to three different interpretations of abhinnavācya given in the ṭīkā

(ĪPVV p. 191): sasthī° or pañcamīsamāsa and karmadhāraya, that is, ‘in (or from)

which the expressible is undifferentiated’, ‘designated as undifferentiated’.

28 Sphurattā and the closely related term spanda are specially fit to designate

the highest śakti of Parameśvara due to their paradoxical nature of immovable

movement - ‘oscillation sur place’ (Hulin 1978: 290) -; kimciccālanam, Abh. says,

where kimcit stands for the appearance of movement, agitation in something which

is, however, motionless, firm. In ordinary usage, too, the adjective spandavat is

referred to one who is firmly established though being full of the freedom of many

deliberations. In comparison with spanda, sphurattā (Abh. records the use of this

term and concept by Śaṅkaranandana) also contains a ‘luminous, shining’ connotation.

One may say mama sphurati ghataḥ only in the sense that this shining-vibrating

which belongs to the I is transferred to the object. Truly speaking, it is the I (whose

essence is light) who - partially departing or, in a sense, not departing at all, from

his nature of consciousness - shines-vibrates as an object (ibid. p. 200).

29 Cf. the concept of mahāsattā in Bhartṛhari. «It is being [sattā] which, being

differentiated according to the object in which it is present, is called the universal.

All words are based on that. That is the meaning of the stem and of the root; it is

eternal it is the great Soul; it is the meaning of the suffixes -tva and -tā.» (VP III.1.33-

34; transl. Iyer 1971: 25-26). Words, whether one accepts the thesis that they denote

the universal (jāti, sāmānya) or the individual substance (dravya), all express, in the

end, the pure being in the most general form, the mahāsattā, i.c. the sattā not qualified

by any particular content. This is valid not only for nouns - which refer to accom-

plished realities - but also for verbal roots - which express processes to be accom-

plished. This, so-to-speak, universals’ universal (mahāsāmānya), present in all things

and embracing even non-being (cf. Prakīrṇakaprakāśa p. 41 sarvabhāvesu sadṛūpam

sāmānyam anugataṃ / abhāvyasyāpi buddhyākāśena nirūpaṇāt), is Brahman itself.

I 5.14 - It is, by nature, luminous vibrating, it is the subject of this luminous vibrating; it is not to be understood as the counterpart of non-being (abhāvāpratiyoginī) [but] it also pervades non-being; it is existing, being, the subject of the action of being; permanent, because untouched by space and time. It is the power of the activity of consciousness, whose essence is reflective awareness. It constitutes the foundation of the self of the supreme Lord, who is all things; the various āgamas call it the 'heart'30. - 14 -

15. Precisely by virtue of this he transforms himself into the reality which is the object of cognition (jñeyīkuryāt); but this object does not subsist independently [of the subject]. If he were to tend towards it31 his freedom would cease.

• By virtue of such pure freedom, he knows no object that has any foundation outside himself, but, on the contrary, his power being unrestrained, he transforms his very self, which is not the object of cognition, into cognizable reality. If he had to depend on a cognizable object separate from himself, his state as the subject of cognition would lapse32. - 15 -

16. The Lord, thanks to his freedom which is absence of duality, by creating a self not devoid of freedom variously representing him in the forms of Īśa etc. renders the carrying out of practical activity possible.

• Therefore the Lord, by virtue of his freedom which is perfect fullness characterized by unity with the cognizable object, manifests himself as that particular cognizing subject of that particular moment, representing him in the forms of Īśvara, Śiva, the perceiving subject and so on, acting thus for the purpose of the various practical activities such as creative meditation etc.33. - 16 -

30 On the 'heart' in the Śaiva texts and, in general, in Indian tradition see Muller-Ortega 1989.

31 I.e. towards an object conceived of as independent of him.

32 The subject is by definiton free (svatantra).

33 It is only the Lord's power of freedom that renders possible this paradoxical blend of objectivity and true subjectivity. The terms of the question are lucidly expounded by Abh., in an earlier passage of ĪPVV (I p. 108): «Although freedom and dependence are two diametrically opposed realities and mutually exclude one another, Parameśvara in his supreme freedom combines them (yojayati) when he

17. The variety of notions such as 'I etc., does not entail diversity in I 5.17

the nature of the self, because a self is created precisely as he who lends himself to being the object of the reflective awareness 'I (ahaṃmrśyataiva), like action which is expressed by personal endings (tiṅvācyakarmavat).

– He who is the object of the reflective awareness 'I' on the plane of the present cognizing subject does not have the nature of 'this' (idantā), which he should have being the object of cognition34. In the diversity of ways in which the self is definitely apprehended and in which he appears to perception, what is expressed [e.g.] by the word 'īśvara' is the cognizing subject and no other; only, this happens indirectly, through the reflective awareness 'I'35. Similarly, that reality, for example, that is

creates differentiated representations such as Īśvara, the self etc. Their idantā component makes them able to become the object of meditation, worship, teaching and so on, and, on the other hand, their unveiled ahaṃtā component ensures the attainment of their true nature».

34 The power of freedom of Parameśvara, in the sphere of the general process of the objectification of the self - from which all stems - causes a certain layer to remain as though suspended between the 'I' and the 'this'. 'This' insofar as constructed, 'created' (nirmita) just like any reality posited as external, but unlike these, closer to pure subjectivity and capable of reimmersing itself in it by virtue of the contact with the 'I. This is the case with notions (and words) such as Śiva, Īśvara, self, cognizer, which are on the one hand an objectified reality, separate from consciousness, and on the other not completely so (ĪPVV II p. 210 na ca atyantam prthak svātmany eva): na ca anīśvaram, Abh. adds elsewhere (ĪPV I p. 272), otherwise meditating on īśvara or anīśvara would be the same thing, which is contradicted by the different fruits that derive from it.

35 The real self, consciousness, is characterized by aham-parāmarśa; on the contrary, reality and denominations such as Śiva, Īśvara, Bhagavān, I etc. are characterized by idam-parāmarśa whose sphere includes the individual parāmarśas (śivaparāmarśa etc.). The issue raised is how a single reality can be the basis on which both creator and created (nirmita) rest. According to Utp.'s reply, as developed by Abh., unification is provided by the aham-parāmarśa which also underlies the notions of Śiva, Īśvara etc. It is the I (aham) in fact - as Abh. says (ĪPV I p. 278- 9) - which constitutes the ultimate outcome, the definitive resting place (viśrānti-sthāna) of this kind of cognition (but also, he adds later, of all the others, without distinction); even for the cognition 'blue' the final resting place is the I, though this is more difficult for the average intellect to grasp. This distinction between the two planes is outlined by Abh., following the ṭīkā, in a passage of ĪPVV (II p. 213). In the case of the, so to speak, primary aham-parāmarśa, from the very beginning (upakrama eva) there is awareness of freedom. In the parāmarśas concerning ātman,

īśvara etc. this is true only as regards the final outcome; at the beginning, on the

123

I 5.17 denoted by the verbal form 'he cooks', is expressed by the word 'action'; for, as has been said: «The words 'action', 'quality', 'universal', 'relation' etc. do not express [directly] the various actions etc. because they are denoted instead through the verbal form 'he cooks' and so on»36. From the words 'īśvara' and 'self' one reaches the understanding of one's own self, but only insofar as one contextually has the reflective awareness 'I', that is, not directly. The apprehension of the self, however, effectively takes place, just as the object previously perceived is known by memory, through the direct perception37. - 17 -

18. Owing to the power of māyā, for the Lord it has as its object a cognizable reality differentiated [from self] and is called by the names of cognition, imagination, determination etc.

• By the power of māyā, to Parameśvara38 whose essence is light, the world - which consists of his own self - is manifested as differentiated. This very Consciousness, therefore, appears as sensory cognition; precisely this reality which is manifested as differentiated39 forms the object of memory, imagination (samkalpah) and determination (adhyavasāyah). Thought40 appearing in the form of the functions of the internal sense (manaso) and the intellect (buddhio), what is acting is in fact none other than Consciousness itself. - 18 -

contrary, there is the condition of dependence peculiar to the 'this', insofar as those notions arise marked by objectivity, despite the fact that the aham-parāmarśa is present in them and pervades them from within.

36 Untraced quotation. In tenor it may be compared to VP III 6. 1, which Abh. quotes in ĪPVV II p. 215.

37 The parallelism is clearly explained by Abh. (ibid. p. 213).

38 In departing from what appears to be the plainest interpretation ('the power of māyā of the Lord'), Abh. (ĪPVV II p. 217) follows the indications in the tīkā. In the eyes of the Lord (or rather, of the limited subject in whom he is freely contracted) the world is made to appear as differentiated from the self.

39 According to Abh. (ĪPVV II p. 218) tasyaiva bhinnasya is not to be understood as anubhūtasya, which - it is objected - would be all right for smṛti and adhyavasāya, but not for samkalpa (samkalpe tu nava ekaṣiṣaya utthāpyate). It means the external object in the broad sense of the word, i.e. that which is subject to the power of māyā.

40 The concessive meaning of api refers here, according to ĪPVV II p. 218, to the implicit objection: how can cognition, determination etc. have Consciousness as their essential nature, since they are functions of the intellect, the manas and the senses, which are in themselves insentient (see text note 291)? Cf. ŚK I. 6-7.

19. Even at the moment of the direct perception (sākṣātkāraksane 'pi) I 5.19 there is a reflective awareness (vimarśah). How otherwise could one account for such actions as running and so on, if they were thought of as being devoid of determinate awareness (pratisamdhānā°)?41

• Even at the time of that cognizing which is characterized by direct perception there is in consciousness a reflective awareness of the object (arthapratyavamarśah), in a subtle form42. In fact, a rapid action - such as running and so on - only appears to be so. The discursive articulation is already contained in embryo in the sensorial experience, and so too - still in a 'contracted' (samvartita) form - is the linguistic articulation, inseparably connected with it, for otherwise it would not be able to occur after the perception; similarly, even the most rapid and thoughtless action cannot but contain, however sketchily, an articulated awareness of the ends to be achieved etc. (cf. Hulin 1978: 294-96). In formulating this conception, which may be considered one of the pivots of Pratyabhijñā philosophy, Utp. follows in the footsteps of Bhartrhari (cf. VP I. 131-32 na so 'sti pratyayo loke yaḥ śabdānugamād rte / anuvidharm iva jñānam sarvam śabdena bhāsate // vāgrūpatā ced utkramed avabodhasya śāsvatī / na prakāśah prakāśeta sā hi pratyavamarśinī //) «There is no cognition in the world in which the word does not figure. All knowledge is, as it were, intertwined with the word. If this eternal identity of knowledge and the word were to disappear, knowledge would cease to be knowledge; it is this identity which makes identification [or, in a broader sense, 'reflective awareness'] possible» (transl. Iyer 1965: 110-11; see Matilal 1990: 133-141) and is constantly aware of Dharmakīrti's diametrically opposed position (cf. in particular PV III. 174-191ab; and, obviously, the well-known definiton of pratyakṣa as kalpanāpodham in PS; see Funayama 1992).

42 A potential linguistic articulation (śabdabhāvanā), says Bhartrhari (vrtti on VP I.131, p. 188), is present in various degrees in every cognition; it is therefore also present in the cognition devoid of mental elaboration (avikalpajñāna), though only in a contracted and hidden form (samhrtarūpa). (But at this point one - and first of all a Buddhist - might wonder on which basis it is then possible to draw a distinction between savikalpa and nirvikalpa, and, moreover, what nirvikalpa is. An elaborate answer to these crucial questions is given by Abh. in ĪPVV II p. 239-40; cf. Torella forthcoming). This condition makes the cognition that occurs remain, in one sense, as though it had not occurred, because it is not able to manifest its effects (kārya), namely, to attain the plane of verbal communication and practical use (Vṛṣabhadeva's Paddhati p. 188). That this cognition, however, has occurred is proved by the fact that a suitable occasion (nimitta), such as the subsequent occurrence of a similar cognition (Paddhati p. 189 sadrśagrahanāder nimittāt) may reawaken it and render it manifest. Bhartrhari gives the example of a person walking quickly, wholly intent on reaching his destination, who seems not to notice anything that he meets along his path - leaves, clods of earth -, until, some time later, he finds himself faced with a similar

I 5.19 as reciting, running and so on – necessarily occurs through a determinate awareness (°anusamdhānena)43 of the intention of reaching or leaving this or that place etc. which is seen at that moment44. - 19 -

20. The determinate cognition (adhyavasā) ‘this is a jar’, beyond the linguistic sign and the thing signified (nāmarūpa°), is the very power of the supreme Lord. It is manifested in the same way as the self and not in terms of ‘this’45.

object and this brings to mind the cognition that he did not previously realize he had had. This is possible, says Bhartṛhari, only if we admit the existence within the first perception of a śabdabhāvanā which, after being ‘contracted’ (cf. ĪPVV II p. 226) as it was, has now become fully manifest (vyakta). Then let us consider the case of a sensation taken in its very first moment (Vṛtti p. 170 prathamopanipātī) and not destined to further development. Even the indistinct cognition that derives from it, based on an inner formulation such as ‘this, that’ is made possible only by the presence of a śabdabhāvanā, though in a subtle form. Finally I will quote one last, particularly significant, example from those Bhartṛhari gives. This is the analysis of the process that leads from the perception of a group of words to the comprehension of the meaning of the sentence they form. The text of the Vṛtti (pp. 191-92) is ambiguous in some places and I understand it in the light of Vṛṣabhadeva’s interpretation (which is, however, questionable; cf. Aklujkar 1970: 147-48). What appears at first is the indeterminate reality of a series of separate words-meanings, isolated from one another. Subsequently (uttarakālam; Paddhati p. 191 abhinnapadārthaparicchedād uttarakālam) there is a reflective awareness (pratyavamarśa), namely a determination of the meanings understood from the words (Paddhati ibid. padebhyo ‘nubhūtānām arthānām ākāranirūpanā) and a coordination (anusamdhāna), namely the ascertainment, with mutual connection, of the meanings obtained from the words (Paddhati ibid. padebhyo ‘vagatānām arthānām parasparāvacchedanāvadhāraṇam); finally, the constitution of a single overall meaning (ekārthakāritvam). All this depends on having the word as essence (vāgrūpatāyām baddham). It is to be noted that pratyavamarśa and anusamdhāna are precisely the two terms on which Utp.’s Vṛtti on this kārikā is centred.

43 Anusamdhānu, to be understood also in the sense of ‘coordination’ of the distinct wills to leave one thing and seek another, present in the act of volition (see text note 294). As stated before (I.5.10), volition is a particular form of reflective awareness; it must include a vimarśa ‘this is that’, though in a generic form, touching the desired object in same way. In this state the powers of volition, knowledge and action are united with the light full of consciousness and bliss (ĪPVV II p. 233 and ŚDVṛ pp. 91-92). See also ŚD I.9.11ab.

44 According to Abh. (ĪPVV II p. 233, see text note 294) the expression deśādi° may be understood in two ways: ‘other substances or qualities’ or ‘substances etc. existing elsewhere or of another nature’.

45 The kārikā is to be understood as a reply to an objection which we find made explicit by Abh. in ĪPV and ĪPVV. The objector admits that the determinate cogni-

• The determinate cognition 'this' and 'jar', beyond the linguistic expression and the thing signified, which in themselves appear as differentiated [from the subject], consists essentially in the power of Consciousness: it is manifested, like the self46, in terms of non-differentiation. - 20 -

21. However, insofar as they conform to the time and place of the differentiated cognizable reality, [perceptive] cognition, memory and determinate cognition etc. appear to be successive47.

• Differentiated cognizable reality, such as the jar and so on, is made manifest to the Consciousness48 as differentiated by this or that time and place due to the power of māyā. [Perceptive] cognition, memory etc., closely resting on this differentiated reality, are manifested at different times and places etc. - 21 -

tion in the subtle form the previous kārikā spoke of may be considered not separate from the subject or from the 'light', as we have seen, but he wonders how one can continue to claim this once the determinate cognition (adhyavasāya) 'this is a jar' - becomes evident (sphuta) and, on being articulated in words, fully assumes the form of vikalpa. The word 'jar' - like the thing it signifies - is manifested as an objective reality, separate from the light of consciousness, and this cannot but have repercussions on the determinate cognition the very essence of which is the linguistic expression. The meaning of Utp.'s reply is the following. No one has ever said that it is the ordinary manifested word (which is only the lowest level of language) that constitutes the vimarśa; it is, on the contrary, on the same level as any reality that appears as external. This word and the thing it signifies are only the object of unification by a determinative cognition, which is rooted in the Lord's pure power of consciousness (of which it is one of the manifestations). In unifying nāma and rūpa (śabda and artha) determinate cognition remains distinct from them, on another level, retricving them through the outcome of its very process within the sphere of the 'I'.

46 The various notions of self, cognizing subject, Lord etc. shine directly to the consciousness as 'I' (cf. above kārikās 16-17 above).

47 One can explain this kārikā, too, as the reply to an unformulated objection, which it is possible to intuit from the thread of the argumentation and with the aid of Abh.'s commentaries. If every form of cognizing - the objector wonders - in the ultimate analysis is resolved in the 'I', what results is an absolute unity of cognition in which there is no room for spatio-temporal differentiations (which may characterize only the object). And yet this differentiation is evident at every moment of ordinary experience ...

48 See above n. 38.

I 6.1

CHAPTER VI

1. The reflective awareness 'I', which is the very essence of light, is not a mental construct (vikalpah), although it is informed by the word (vāgvapuh). For a vikalpa is an act of ascertainment (viniścayah) presenting a duality (dvayākṣepī)1.

– The reflective awareness concerning the self, the reflective awareness 'I', which constitutes the very nature of light, cannot be called vikalpa even if it is essentially associated with a 'discourse' (sābhilāpo 'pi)2 since the word that informs it is the supreme word. Indeed, the vikalpa3 is an ascertainment (niścayah) acquired through the negation of the opposite4, and, as regards pure light, there is no possibility of the existence of something that is its opposite5. – I –

1 Without a duality of this - non-this, Abh. remarks (ĪPVV II p. 280), how can the functions peculiar to vikalpa be carried out, that is, how and what to unify, divide, imagine ?

2 Utp. refers here critically to the well-known definition of kalpana given by Dharmakīrti (NB I. 5) abhilāpasamsargayogyapratibhāsā pratītiḥ kalpanā, which in turn derives, with modifications, from Dignāga (PS I. 3d nāmajātyādiyojanā; cf. Funayama 1992). During the discussion of this subject in ĪPVV, Abh. repeatedly quotes passages from Dharmottara, which turn out to be either paraphrases or summaries of passages from the NBT (ad I. 5).

3 On vikalpa in general see Matilal 1986: 313-15. The term vikalpa is analysed in ĪPVV II p. 274 in a few different, but compatible (or even complementary) ways, each of them singling out an aspect of this complex notion. a) vividhakalpanā, to be understood either as putting together what is differentiated in reality (bhinnaśyāpi ... yojanā) or as conceiving as manifold what is one in reality (ekasyāpi vaividhyena kalpanā). b) Cutting (klp- in the sense of vicchedanā) another form which has been imagined (vividhasya sankitasya rūpāntarasya kalpanam vicchedanam). c) After dividing (vibhajya), removing from the other and protecting by making it enter into the cognizer's mind (vi-kal-pa). And later on (p. 281), d) vividham kalpanam, i.e. manifold conceiving, in the sense that at same time it has something being affirmed and something being negated as its object (vidhīyamānaniṣidhyamānatatayā).

4 The main characteristic of vikalpa is that of projecting the object outside the I, of knowing it in terms of separation: 'this'. In analyzing the working of vikalpa, Āhnika VI dwells particularly on its aspect of 'exclusion'. In the avikalpaka state the thing is 'full', just like consciousness, but it cannot become the object of practical

(notes 4-5 next pages)

128

4

To elaborate:

2. In fact, the manifestation of two opposite realities is possible in the case of 'jar' and 'non-jar'. On the contrary, the manifestation of a

usage as regards cognition, communication and so on (ĪPV I pp. 309-310 tadavikalpadaśāyam citsvabhāvo 'sau ghaṭaḥ cidvad eva viśvaśarīraḥ pūrṇaḥ, na ca tena kecid vyavahārāḥ). For this purpose the subject in the māyic world has to break into pieces what is full in itself, distinguish and negate: knowing becomes contrasting.

Also in this case Utp. appears drawn into the orbit of Buddhist thought. The first formulation of the apoha theory is found, as is well-known, in the PS and in the Nyāyamukha, though apparently still limited in its application to the ambit of the inferential process and the denotation of words (for the development of the apoha theory in Dignāga's works see Frauwallner 1959: 99-106). The word expresses things as qualified by the negation of the other meanings (sahdo 'rthāntaranivrttiviśiṣṭān eva bhāvān āha, cit. in PVSV pp. 62-63; the passage comes from the vrtti on PS V.36d, Hattori 1980: 71). Or again: '[The word] expresses its meaning through the exclusion of others' (PS V.1cd ... svārtham anyāpohena bhāsate). It was, then, Dharmakīrti who extended the application of the apoha to all the fields where universals operate, that is, in the broad sense, to the contents of discursive thought (PV I.134 tenānyāpohaviṣayāḥ proktāḥ sāmānyagocarāḥ / śabdāś ca buddhayaś caiva vastuny eṣām asambhavāt //; PVSV pp. 92-93 tasmāt siddham etat sarve śabdā vivekaviṣayā vikalpāś ca). Therefore, words (or vikalpas), do not express universals, undividedly present in single particular things, - as the Realists maintain - but only a 'difference' (bheda) through the exclusion of what is other, namely, of all those things which are joined by the fact of having different effects with respect to the thing intended. The intrinsic nature of the thing constitutes its difference, the exclusion of what is other its (potentially) common dimension (PV I.180ab cintyete svātmanā bheda vyāvṛttyā ca samānatā); the anyāpoha, after all, is seen to share all the essential characters of the sāmānya, that is, ekatva nityatva pratyekaparisamāpti (cf. PS V.36cd; Katsura 1991: 132). Therefore, the word is applied to that difference in which the exclusion of the other has made a common structure appear (PVSV p. 62 ... bhede samānarūpāpratibhāsiny ākṣiptatadanyavyāvṛttir niveśyata iti). However, it is not a question of two distinct operations (with a relationship of viśeṣya-viśeṣaṇa), Dharmakīrti says - and Jinendrabuddhi will repeat, a few centuries later, in his commentary on the PS (Stcherbatsky 1930²: 464) - since the denotation of its own meaning entails by itself the exclusion of others, precisely inasmuch as the nature of its own meaning is the 'difference' (PVSV p. 63). But, Dharmakīrti adds a little later (PV I.128), the difference is not a real thing (vastu) - nor is exclusion -; it is only a relative term. Something real would be the form (rūpa), but the object of the word is not rūpa but bheda (abhāva, viveka; PV I.185 and PVSV thereon).

Almost all the debates of the later Buddhist thinkers, who will attempt to recast the apoha theory taking into account the Nyāya and Mīmāṃsā criticism, are centred on these points. The current division (Mookerjee 1935:132-33) between 'negativists' (Dignāga and Dharmakīrti; for a distinction of their respective positions see Katsura 1991: 142-143), 'positivists' (Śāntarakṣita and Kamalaśīla: the word expresses firstly

129

I 6.2 reality that is other and differentiable6 from light, on the same plane (iva)7, is not possible.

a positive entity and only secondly, by implication, the exclusion of the other, TS 1018 tasya ca pratibimbasya gatāv evāvagamyate i sāmarthyād anyaviśleṣo nāsyānyātmakatā yataḥ //) and 'synthetists' (Jñānaśrīmitra and Ratnakīrti: the word expresses a positive entity qualified by the exclusion of the other; Ratnakīrti's Apohasiddhi p. 59 nāsmabhir apohaśabdena vidhir eva kevalo 'bhipretaḥ / nāpy anyavyāvṛttimatram i kintu anyāpohaviśiṣṭo vidhir śabdānām arthah, then taken up again by Mokṣākaragupta) seems too simplistic and fails to grasp what appears as a substantial unity with different formulations (Katsura 1986, Siderits 1986). The Pratyabhijñā position is very close to Ratnakīrti's: the words (and vikalpas) rest on a reality which is mainly affirmative, qualified by the implicit negation of the opposite (ĪPVV II p. 289 antarlīnikrtapratiyogyabhāvaviśiṣṭavidhipradhānavasuniṣṭhāḥ; p. 280 vidhau viśrāntatvāt niścayasya). In this process three moments may be singled out: imagining what is other than the object intended, then superimposing it onto the object and lastly negating it. (ibid. p. 289 vikalpayet, āropayet niṣedhec ca). Or, according to another formulation: every apprehension includes the superimposition of the manifestations to be excluded and their negation (ibid. p. 290 sarvā prameyeṣu pratipattir apohyābhāsāropanīṣedhavyāpārāntahkāriṇī).

5 The countercorrelate of light, its contrary, cannot in fact be either aprakāśamāna - because in this case it would simply not exist - or prakāśamāna - because it would be of the same essence as light (ĪPVV II p. 279).

6 Cf. ĪPV I p. 307 bhedinah ... apohanātmaka bhedanavyāpārasahiṣṇoh [both Bh and KSTS Ed. read incorrectly °vyāpārāsahiṣṇoh ]. On the plane of knowable reality it is possible to imagine, with respect to a 'blue', a 'non-blue' (and what is other than blue indeed exists); why then, on the plane of the knowing subject, imagining a reality other than consciousness should be inadmissible ? This objection is put forward by Utp. and Abh. (ĪPVV II p. 288) and answered as follows. On the plane of prameya, created by māyā and made of a multitude of ābhāsas, anekatva and vaicitrya are not surprising, whereas in pure light a reality different (cf. bhinnasya in the vrtti) from the nature of pramātr cannot possibly exist. But even admitting that lights may be differentiated in their nature owing to a differentiation of space and time, nonetheless, just as, when a certain 'blue' is ascertained, only yellow etc. are negated - not the other 'blues' -, likewise, when a light is ascertained the other lights are not negated. The two parts of this argument unfold what is implicit in dvitīyasya bhinnasya in the vrtti. Therefore, even if there were a 'second' (etc.) light, this would not be subject to exclusion. However, a 'second' light is not admissible as a reality (ĪPV I p. 308 na ca prakāśasya svarūpadeśālabhedo yena dvitīyaḥ prakāśa ekasmād apohyate). A 'second' light as the product of mental construction cannot constitute a pratiyogin in the same way as 'being cause, effect, instantaneous' cannot be considered a pratiyogin with respect to 'blue' (ĪPVV II p. 282 svakalpanākalpitena dvitīyena na pratiyogiparipūranam kriyate / na hi nīlasya kāranatvakāryatvakṣaṇikatvādi pratiyogipakṣe vartate, na hi tat pratik kiñcit).

7 What is needed for two entities to be termed pratiyogin is precisely their opposing each other while remaining on the same plane. My translation makes the

130

• Otherness with respect to light (prakāśetaratvam) is not possible, I 6.2 since an opposite reality, second to and distinct from light, called non-light, is not manifested. There being no possibility of exclusion (vyapohanāyogāt), since such an opposite reality does not exist, one cannot, therefore, speak of mental construct (vikalpatā°) [with reference to ahampratyavamarśa]. - 2 -

Likewise:

3. For we call vikalpa the ascertainment of a certain thing (tanniścayanam) [e.g. 'jar' arrived at through the exclusion of its opposite (atadvyapohanāt) by the knowing subject only (mātraiva), in whom that and the opposite of that are manifested (tadatatpratibhābhājā)8.

• The ascertainment 'jar', brought about by the free knowing subject only within whom lie the manifestations of that [the jar] and its opposite, and that takes place through the exclusion of the opposite: this is the activity called vikalpa. - 3 -

4-5. Having left the plane of Consciousness because of the influence of māyā, that reflective awareness 'I' which addresses differentiated realities - e.g. the body, the intellect, the vital breath or that imagined entity, similar to ether - understood as the knowing subject, that reflective awareness, excluding what is other than its object, is a vikalpa,

value of iva explicit and in so doing follows Abh.'s commentary. Bh., too, insists on this point (II p. 307) tulyakaksya eva hi pratispardhitayā bhedena purah sthātum śaknoti iti tulyakaksyasyasyety uktam; Abh. (ĪPVV II p. 282) adds pratiyogitvam hi tat-sthānāpannatvam.

8 The Buddhist description of vikalpa's modus operandi thus seems liable to be shared also by the Śaivas. Moreover, on a close scrutiny, it is only within the śaiva context that it may logically stand and elude criticism. Indeed, if one accepts the Buddhist presupposition of things being self-confined, it results that what appears from the cognition of a thing is only that same thing and the awareness of its cognition, and one does not understand where the 'other' things and their negation spring from. But this difficulty concerns only the Buddhist (ĪPV I p. 308 evam śakyah paryayojyo na tu vayam), not the śaiva conception centred in a cognizer who is not bound to the present object and its cognition (ĪPVV II p. 291 visayataddarśa-nādipāratantryam ujjhatā) but is free to dynamically aggregate, separate and merge all the various cognitions. The centrality of the I, which characterizes the śaiva conception, is pointed out by the particle eva in the kārikā (mātraiva) and in the vrtti (pramātṛ eva).

I 6.4-5 which arises precisely from the manifestation of an opposite reality to be excluded9.

• The reflective awareness 'I' as the knowing subject, which because of the power of māyā (māyāśaktyā) of the Consciousness-principle itself, addresses realities that are manifested as separate such as the body, the intellect, the interior tactile sensation10 or that imagined11 entity which is the void (śūnya), beyond them, similar to ether12 - this form of reflective awareness 'I' is nothing other than a vikalpa, like 'this is a jar', since it excludes the various opposite entities that are manifested, such as the body and so on. - 4-5 -

6. The connection, in the case of an entity that is manifested at a certain time (kādācitkāvabhāse), with something that was formerly manifested etc., this connection, too, which occurs thanks to the latent impression, is a mental construct (kalpanā) since it is applied to realities that are manifested as differentiated13.

• Within the sphere of the various totally differentiated and discontinuous manifestations such as the body, the void etc., the connection in tinuous manifestations such as the body, the void etc., the connection in

9 In his commentaries Abh. introduces these two stanzas with an objection to which they furnish the reply. The reflective awareness 'I' (ahampratyavamarśa) occurs through the exclusion of non-I; is it not therefore a vikalpa itself? The answer is that there are two kinds of ahampratyavamarśa, one is pure and the other is impure; the objector's remark is pertinent only to the latter, which is the object of the two stanzas.

10 Sparśa (but āntara) is the sensory faculty that reveals the existence of the prāṇa, just as antahkarana reveals that of the mind and the sense of the I, or sight reveals that of the body. But Abh., in commenting on the tīkā, grasps other meanings of this āntaratvam of the prāṇa (not to be confused with physical breath), namely its not being directed outwards, its being the essential element and its permeating all the senses (cf. ĪPVV II pp. 295-7).

11 Imagined insofar as, unlike the other realities with which subjectivity is wrongly identified, it cannot be directly known by the internal and external senses (ibid. p. 297).

12 For example, it shares with ether 'vacuity' and the fact that it is not a pure non-entity (ibid. p. 297). On the void see below pp. 203 ff.

13 Once again it is useful to resort to Abh.'s commentaries to understand the link between the present stanza and the previous one. The notion of the I (ahampratyaya) is not only to be divided into pure and impure (śuddha - aśuddha), but each of these two forms has in turn two aspects: an immediate awareness through a direct internal experience and a mediated one which is the result of a connection

unity (aikyayojanā). with former manifestations, with an individuality, I 6.6

with a name, (a connection) made possible by the latent impressions which ensure the internal permanence of the former manifestations, - this connection in unity, which is essentially an activity of the knowing subject (pramātrvyāpararūpā), is merely a mental construct and precisely that called 'recognition'. - 6 -

7. Thus also in the course of ordinary reality14 the Lord, entering the body etc., renders externally manifest by his volition the multitude of objects that shine within him.

• At the moment of the original creation, as in the course of everyday reality, Maheśvara, by virtue of the power of māyā15, by entering the body etc. conceived of as self, creates the [limited] knower and thanks to the power of doer (kartrśakryā) gradually16 renders the various objects

(anusamdhāna) between several elements. Now in the case of the aśuddha-ahampratyaya (which concerns us here), the direct form (anubhavarūpa) - e.g. 'I am thin', 'I am old' etc. - has already been spoken of and acknowledged as vikalpa (see above kārikās 4-5). But it might be objected that the second form, since it works through unification - 'that which was previously a child, then an adult, with a determinate name, character etc. is the I - and not exclusion (as was seen in the case of the first), does not have the characteristics of the vikalpa. The present stanza answers this objection by essentially saying that an anusamdhāna that has as its object realities conceived as bāhya and bhinna cannot bring about a real unification as it is polluted by their appearing in the sign of vikalpa. On the contrary, it is on the plane of Sadāśiva where the 'this' is not yet bhinna that anusamdhāna brings about a śuddha-ahampratyaya. However, Abh. concludes (ĪPV I p. 327), in every anusamdhāna, namely in every act of unification we perform in ordinary reality, the light of vidyāśakti (see below pp. 193-194) shines for an instant: each of these may be taken by the limited subject as the first step in the process leading to the overcoming of differentiation and duality.

14 The five activities of the Lord take place not only on a cosmic level but also in everyday reality, which is, in various ways, continually penetrated and supported by them (cf. ŚSV pp. 10-12, PH pp. 22-26).

15 Utp. glosses icchayā with māyāśaktyā and kartrśaktyā; as Abh. explains (ĪPVV II p. 323) māyāśakti is responsible for the identification of the I with the various seats of limited subjectivity (body etc.), kartrśakti - in the form of kalā and vidyā - for the manifestation of reality as external.

16 I.e. linking the various manifestations with particular subjects and then removing them from those to link them with others and so on (cf. ĪPVV II pp. 316-7); this pertains to the microcosmic level of the five divine operations (creation, reabsorption etc.). Or, according to another interpretation (ibid. p. 323), gradually connecting on the pure, mixed and impure levels of creation.

I 6.7 that shine within him externally manifest. Creating is precisely rendering manifest in this way (tathābhāsanam eva). On the contrary, if he does not enter the body etc., then the whole manifestation of objective reality will take place spontaneously and simultaneously in the form of 'I am all this'17. - 7 -

8. Therefore it is definitely ascertained that whether in memory, or in vikalpa - that depends on exclusion -, or in direct perception, the manifestation of the object is internal.

• In every kind of cognition (sarvasamvitsu), all objective manifestation is always shining as absorbed in the cognizer (pramātrsamlīnah) like the consciousness-principle, in an appropriate way (ānurūpyeṇa)18. - 8 -

9. However, in direct perception, which consists in rendering the object externally manifest, [this manifestation] occurs spontaneously (naisargikah), while in memory etc. this is informed by a former perception.

• In direct perception, consisting in rendering the object externally manifest19, the manifestation of the object within the consciousness (cittattvasyāntar) is spontaneous (sahaja)20; on the contrary, in memory etc. it has as its essence a former perception, and it is for this very reason that memory is said to arise from latent impressions21. - 9 -

17 It is the plane of Sadāśiva-Īśvara.

18 When, for example, the cognition appears as 'this is a jar' and there does not seem to be any reason to claim a connection with the subject, even then, on the contrary, the fact that the cognition and its object rest ultimately on the cognizing subject reveals that their nature is essentially consciousness. The expression ānurūpyeṇa seems to refer to the possibility of relative differentiations of degree within a basic identity; see also above I. 5. 10 and vrtti.

19 In the sense of perceiving it as other than self, as 'this'.

20 I.e. it does not depend on external factors (former perceptions), but, as Abh. says (ĪPVV II p. 329); it stems from the freedom of the Lord himself (cf. text note 353). Also see PTV p. 221 (svasāmarthyodbhūta°), ĪPVV I p. 114 (svasāmarthyot-thanirvikalpakajñāna°).

21 The criterion on the basis of which it is possible to make a general distinction between the various kinds of cognition lies in the spontaneous or mediated manifestation of the object. The object is manifested spontaneously in direct perception, and mediately in vikalpa (or rather in a certain kind of vikalpa; see the following kārikā), where the manifestation of the object must necessarily pass through the reawakening of a former perception. Therefore we have as homologous terms

10. It is also spontaneous in the case of the vikalpa that acts I 6.10 autonomously (svairacāriṇi) rendering particular configurations manifest in the mental sphere at will.

– However, the independent (svatantrah) vikalpa22 renders this or that thing manifest, whether new or characterized by a different organization of its parts, making it the object of the mind even if it has never entered the field of the senses such as sight etc.; and it renders it manifest at will without considering its having been previously perceived23. Within the sphere of this form of vikalpa the manifestation of the object is spontaneous, too. – 10 –

11. Therefore, owing to the fact of rendering manifest at will, what is mentally conceived, knowledge and action have been proved to be fully evident (sphuṭe) in every being.

– By virtue of the capacity for creating and knowing objects not formerly experienced, found in the vikalpa, the presence of the power to know and create everything is evident in everyone. – 11 –

naisargika-avyavadhāna-sphuṭatā on the one hand and samiskāraja-vyavadhāna-asphuṭatā on the other (cf Bh I p. 337). The fact of making the object manifest as external, which is mentioned as characteristic of direct perception, does not constitute an equally valid criterion; in fact, as Abh. notes, in the other kinds of knowledge, too, the object is projected outside the I, into the antahkarana.

22 There are two kinds of vikalpa, according to whether it depends or does not depend on a former direct perception. The first kind includes, of course, memory (smṛti), determination (adhyavasāya), conjecture (utpreksana); the second, to which the stanza refers, – the autonomous (svatantra) vikalpa – includes fantasy (manorājya) and imagination (saṃkalpa).

23 Both kinds of vikalpa work on material furnished by former perceptions. The 'novelty' (navatā, apūrvatvam) which characterizes the object of the vikalpa called 'independent' is not therefore to be found in an impossible 'novelty' of the individual ābhāsa, but in not considering the former manifestation of the object at a particular time and place (which is, on the contrary, essential in the two main kinds of 'dependent' vikalpa – memory and determination) and in the different connection (yojanā) of the individual ābhāsas – different compared to the former experience – with the consequent creation of new arrangements (sanniveśa). The example Abh. gives of this is imagining a white clephant, with two trunks, a hundred tusks etc. Another element that characterizes the autonomous vikalpa is that it comes into being and ceases independently of the subject's will, as the mind's free wandering (see ĪPVV II p. 332; in the compound samsthānayojanāṃśena, l. 14, the final na is to be read separately).

I 7.1

Chapter VII

1. And this intuitive light (pratibhā)¹ influenced (ruṣṭitā) by the succession of all the various objects is the knowing subject, which is consciousness devoid of succession and limit², Maheśvara.

• And this light (ābhāsah) variegated (ācchuritah) by the succession of all the various objects is, insofar as it rests internally, the knowing subject, called the 'self', which pervades the times of all the cognitive acts (sarvasamvitkālavyāpī) and consists of consciousness devoid of succession and limit; and it is Maheśvara in that it is the creator as regards the cognizable reality that constitutes its 'body' (svāṅgabhūte). - 1 -

2. Relation (jñāteyam) is possible if things, through the paths of the various and distinct cognitive acts, come to rest on a single knowing subject.

• Once they have merged and been absorbed (upalīya) into the sea of the single knowing subject through the currents of the many cognitive acts [flowing into him]³, the different objects are set in the network of

¹ Pratibhā, which I have translated vaguely as 'intuitive light', is the revelation, the presentation of the object to consciousness, its illumination by the light of the subject (prati-bhā), in which in the final analysis it rests. The succession which seems to be projected on the light when the objects shine within it (or rather when it shines as objects) disappears at the moment when the perception of the object as external reality 'this', becoming introverted, dissolves and finds its root in the I: 'I perceive this'; cf. the formulation in APS v. 22, often quoted in Śaiva texts. pratibhā is here to be understood as synonymous with prakāśa, ābhāsa (cf. I.6.3) and not in the pregnant sense of 'intuition', that is, a special, superhuman kind of direct vision (cf. Gopinath Kaviraj 1966: 1-44, Gonda 1963: 318-48, Iyer 1969: 86-92; Tola-Dragonetti 1990).

² In his two commentaries Abh. divides the third pāda into akramā (referring to pratibhā) and anantacidrūpah (referring to pramātā). The vrtti, on the contrary, takes it as a single compound, referring to pramātā.

³ Or also, following Abh.'s commentary (see text note 367): 'by the means represented by the currents of the many cognitive acts', currents that force the objects from the external to the internal plane, into consciousness.

136

connections pertaining to practical reality (vyavahārasamanvayam), such as, for instance, the cause-effect relation and so on. - 2 -

3. A connection between objects having a spatio-temporal succession and being self-contained is established by their manifestation in terms of unification (sakrdābhāsa°).4 Otherwise, no connection at all is possible.

• Things, both in their own nature and manifestation5, are self-contained. A connection between them depends on their simultaneous and unitary manifestation (yugapadekābhāsa°). This undivided [manifestation, i.e. connection] is possible only if absorbed into the knowing subject. - 3 -

4. The perceptions and non-perceptions (pratyakṣānupalambhānām), which in themselves concern this or that separate part, may cause the establishment of the cause-effect relation only if they rest on a single knowing subject.

• Perceptions and non-perceptions determine the cause-effect relation and also its establishment (tatsiddhiḥ)6, [only] once they have entered into a connection with one another (samanvayam āgatya) through a single knowing subject. If they do not enter into a reciprocal connection (asamanvitāḥ) perceptions and non-perceptions fade away at the moment

4 Cf. ĪPVV II p. 343 sakrcchabdo bhedaniṣedhatātparye prayuktah; ibid. sakrdābhāsena deśakālasvarūpābhāsayojanaikyābhāsena. On sakrdābhāsa see also below p. 155.

5 Namely, both in themselves and as regards the cognitions that have them as object, or, as Abh. says, both in perception and ascertainment (see text note 369).

6 Utp. is referring here to the criterion adopted by the Buddhist logicians (Bh attributes it to tarkikādi) to ascertain the existence of a cause-effect relation. This criterion is based on the combination and interrelation of a certain number of perceptions and non-perceptions (cf. PVin II pp. 33-4, HB p. 4, SP 13, PVSV p. 22). Fire, for example, may be considered the cause of smoke since, fire having been perceived, smoke, not initially perceived, is perceived later. Dharmakīrti does not explicitly state the number of these perceptions and non-perceptions: though their division into five seems the most likely (this is what Dharmottara in the PVin-ṭīkā, Mādhava in the SDS, and Abh. in the ĪPVV understand), some commentators - Śākyamati, Karnakagomin, Arcata - only recognize three as being essential (see Steinkellner 1967: II. 97 n. 49). The dispute between these two views has been studied in Kajiyama 1963, where the Kāryakāranabhāvasiddhi of Jñānaśrīmitra (a trikayādin) has been translated.

I 7.4 in which they make known one after the other their respective objects, and are not able to display any mutual dependence7. - 4 -

5. The self-consciousness (svasamvittiḥ) which in memory is the proof of memory's very being, that same self-consciousness and nothing else establishes the subsistence of the former direct perception.

• Since [at the time of the memory] the self-consciousness of the former direct perception no longer exists, it is the self-consciousness of the memory and nothing else which establishes the existence of this former perception, as it does for the memory itself. This self-consciousness consists in the knowing subject characterized by the manifestation of a single identical object8. The status of effect - acting as logical reason - (kāryalingata) may not be attributed to memory since, as the manifestation of the former perception is no longer present, it cannot be said that a cause-effect relation has been established9. - 5 -

7 Here the allusion is to the difficulties inherent in the Buddhist conception of causality; the whole of the fourth āhnika of the Kriyādhikāra will be devoted to the discussion of this.

8 The argument that memory is the manifestation and proof of the existence and continuity of the knowing subject, already extensively dealt with earlier, is here incidentally taken up once again. It is in effect through memory that the subject may establish that link between perceptions and non-perceptions capable of proving the existence of the cause-effect relation. This connection, as has been seen in the previous kārikā, does not really pertain to the cognitions themselves, but is projected onto them by the subject. The continuity of self-consciousness thus reveals itself in memory. In fact, it is the very same svasamvedana which illuminates that illuminates, makes manifest, the former perception - now no longer existent - on which the memory is based. Abh. lucidly distinguishes this type of argumentation supporting the existence of the self, based on memory (insofar as it reveals the oneness of self-consciousness), from that also based on memory, put forward earlier (cf. IPV I pp. 363-4).

9 Utp.'s aim here is to reject a possible objection - namely that there is no need to resort to the work of a continuous svasamvedana, but that memory is enough in itself to connect up with the former perception, since by general consensus the effect can lead to the knowledge of the cause. What Utp. does not acknowledge is precisely the admissibility of such an inference, whose logical reason (linga) is represented by the nature of effect (kārya) attributed to the memory: between the perception that occurred in the past and the present memory there is no cause-effect relation, which presupposes a continuity (see text note 379 for the development of this argumentation). Therefore those who invoke memory as the unifier of cognitions which are in themselves separate implicitly finish up by involving precisely

6. Also the invalidating-invalidated relation (bādhyabādhakabhāvah) I 7.6 between cognitions which are restricted to themselves and do not contradict one another (svātmaniṣṭhāvirodhinām) may exist only by virtue of their resting on a single knowing subject.

• What contradiction can there be between cognitions which are only directed to the single manifestations that are their own (bhinnaśyābhāsa-mātraniṣṭhānām)10 and, consequently, how can one speak of an invalidating-invalidated relation with reference to them?11 On the contrary, this relation is admissible if they rest on a single knowing subject. - 6 -

7. [Someone might however object that] just as the cognition of an empty surface entails at the same time the cognition of the absence of the jar, so the cognition of mother-of-pearl entails the non-validity of the cognition of silver (rūpyajñānāpramātvavit)12.

that knowing subject they wanted to avoid, who, as we have seen, is identified with the continuity of self-consciousness on which memory is based.

10 Or also: «that insist only on themselves (svataḥ) - i.e. on their own self-awareness - and on the manifestation (ābhāsaḥ) that is their object», following ĪPVV II p. 371 svasmin visaye ātmani ca svasamvedane (also ĪPV I p. 365. which substitutes svarūpe for svasamvedane). This is an interpretation that does not basically differ from the text of the vrtti but constitutes the development of what is there implicit, apparently following the indications in the ṭikā (cf. ĪPVV II p. 372 vrttau vivṛtau ca svaś ca ābhāsaś ca iti svaś ca viśayaś ca iti yojyam). Or, Abh. continues, only the ābhāsa (the content of the cognition) may be considered mentioned in this expression, since the invalidation can concern only that and not the 'self-awareness' component (pp. 372-73 atha vā svasamvedanabhāge bādhāsambhavād ābhāsamātram eva sambhavadbādham iha nirdiṣṭam); in this case the svātmāḥ of the kārikā is to be understood in the sense of ābhāsa (p. 373 tadā ca sūtre 'pi svātmagrahaṇam ābhāsatātparyeṇa yojyam). Abh. (ĪPV I pp. 265-6) indicates yet another possible interpretation of the kārikā as a whole, which is really rather contrived, but is, in the end, not far removed from the general meaning considered so far. As regards the sense in which the expression svābhāsa is used in the logical schools of Buddhism (grāhakābhāsa and svasamvedana) cf. Hattori 1968: 101 ff.

11 Also the invalidating-invalidated relation is therefore possible only thanks to the subject's freedom to act on cognitions that are themselves unrelated and therefore incapable, except in a uselessly general way, of being in opposition to one another. According to the Śaiva view, an invalidating condition is one which interrupts the continuation of the result of another (ĪPVV II p. 372 yat yasya phalānuvṛttibhaṅgam karoti, tat tasya bādhakam; cf. Rastogi 1986) which presupposes precisely the continuity and oneness of the knowing subject. See also SD IV.15 ff.

12 This kārikā voices an objection expressed from the point of view of the Buddhist logicians (ĪPVV II p. 376 dharmottaropādhyāyādidarśitam), according to

139

I 7.7 - The cognition of the absence of the jar - 'on this surface there is no jar' - is precisely the cognition of the empty surface (kevala-bhūtalajñānam eva), since the empty surface appears as the absence of the jar. Analogously, one may say that the cognition of mother-of-pearl

whom the invalidating-invalidated cognition relation does not need any single subject on which to base itself, but it is the correct cognition that entails by itself alone, implicitly, the non-validity of every other different cognition of its object. The principle is the same - as pointed out in the kārikā - as the one underlying the explanation of the concept of non-being (abhāva) and non-perception (anupalabdhi) given by Dharmottara in the NBT. The cognition of the absence of the jar, though being in itself distinct from that of the empty surface, is essentially connected with it, just as determined cognition is linked to perception and represents its subsequent stage: the same cognitive act embraces both of them (cf. NBT pp. 122-123). Dharmottara here depends directly on Dharmakīrti, who had extensively debated the question in several of his works (PV, PVSV, HB, NB). The essential points of Dharmakīrti's conception - which he sharply distinguishes from his teacher Īśvarasena's adarśana-mātra (Steinkellner 1966: 75 ff; 1992: 315) - may be summarized as follows (I refer to HB pp. 21-28*). Non-perception is not a distinct pramāṇa connected with a particular prameya of its own (abhāva), as for instance Kumārila maintains. It consists in a positive perception of something that exists, a perception and object that however are different from the cognizer's expectation (anyopalabdhi, anyābhāva, vivaksitopalabdher anyatvāt) and in this way reveal an absence. From the positive perception of an 'other' thing one may pass to the cognition of the absence of the thing intended only on determinate conditions: first of all, that the two things possess an equal capacity to suscitate a certain cognition (avisiṣṭayogyatā) and that therefore they are necessarily associated in a single cognitive act, and. furthermore, that for the thing that is the object of negation there subsist all the conditions necessary for its perception (listed in NB II. 13-14). The cognition of the absence is immediate: the perception of the one thing alone entails the determination of the absence of the other (bhāvasiddhir evāparasyābhāvasiddhih). Seeing that a surface is empty - in the classical example - is knowing that the vase is not there (tasya kaivalyam aparasya vaikalyam). If this is possible, it is because it is the very nature of every cognition to define its object contextually by excluding the other (tat paricchinatti tadanyad vyavacchinatti). As Dharmakīrti states several times using different arguments, we are not dealing with an inferential process here, by which from the being there of the other we arrive (via mediation) at the not being there of the thing aimed at, because between the two there is no real sambandha on which the relation of linga-lingin may rest (na vai kutaścit sambandhād anyabhāvas tadabhāva-bhāvagamaka iṣṭah, kim tv anyabhāva eva tadabhāvah). The non-perception therefore directly reveals the absence of a thing (abhāva) and promotes the use (verbalization etc.) of this notion in practice (abhāvavyavahāra). Anupalabdhi may be used as a hetu to establish abhāvavyavahāra, but only when it is a question of illuminating and guiding a torpid intellect (mūdhapratipattau) that needs to arrive by degrees at a notion that is, on the contrary, immediate in a normal person.

140

is, at the same time, the cognition of the absence of silver, since there I 7.7 is no mutual identification between mother-of-pearl and silver. And therefore it is the very perception that constitutes by itself the invalidating cognition (pratyaksam bādhakam). - 7 -

8. That is inadmissible. From the cognition of the empty surface all that is proved is merely that this surface is not a jar (tasyāghatātmatā), and not the absence on it of a jar that is accessible to perception (upalabdhyyogyasya)13.

• From the cognition of an empty surface it is proved that this surface is not a jar, but not that on this substratum there is not a distinct jar, capable for its part [if it were there] of being the object of sight. - 8 -

13 Utp.'s reply is based on the distinction between two types of abhāva: tādātmyo 'the surface is not the jar' and vyatireko 'on that surface there is no jar' (the two terms recur in ĪPV and ĪPVV; the Navyanyāya would say samānādhikaraṇo-vyadhikaraṇābhāva or anyonyao-samsargābhāva, cf. Matilal 1968: 50, 116, 148; or tādātmyo - samsargābhāva, cf. Jha 1981: 184-187); in the first the possibility of perception has no part, in the second it is essential. The example proposed by the Buddhists would be valid only in the first case, but not in the case that interests them here, in which the negation is not generalized but concerns, according to their own formulation, an object characterized by hypothetical visibility (drśyatva). Its application to vyatirekābhāva is erroneous for atiprasanga (ĪPV I p. 374). The same point had already been made by Kumārila in the ŚV (Abhāvapariccheda, in particular vv. 38-39 and Pārthasārathimiśra's lucid explanation in the Nyāyaratnākara). Dharmakīrti had also foreseen this possible objection to his own theory: if the reference to the cognition that simultaneously defines its object and excludes that which is other may serve to account for the immediacy of the cognition of absence, what results, however, is a generalized negation concerning all that is other than the object effectively perceived, thus rendering impossible the application of those distinctions (ekajñānasamsargi, upalabdhiakṣanaprāpta) mentioned above (HB p. 26* yady ekaparicchedād evānyavyavacchedah siddhyati, sarvasyānyasyāviśeṣena tatrābhāvasiddhir bhavec, na tulyayogyāvastashyaiva; upalabdhiakṣanaprāptasyānu-palabdhir abhāvasādhanīti viśeṣaṇam ca na vaktavyam, anupalabdhiakṣanaprāptā-nām api tatra vyavacchedāt). Dharmakīrti replies by referring to the narrow sense in which the term 'other' is to be understood in the context of the anupalabdhi, as explicitly stated at the beginning (ibid. p. 22* aviśiṣṭayogyatā rūpayor ekajñāna-samsarginoh parasparāpeksām evānyatvam ihābhipretam). In conclusion, from the non-perception, as it has been defined, there follows the determination of the absence of something definite in a definite time and place (p. 27 tasmād yathokto-d evānupalambhāt kvacid kadācit kasyacid abhāvasiddhih).

I 7.9 9. The surface is always a separate and distinct reality (viviktam), since things are self-contained (svātmaniṣṭhiteh); and therefore how can the cognition of it sometimes (jātu) lead to establishing the not being there of a thing distinct from it?

• The surface is always separate from what is other than itself and so how can the cognition of it establish only in some cases (kadācid eva) the absence on it of a distinct entity, such as the jar14? The separation from a distinct jar (bhinna-ghata-viviktatā) might be an occasional aspect (kadācitkaṃ rūpam) of the surface only provided that also the association with that jar (ghatasahitatāpi) sometimes constitutes its own form. But this is not the case: there are always two things, each self-contained, and the association (sāhityam) between them is not another individual entity which, though distinct from them, comprises both of them (tadatik-tam ubhayātmakam)15. However [the objector replies], a single cognitive act in which the two manifestations appear associated (ubhayābhāsasam-sargātmakam) constitutes in itself the absence of another cognition in which only one of the two is manifested (ekābhasajñānāntarābhavāru-pākṣam ekajñānasamsargi kathyate).

14 Utp.'s objection serves to induce his opponent to play the last card he was saving, in a sense his most important one, that on which the Buddhist conception of the anupalabdhi rests, and which gives it its originality. In order that the non-perception of a thing A may define its absence, all the conditions of perceptibility must exist (both as regards the thing and the subject), and there must be the positive perception of a thing B (and this is the essential point) which enters into such a relation with A that a single cognitive act must necessarily grasp both of them (NBT p. 101 ekendriyajñānagrāhyam locanādipraṇidhānābhimukhaṃ vastudṛṣṭaṃ anyonyāpekṣam ekajñānasamsargi kathyate). If this cognitive act takes place and only B is perceived this means that A is absent. It is in this sense that the Buddhist equation - perception of B equals absence of A - must be understood (NBT p. 118 arthajñāne eva pratyakṣasya ghaṭasyābhāva ucyate). This is - as we have seen (kārikā 7) - perfectly homologous to the Buddhist explanation of bādhā: cognition of mother-of-pearl equals invalidation of cognition of silver.

15 What guides Utp.'s criticism, here as elsewhere, is his intent to underline the immobility of phenomenal reality (objects, cognitions taken in themselves etc.), in order to make the affirmation of an 'I' necessary, as the only source of that dynamism which is present at all levels of practical reality, and which renders it possible. In doing this he cannot but find himself continually disputing with the Buddhists, who in certain respects take the opposite path. Here he denies that sāhitya, ghatasahitata is something that may be perceived when space is perceived, as its nature; it is not an objective given, it is merely a mental construct (ĪPVV II p. 384 vikalpanamātram etat). This subject is dealt with extensively by Abh. following on from the țīkā (cf. ibid. p. 383).

142

pam)16. [A cognitive act] nevertheless [it is replied] delimits the object I 7.9

as being self-contained: the determination of the absence of the jar does

not occur as a direct consequence of the datum (na vastubalena), as is

the case, on the contrary, in the determination of the existence of the

place, which depends on the perception of the place. However [the ob-

ject or says] it can be arrived at through the absence of the effect, that

is, through the absence of a single cognition in which the two things are

manifested17. Not even this is admissible: it would be a cognition arrived

at through mediation (vyavadhānena pratītiḥ), whereas the establishment

of the [empty] place derives immediately from the sight of it (pradeśa-

darśanād eva)18. - 9 -

10. However, on the surface in question there is a beam of light, or

• for the blindman - a soft, warm, etc. tactile sensation: it is these per-

ceptions that establish on this surface the non-identification of their ob-

ject with the jar19.

16 The opponent shows that he is perfectly aware (who can be more so than a

Buddhist?) that sāhitya does not belong to the sphere of things, but to the sphere of

cognitions (cf. e.g. PV I. 87 samsrjyante na bhidyante svato 'rthāḥ pāramārthikāḥ |

rūpam ekam anekam ca teṣu buddher upaplavah) and reformulates his position in

prativiṣayam paryavasyaḥ ...) - 385. Dharmakīrti for his part might say that if such

an 'association' (sahitā, samsarga) is not a given fact (vastu), it is however present

in cognition and it is reverberated by cognition on the things that gave rise to it (in

the background lies the principle expounded in PV I. 109 ekapratyavamarśasya

hetvyād dhīr abhedinī | ekadhīhetubhāvena yaktiñām apy abhinnatā //).

17 The Buddhist opponent, put on the spot, thus ends up by presenting the nega-

tive judgement «on the surface there is no jar» as an inference, coinciding substan-

tially with the first of the eleven types described in NB II. 30 ff. (reduced to three

in IIB), svabhāvānupalabdhiḥ: as Kamalaśīla points out in TSP p. 587 and as results

precisely from its formulation in the vrtti, it is resolved in a kāryānupalabdhi. The

Buddhist has therefore illicitly changed plane and Utp. promptly points this out. Bud-

dhist logic is perfectly aware of the dual level of the negative judgement (NBT p.

123 ato drśyānupalambho 'bhāvajñānam krtam pravartayati na tu akṛtam karoṣi

abhāvamiścayo 'nupalambhāt pravṛtto 'pi pratyakṣeṇa krto 'nupalambhena pravartita

ukta iti) and, as has been seen (cf. above n. 12), admits the presentation of the nega-

tion in terms of inference only to demonstrate the abhāvāryavahāra in certain cases.

This final shift in perspective is thus equivalent to a surrender.

18 This is also the Buddhist position restated so often (cf. e.g. HB p. 27*

darśanānantaram vyavadhānena vinedam astīdām tu nāsīti), that the Buddhist in-

terlocutor ended up, however, by contradicting.

19 After criticizing the opposing thesis, Utp. now puts forward his own; it is

'unknown to the ancients' (ĪPV I p. 376 cirantanair apadiṣṭam; Bh. ibid., ciran-

143

I 7.10

– The verbal formulation and relative behaviour (vyavahartum) ‘there

is the absence of the jar on this surface, there is no jar’ can be legitimately

brought about by sight or another sense (ālokādi)20 only once they have

directly perceived in the place in question a beam of light or, in the

shadows, a warm, soft etc. tactile sensation constituting in the first case

the absence of the shape of the jar, and in the second of its tangible form.

– 10 –

11. A piśāca is a different thing from light (anālokaḥ); [nevertheless]

being invisible, he may reside within the light, just as he may be within

the surface in question: from every point of view (sarvathā) his presence

may not be denied.

– On the basis of what has been said, the fact that light is a different

thing from the piśāca does not entail the negation of its presence within

the light. The piśāca is, in fact, invisible and, though it is different from

the light, it can reside within the light, in the same way in which it can

without any difficulty also reside within a ball of clay. Therefore the ab-

sence of the piśāca, as he is not visible, is not proved both from the

point of view of the opposing doctrines and of our own21. – 11 –

tanair hi śuddhabhūtaladrṣṭer evābhāvasādhakatvam referring to the masters of

Buddhist logic, whose conception was here the object of his criticism. It is not the

cognition of the empty surface in itself that may lead to the establishment of the ab-

sence of the jar, but the perception, through any of the senses, of something else

that must necessarily be in its place (this principle is categorically expounded by

Abh., ĪPV I p. 376 iha bhāva eva bhāvāntarasya abhāva). In this way the presence

of all the factors essential for perception to take place (upalabdhi-lakṣaṇa) is ascer-

tained in the subject – which the Buddhists also demand (drśya = upalabdhilaksa-

naprāpta) –, since a visual or auditory etc. perception actually takes place and may,

by perceiving a particular object, legitimately imply the non-perception of another

(the object whose absence one wishes to establish). All without unduly confusing

ādhāra and ādheya (ĪPVV II p. 382 ekasyaiva hi katham ādhārādheyatā): the per-

ception required to establish an absence is not that of the surface (ādhāra), but the

space above it, which may or may not be occupied by a jar etc. (ādheya; cf. ibid.

p. 388).

20 It is to be noted that here the word āloka is used in the same sentence first

in the sense of ‘light’, then in that of ‘sight’. [I was mistaken; see above Foreword]

21 In the opponent’s eyes, the Śaiva conception runs the risk of making the

vyatirekābhāva, which is essential in practical reality, become an undifferentiated

tādātmyābhāva, consequently including even the negations of what is imperceptible

by nature. Indeed, this verse confirms that the Śaiva and Buddhist theories fully agree

on this point.

12. Thus the cognition of mother-of-pearl may appear as the non-being of the cognition of silver; however, it cannot reveal the non-validity of the former cognition of silver 22.

• The cognition of mother-of-pearl in itself appears as the non-being of the cognition of silver; but through the direct cognition of mother of pearl which takes place at that particular moment (tadāntara°) one does not obtain the non-validity (aprāmānyam) of a distinct cognition of silver that occurred in the past. - 12 -

13. Not even inference can account for the invalidation of a cognition, due to the property-possessor not being established (dharmyasid-dheh)23; on the contrary the invalidation, established on the basis of everyone's inner experience (svasamvedanasiddhā)24, is coherently explained (yuktā) as deriving from a unitary knowing subject.

• At the moment of the cognition of mother-of-pearl the former cognition of silver no longer subsists. Therefore, since the property-possessor is not established, the invalidation may not be explained in terms of inference25. On the contrary, it is explained (sidhyati) by maintaining that it is in the self-awareness (svasamvedane) - consisting in the unitary

22 The discussion now returns to the invalidating-invalidated cognition relation, after this long digression on the concept of negation has furnished the elements needed to clarify the crux of this relation and its implications.

23 The preceding kārikās have shown that, remaining in the sphere of the Buddhist conception, perception alone is not able to account for the invalidation (bādhā) of a cognition. Nor, this kārikā adds, would a possible Buddhist recourse to inference achieve this purpose. 'Also' (api), as Abh. suggests, may also refer to dharmyasid-dher; in this case there would be an allusion to the asiddhi of the reason adduced, as well as of the property-possessor. The dharmyasiddhi is a form of logical fallacy (hetvābhāsa), included by Dharmakīrti in the asiddha type (NB III.65 dharmyasid-dhāv apy asiddhah - yathā sarvagata āmeti sādhye sarvatropalabhyamānagunatvam).

On the meaning and translation of the term dharmin see Kitagawa 1960: 387.

24 I translate it in this way on the basis of the immediate meaning, which - I believe - remains valid, even if the vrtti chooses to make a so-to-speak concentric meaning emerge in its place, which contains the analytic explanation of bādhā.

25 In fact, the inference would be as follows: the cognition of silver (dharmin) is faulty (dharma) because of the cognition of mother-of-pearl, or of the cognition 'it is not silver', or also because the thing in question becomes the object of two contradictory cognitions (three possible hetus). The fault, as the vrtti makes clear, lies in the fact that at the moment of the inference the cognition of silver, produced in the past and momentary according to the Buddhists, no longer exists. The fault of asiddhi may also be attributed to the possible hetus (cf. ĪPVV II p. 397).

I7.13 knowing subject - that the relation constituted by the two cognitions referring to a single object is manifested (ekadeśāvaṣṭambhyubhayajñā-namayasambandhabhāsanāt). The congruence with reality which is subsequently ascertained (paścātsamvādah), and 'which causes one to say that the one cognition is valid and the other is not, can take place insofar as the former cognition is also manifested in the self-awareness of the present direct perception. The determination of the congruence also depends on the unitary knowing subject26. - 13 -

14. Thus ordinary worldly activity, whether pure or impure27, is experienced as resting on the Lord associated (khacite)28 with the manifestation of greatly differentiated objective realities.

• All practical activity, based on differentiation due to the power of māyā, is pure for those who possess such knowledge, impure for those who are blind because of nescience29. It is possible to grasp, through direct experience (sambhāvyate 'nubhāvena), that all this daily practical activity is founded on the Blessed One30, engaged in the manifestation of the various differentiated realities. - 14 -

26 Two stages may be distinguished in the invalidating process: firstly a conflict (virodha) between two cognitions, and then the ascertainment of their congruence with reality (samvāda) on the basis of their ability to produce the expected results etc. Virodha is possible. Abh. continues on the basis of the tīkā (ĪPVV II pp. 403-4), only if there is a single and unitary subject of the two cognitions (virodhasya kāranam yad ekāśrayatvam tat sati ekasmin pramātari jñeyam); similarly, samvāda, taking place only after (paratah) the cognitions, requires the unifying action of the subject. Therefore, since the invalidation depends on virodha and samvāda and these in turn depend on a single cognizer, it follows that the invalidation also depends on this.

27 In the activity of everyday life (vyavahāra) there is an impure aspect, consisting of sales, purchases, quarrels etc., in which the vulgar man is immersed, and a pure aspect, consisting of teaching, meditation, initiation, ritual ceremonies, pertaining to the man who has entered on the path of freedom. The value of this division is, however, far from being absolute and diminishes with the gradual elevation of consciousness.

28 Glossed (ĪPV I p. 390, ĪPVV II p. 404) with svarūpānanyathābhāvena uparakte.

29 I.e. sakala, pralayākala and vijñānākala (ibid. p. 406).

30 Practical reality is possible as such in that it is rooted in the Supreme Consciousness and this may be known through direct experience itself (i.e. not through inference). This experience is accessible also to obtuse souls incapable of arriving at it immediately, through meditation and the repeated practice of this teaching. (cf. ibid. p. 405).

CHAPTER VIII

1. However (kevalam)1, the manifestations sometimes depend on the direct sensory experience which takes place at that very moment (tātkālikākṣasāmaksya°) at other times they do not, as in the case of the blindman, darkness etc.2

• Sometimes the manifestations are conveyed by the present direct experience, thus determining the cognition that may be formulated as 'this is a jar'; on the contrary, in the case of darkness etc. they derive from a former perception. - 1 -

2. On the contrary, there is no difference (viśeṣaḥ) in the intrinsic existence (sattāyām) of the manifestation of the object within the sphere of the mental constructs regarding future, present or past objects.

• In the mental constructs that follow direct perception (pratyakṣaprṣṭhapātiṣu), such as memory and conjecture, or in the others that are not dependent on it (svatantreṣu), concerning the three times, the manifestation of the object remains internally the same3. - 2 -

3. Even if pleasure etc. and the causes of the states of pleasure etc. are real (vastuṣu) and their manifestation exists (sadbhāve), nevertheless,

1 Here kevalam introduces a qualification of the general statement that concluded the previous āhnika: the whole fabric of phenomenal existence has its reality within the Lord.

2 This explains how the ābhāsas differ in clearness or vividness (sphuṭatva; ĪPV I p. 402-3). The vividness of an ābhāsa depends on its association with the ābhāsa 'direct experience'. This is missing, therefore, when the subject is in the dark or blind (but not born blind) and is replaced by its reconstruction on the basis of former direct experiences. On the contrary, this is altogether absent in the case of a person who is born blind, unless one acknowledges (ĪPVV II p. 408) that it occurs through the latent traces left by former existences.

3 The ābhāsa 'blue' is essentially the same whether it appears associated with the ābhāsa 'I see' or 'I remember' or 'I imagine'. Similarly the ābhāsa 'seeing' remains essentially the same even if it is associated in turn with the āhāsa 'blue', 'jar' etc. In the end everything, without distinction, takes on the form 'I' (I perceive, I remember, I imagine.the colour blue; cf. ĪPVV II p. 409).

I 8.3 because they belong to the past, an effective similar4 condition of pleasure is not5 produced (tathāsthitiḥ).

– The manifestations of pleasure, pain etc. and those of the factors that cause them, though they always exist internally6, do not bring about a state of joy etc. because, due to their belonging to the past [in the case in question], they have no external existence at the present moment and viśiṣṭānām) are able to produce the above mentioned state. – 3 –

4. If, however, pleasure etc. is intensely reproduced (gāḍham ullikhyamāne) through mental representation, then that same state (tathāsthitiḥ) comes about in the subject, in the same way7, since he vividly experiences that pleasure.

– Reproduced through an autonomous mental representation8 with particularly intense effort, pleasure etc. becomes completely vivid and causes, in the same way, joy (vikāsa9) etc. – 4 –

5. External existence (bāhyatā) is to be considered an accessory condition (upādhiḥ) and not the very essence (ātmā) of the manifestations of being and non-being. These, therefore, insofar as they are inner manifestations, always exist.

4 Similar to that experienced in the past, when that particular pleasure and its causes were present.

5 The negative, required by meaning and supported by all three commentaries, is based on reading the compound tathāsthitiḥ as tathā-asthitiḥ, thus attributing to the negative particle the sense of prasajyapratiṣedha (the only one to say so explicitly is Bh I p. 406). Tathā may also not be considered part of the compound that follows, but tathāsthitiḥ must, in any case, appear joined (as it stands for tathā asthitiḥ); however it is only so in Bh's yojanā and never in the kārikās or quotations in ĪPV and ĪPVV.

6 United with the I.

7 I.e. like that deriving from an 'external' pleasure.

8 Utp. specifies that it is vikalpa 'autonomous' in order to exclude memory, which – like determinate knowledge or judgement – is strictly dependent on direct perception (cf. text note 425). The role of memory is limited to supplying the object for the mental evocation, which having freed it from its ties with a particular past time, place and subject, brings it so vividly back to consciousness that it appears as if it were altogether new (see ibid. p. 411; cf. also kārikā I. 4. 2).

9 On vikāsa ('expansion', joy) see Gnoli 1968: 46-47.

148

• All the various manifestations, concerning both being and non-being, even if they do not appear as external, nonetheless have an existence (sattāsty eva), for instance in memory and so on. The fact that they exist externally is only an accessory condition and not their own form. The same is true of the manifestation of 'non-being', even though this [only] has an internal existence (antahsattāyām api), since it does not exist externally10. - 5 -

6. These manifestations being internal (āntaratvāt), insofar as they are one with the knowing subject, cannot have causal efficiency which presupposes differentiation. Even when the condition of externality [of the manifestations as regards the knowing subject] exists, the causal efficiency is differentiated on the basis of the differentiation of the manifestations (bhinnābhāsabhedatah).

• Even if all the various manifestations, such as blue, pleasure etc. permanently exist within, nevertheless, precisely because they consist only in the knowing subject, they do not possess causal efficiency, which is based on the differentiation typical of the relation of cause and effect etc. Even if there is the condition of differentiation from the knowing subject, causal efficiency, being linked to the manifestations, is seen to be differentiated according to whether it pertains to the form and so on, conforming with the diversity of the manifestations (e.g. their being mental, sensory etc.)11. - 6 -

7. Insofar as they are essentially constituted by consciousness (cinmayatve) the manifestations permanently reside internally; insofar as they are manifested as external owing to the power of māyā, they also exist externally.

• As their own form is consciousness the manifestations permanently reside internally. One speaks of external manifestation when the things, remaining unseparated from light (prakāśāvyatirikteṣu) are made

10 Cf. ŚDVr p. 130 tad avastv api prakāśamānam cidrūpam eva, kevalam bāhye prakāśanābhāvād avastu ucyate. See ŚD III.78ab.

11 In other words, this is a necessary but not a sufficient condition: in order for a manifestation to produce a determinate effect a series of specifications and relations with other ābhāsas are also required. On causal efficiency (arthakriyā) see below p. 166.

I 8.7 by the power of māyā to appear (prakāśyamāneṣu)12 externally as the object of direct perception13. But even when they are directly perceptible (tadāpi) they do not cease, for this reason, in the final analysis to be internal. On the contrary, in the case of intending to create and so on (sṛṣṭyādau) even things come to be internal. – 7 –

8. Even the mental representation (ullekhaḥ) which occurs in the sphere of discursive thought (vikalpac) is external, because it appears in terms of separation (prthakprathāḥ). Indeed, internality is unity with the

12 By using the causative form the vṛtti seeks to underline the heteronomy (pāratantrya) of the manifestations of objective reality which ordinary experience dominated by the power of māyā tends to consider extraneous to and independent of consciousness. On the other hand, grammatical speculation is able to place the correct limits on a statement such as this, which, if taken literally, runs the risk of responsibility for the action of becoming manifest, of shining (prakāś-) rested solely with the primary subject of the causative (hetu) or, in other words, with consciousness, this would mean that the subject were completely extraneous to the reality of light – and in that case truly bāhya, or from the Saiva point of view, a mere nothing. But as Bhartrhari says (cf. VP III. 7. 122-128), the subject of the action induced, in the causative form, can only be someone who possesses the requirements to act as a free agent of it (cf. the passage in Abh.'s lost Śivadrṣṭyalocana quoted in PTV p. 225, preryo 'pi sa bhaved yasya saktatā nāma vidyate) (Torella 1987: 157-58). The meaning of the causative does not go beyond the, more or less, cogent stimulation of the hetu so that the kartr performs the action in question: only what is in itself able to shine or, in other words, that which is essentially light, may be made to appear or shine (cf. also TĀ X. 38-45a). The status of the ābhāsa is thus to be found in this delicate balance between svātantrya and pāratantrya.

13 Abh. (ĪPVV II pp. 21-2), following the thread of the śikā, deals at length with the various implications contained in the expression 'as directly perceptible' (pratyakṣatvena), and it begins by imagining the surprise of those who expected to see also the vikalpas (particularly those termed 'independent') mentioned in this context. Utp. is here primarily concerned with underlining what contrasts with what is commonly accepted – in this case claiming the ultimate internality of what is considered external par excellence (the object of pratyakṣa) –, whereas the internality of the vikalpas is generally taken for granted and thus it is not worth expressly stating it. On the contrary, what has to be demonstrated – and this once again conflicts with the general opinion – is precisely the real externality of the vikalpas and this is done in the following stanza. Later this externality ultimately comes to rest on a fundamental internality, which is common to every reality. But pratyakṣatvena in the vṛtti may also have another function: this expression may re'er to all types of ābhāsa – whether sensory or mental – being directly present to the consciousness, and thus to their indiscriminate satyatā.

knowing subject (pramātraikātmyam), externality is separation from I 8.8 him14.

• The mental representation of the jar etc. in the sphere of discursive thought, though not the object of the senses such as sight, is non-etheless external insofar as it is manifested as separate [from the subject]. Indeed, internality is the reflective awareness 'I' (ahamvimarśah), externality is the reflective awareness 'this'15. Thus for things - such as the jar and so on - there are two kinds of externality, that is, they are the object of cognition through both the external and internal senses: for pleasure and so on there is one, in that they are cognised only by the internal senses. - 8 -

9. Owing to the will of the Lord mental representations and pleasure etc. are manifested to consciousness as external, in the same way as [also owing to the will of the Lord] the manifestation - in the form of direct perception - of what pertains to the planes of realities perceptible by the eye etc. (akṣyādibhuvām) is external.

• The manifestation (prakāśah)16 of mental representations and pleasure, pain, shame etc. that takes place through the mind (bauddhah)17 is no different from that which, having the form of direct perception due to the power of the Lord18, concerns those realities that can be perceived by the eye, such as shapes etc. - 9 -

14 One must keep in mind the background against which this discussion of the externality of the ābhāsa in direct perception and in vikalpa takes place. Indeed externality has been declared the sine qua non condition on which the existence of causal efficiency depends (and it cannot be denied that the vikalpas, too, are able to produce effects); and this dependence is invoked in order to refute the satyatā-arthakriyākāritvam identification, upheld by the Buddhists. The vrtti distinguishes between two forms of this general externality, according to whether the ābhāsa is the object of both the external and internal senses, or internal only.

15 Or to put it more clearly, becoming conscious of the object in terms of 'I' (unity, internality) or of 'this' (separation, externality).

16 Once again the problem arises of how to translate the term prakāśa, whilst preserving its density of meaning (light, manifestation, presence to consciousness and therefore also coming to light, perception, knowledge etc.).

17 If this knowing is called bauddha - in spite of the fact, a hypothetical objector notes, the actual grāhaka is in this case the manas - it is because it is concluded by the act of the niścaya, which is pertinent to the buddhi (ĪPV/ ĪPvv II pp. 430-31).

18 Īśvara is the ultimate source of all manifestations. This is so not only in the case of those that are the object of perception as external objects, which appear more

151

I 8.10 10. On the basis of what has been said (tad) without a unity of cognitions ordinary worldly activity would not be possible. It is on the unity of light that the unity of the various cognitions is based and this unitary light is precisely the one knowing subject. This has been definitively established.

– Indeed, practical activity is based on the unification (°anusamdhānena) of the various cognitions, which are in themselves separate from one another, and the unification of the cognitions is constituted by the light-principle, which is unitary in itself. This unitary light is precisely the knowing subject, one, called the Supreme Self. – 10 –

11. And precisely this knowing subject is Maheśvara, insofar as he is characterized by indefectible reflective awareness (vimrśattvena niyatena). In fact, in God this awareness is pure knowledge and action.

– Maheśvara is the supreme self, whose nature is consciousness, with his own primary indefectible nature constituted by reflective awareness. In this consciousness-principle, called Śiva, whose essence is everything, there is absolutely pure knowledge and action, being none other than the reflective awareness 'I' (ahamvimarśanam eva). In Īśvara, on the contrary, knowledge and action, connected as they are with an object assumed to be partly differentiated (bhinnābhinna°), are to be considered partially pure (śuddhāśuddhe). Referring then to an objective reality assumed to be differentiated, as happens on the plane of the individual subject, knowledge and action, informed respectively by sattva and rajas, or in other words by light and activity, contracted by tamas, are called impure.

– 11 –

as independent from the limited subject, but also in the case of those that might seem solely determined by the individual's will, such as mental representations, fantasies etc. Similarly, as Somānanda says (see ŚD I. 44 and vrtti), the jar is produced by the potter but only insofar as his creativity is contained within the all-pervading creativity of the Lord.

SECTION II. ACTION

CHAPTER I

1. Therefore, also the objection raised earlier - 'an action that is both unitary and successive, belonging to a unitary subject is not admissible'1 - was rejected through the demonstration of the existence of the unitary [consciousness].

• Through the demonstration of the unitary consciousness-principle the activity related to the unitary subject is proved unitary. Thus even the objections against action are found to be confuted. - 1 -

2. Succession pertains to ordinary action, which is dependent on the power of Time; it is not, however, admissible for divine eternal action, as it is not for the Lord.

• The action of entities that, due to the power of māyā, appear differentiated, is successive, being dependent on the power of Time2; but that action, informed by the awareness of the self, without beginning or end, which is the very essence of the Lord, is not. - 2 -

3. Time is the transit of the sun etc., or the birth of this or that flower and so on, or also heat and cold. Or time, in reality, is nothing but the succession (krama eva) characterized by these elements.

1 This is the Buddhist objection expressed in kārikā I. 2.9.

2 On the power of Time as the cause of the passing from the One to the many the 'locus classicus' is VP I. 3 and vrtti; cf. also III. 9. 46. The distinction between the working of māyāśakti (the differentiation between things, and between things and the subject) and kālaśakti (the introduction of succession into activity) is found in almost the same terms in Hclarāja's commentary on VP III. 3. 83 (akrame 'pi brahmani bhedāvabhāsanam avidyākrtam / tatraiva kramāvabhāsanam kālākhyasvātantryaśaktikrtam); cf. Akulujkar 1970: 122 n. 39; Iyer 1969: 125-26. Abh. (ĪPVV III p. 10) synthetically defines the power of Time in the following words: bhagavata eva ābhāsānābhāsaprathanasamarthatā.

II 1.3

• Time consists of various actions that are common knowledge in everyday experience3, or of cold and so on. Or rather, time is the succession itself - characterized by these - which is an accessory qualification (upādhi) of all the various realities in their being manifested as differentiated; time is nothing but this succession, because it is through it that temporal differentiations are established. - 3 -

4. Succession is based on differentiation and this in turn derives from the existence or non-existence of a certain manifestation (ābhāsasadasattvatah). The existence or non-existence of manifestations depends on the Lord, who is he who determines their multiform appearance.

• The cause of succession in things is the very action of the Lord, directed to the multiform appearance of the manifestations differentiated in their nature and mutually exclusive4. - 4 -

5. Through the variety of physical forms he causes spatial succession to appear; through the manifestation of the variety of actions the Lord also causes temporal succession to appear5.

• In the case of many entities, there is spatial succession through the manifestation of mutual diversity and temporal succession through actions. However, with reference to a single entity there is only temporal succession due to different actions such as birth, existence, change etc.6 - 5 -

3 The same idea (that the name 'time' is given to those realities able to act as a reference point, or 'measure' for the others) is found expressed in VP III. 9.77 kriyāntarapari cc hedapravrttā yā kriyām prati / nirjñātaparimānā sā kāla ity abhidhīyate //. 'That (activity) the measure (of time) of which is well-known and which is employed in measuring (the duration of) any oyster activity, with regard to (any other) activity, is called time' (transl. Sharma 1972: 101).

4 This is my translation of the expression °anyonyābhāsasūnya°, guided also by a passage in ĪPV (II p. 16), which recurs in almost the same terms in ĪPVV. Different manifestations may implicate each other, referring to a single substratum, such as 'red', 'jar' etc. or may be completely distinct from each other such as 'jar', 'cloth': obviously not in the first case, but not even in the second - where the two things, though unrelated, may coexist in a single cognitive act (yugapad vaikajñānena sphuranāt Bh II p. 16) - is there succession. On the contrary, succession is only possible if the manifestations are such that the presence of the one necessarily excludes that of the other, as in the case of autumn and winter.

5 On kāla / kriyā and deśa / mūrti see VP I.1 Vrtti pp. 2, 9 (and Paddhati pp. 5, 9); see also VP III.7.153 and Helārāja's commentary; Prakīrṇakaprakāśa I p. 117.

6 Together with growth, decline and disappearance they constitute the six modifications of existence (bhāvavikāra; cf. Nirukta I.2.8-9).

154

6. In all things (sarvatra) the diversity of the manifestations is the source of temporal succession for those knowing subjects such as the void etc., whose light is discontinuous (vicchinnabhāsah), but not for the knowing subject who shines once and for ever (sakrt).

• In every object diversified manifestation determines the appearance of temporal succession [exclusively] for the knowing subject who identifies himself with the void, the body and so on. Indeed, apart from memory, what was manifested in the past (prākkālah) does not shine in the same manner at the present moment7, and it is with reference to his own present (svavartamānāpekṣayā) that the subject may use the notions of past and future in everyday experience. On the contrary, for the [true] knowing subject who shines uninterruptedly once and for ever (sakrdvibhātasya)8, there is no temporal differentiation, neither in himself nor as regards objects, since, as the shining is never interrupted, it is not possible to speak of repetition. - 6 -

7. Also spatial succession in things is manifested [only] to the limited knowing subject; on the contrary, to the unlimited subject things appear full of his own self, like his own self.

• With reference to the limited subject [only] things appear differentiated from the subject and from each other, and they are said to be far away etc., for the Lord, on the contrary - his essence being solely light - cannot be non-light (aprakāśasambhavāt) with respect to anything, not even to an atom. The separation or distance from himself, or between things, does not appear to him who is light and only light9. - 7 -

7 That is, connected with preceding manifestations of the limited subject (cf. text note 17). Only through memory can the former shining of the object be revived, in some way; but this is a totally different question, already dealt with at lenght above (Jñānādhikāra, āhnikas III-IV).

8 This is a reference - through a śaiva text (Sārasvatasamgraha according to ĪPVV III p. 23, Śivasūtra according to Spandasamdōha p. 25) - to sakrdvibhāta in Chāndogya Up. VIII.4.2. later taken up again by Gaudapāda in Māndūkyakārikā III.36. IV.81; see also PTV p. 199. Consciousness shines 'once only' (this is the primary meaning of sakrt ), but as this shining is never interrupted the expression may also be taken in its secondary meaning of 'constantly, always, once and for ever (cf. Abh.'s explanation in ĪPVV III pp. 21-22; see also ĪPVV II p. 262 sakrcchabdena tu ekavāravācinā sadārthaparyāyena vā vicchedasaṅkā nivāritā). Śankara similarly glosses sakrt in the passages quoted above with satatam.

9 Lit. 'a compact mass of light' (prakāśaghanasya).

II 1.8 8. Nevertheless, the manifestation of such a differentiation between subject and object of knowledge is the very creative power (nirmāṇaśaktiḥ) of the Lord (īśituḥ), who knows thus.

• The fact that the various entities existing as subject and object of knowledge are determined as differentiated from each other and from Parameśvara (or from the self), is [due to] the creative power of Parameśvara, who, yet, thus shines in absolute fullness, neither does his own permanent form ever in any way cease to exist as a consequence of this determination. - 8 -

156

II 2.1

Chapter II

1. The ideas of action, relation, universal, substance, space and time1,

which are based on unity and multiplicity, are to be considered real

(satyāh), because of their permanence and efficacy (sthairyopayogābhyām).

– The ideas of relation etc. are also real manifestations, just like

action2, even if they regard both unity and multiplicity [at the same

time]3. They can, in fact, be employed permanently with determinate

practical results, and as such they are undeniable4. – 1 –

1This second āhnika aims at establishing that the categories acknowledged by

the bāhyavādins (the reference is particularly to the Nyāya-Vaiśeṣika) are acceptable

only if seen from the Śaiva viewpoint. Otherwise they do not stand up to criticism

(and the criticism par excellence is that of the Buddhist logicians). Cf. Torella 1992:

328-29.

2The demonstration of the reality of action was implicit in that of the power

of action discussed earlier. In the ṭīkā (cf. ÍPVV III pp. 36-37) Utp. refers to

three different interpretations of the compound satyābhāsa – as karmadhāraya,

saṣṭhītatpuruṣa and bahuvrīhi.

3The principal objection the Buddhists make to these categories, which they

place in the sphere of relative truth, is precisely that of viruddhadharmādhyāsa.

4The truth of a cognition is established here precisely by its not being contradicted

by a subsequent cognition (sthairya) and by its leading to expected practical

results (upayoga); see Abh.’s lucid formulation (ÍPVV III p. 29), satya eva yatah

sthiro bādhakenānunmūlyamānāvimarśaḥ samvādavāmś ca abhisamhitāyām grāma-

prāptilakṣaṇāyām kriyāyām upayogī. The key-term samvāda may be applied both to

sthairya and to upayoga (cf. ibid. p. 69, asyām vyākhyāyām bhāvasya arthakriyām

kurvata ekānekarūpatā ... iti kartavyatā. – ity evambhūta upayoga uktaḥ, sthairyam

tu samvādaḥ; uttaravyākhyāne tu sthairyam bādhakābhāvah, upayogas tu samvādah

iti bheda iti). This may be compared with one of the classical explanations of the

term, that given by Dharmottara (NBT p. 17, avisamvādakam jñānam samyagjñānam

/ loke ca pūrvam upadarsítam artham prāpayan samvādaka ucyaate, tadvat jñānam

api svayam pradarśitam artham prāpayat samvādakam ucyate (behind which lies

Dharmakīrti, PV II. 1 ... arthakriyāsthitih / avisamvādanam...). On samvāda and

related terms, such as avisamvādin etc. (and their different acceptions in the Buddhist

exegetical tradition) see Katsura 1984: 222-24, Dreyfus 1991, Franco 1991: 49-50, van

Bijlert 1989: 122-28.

157

II 2.2 2. In action etc. (tatra)5 there is one internal reality (āntaram tattvam); this, once it has become the object of sensory knowledge, becomes manifold depending on place, time and the specific shapes it assumes.

– The essential reality (tattvam) – which is by nature absolutely undivided (abhinnam eva) – becomes one and manifold because of the division into internal and external manifestation6. [It becomes manifold] because of the multiplicity of the manifestations of the individual realities ("svalakṣaṇa"), constituted as they are by the different combination of external manifestations such as place, time, shape. – 2 –

3. The mind (manas), performing the function of determination following [the direct perception] (anuvyavasāyi)7, brings about conceptual

5 The vrtti does not indicate the interpretation to be given of the word tatra. Abh. in his two commentaries suggests three interpretations that he considers equally possible: 'in action etc.', 'given the reality of action etc.', 'as regards the unity and multiplicity of action etc.' The țikā, from which Abh. quotes the expression evam sati (probably referring to tatra) glossed with vidyāmāne kriyādau tadgate ca satyatve (ĪPVV III p. 45), seems to indicate the second interpretation.

6 A reality, when it is completely identified with consciousness, in other words, when it is only consciousness (cinmātratayā), may not even strictly be referred to as one since the unity-multiplicity opposition only arises at a later stage. This is seen from the vrtti, from its paraphrase in ĪPVV (see text note 36) and from what Abh. states a few lines earlier in ĪPVV III p. 45 nīlādinām bhāvānām pāramārthikam yad āntaram samvinmātralakṣaṇam tattvam, tatra nā kalanā kācid abhedaghanatvād ekatāyā api tatra abhāvāt. The kārikā and the vrtti become clearer once various levels of meaning have been recognized in the word āntaram. The primary meaning of āntaram is the absolute reality identified with the I; it coincides with the first of the two types of āntaram spoken of in ĪPVV II p. 416 (āntaratvam dvidhā ahan-tayā samskāratayā ca), later articulated (pp. 416-417) and dislocated along the scale of the principles (the highest āntaratvam corresponds to the plane of Parameśvara): it is called abhinnam in the vrtti. This absolute āntaratvam constitutes the ultimate foundation from which unity and multiplicity come into being on the phenomenal level. Unity is understood as a relative āntaratvam, which may be taken in more than one way (see ĪPVV III p. 45 yat tu āntaram ābhāsāntarānāmiśram antarangam, āntaram ca antahkaratvavedyam tat tattvam ekam, tad ekam), and multiplicity is knowability on the part of the external senses, temporality, spatiality etc. Thus it becomes one-many (ekānekam).

7 This is apparently the meaning indicated by ĪPV II pp. 43-44. aindriyake nir-vikalpake ... / paścādbhāvinam vyavasāyam niścayātmakam vikalpakam anuvyavasāya-śabdācyam. According to ĪPVV III p. 46 anuvyavasāya is the mental operation that occurs after the determinate awareness of the direct experience (anuvyavasāyam sākṣātkāvimarśasya paścādbhāvinim vimarśam ...). In any case, the term is not used here in the particular sense attributed to it in the Nyāya.

elaborations such as action and so on, which are based on these two8 II 2.3 and substantiated by the activity of the knowing subject9.

• The conceptual elaborations such as action etc., carried out by the mind (mānasyah) and regarding both the external and internal plane - the mind being half way between them - are essentially the activity of the knowing subject (pramātṛvyāpārarūpāḥ). - 3 -

4. Things that are self-contained (svātmanisthāḥ)10 and are manifested separately (viviktābhāḥ), possess a unity in the sense of mutual connection within the unitary knowing subject. This is the basis of the ideas of relation.

• The various ideas of relation (sambandha°) such as that expressed in the phrase 'the king's servant', rest on the unity which derives from the connection on the internal plane (antaḥ samanvayāt) and the differentiation of the two related terms externally11. - 4 -

5. The conceptual elaborations of the manifestations of 'universal' and 'individual substance' also rest externally on unity, as well as on the differentiation of the individual realities and parts.

• The notions of 'cows' and 'Caitra' have as object, externally too, both the unity of manifestation represented respectively by the general and the differentiation of the individual realities and parts.

8 I.e. 'on the external and internal plane' (according to the vṛtti), 'on unity and multiplicity' (according to Abh.). The two interpretations are equivalent in the light of the preceding kārikā.

9 As Abh. clarifies very well in the introduction to this stanza in his commentaries, Utp.'s intention here is to reply to the objection that might be raised by a Buddhist: these conceptual elaborations - which is what they are - do not take place at the time of the direct perception but later, when any contact with reality is precluded; therefore they are, in the absolute sense, unreal. In Utp.'s reply it is the continuity and dynamism of the I that guarantees the contact between these two moments of time, thus legitimating the working of the mind.

10 The absolute difference and self-containment of things, which is one of the pivots of Buddhist doctrine and the school of Dignāga in particular, finds its most well-known and rigorous expression in PV I 40ab (later taken up again in PVin; see Steinkellner 1973: I, 40-41) sarve bhāvāḥ svabhāvena svabhāvāvyavasthiteh ...

11 The two terms of the servant-king relation have an autonomous reality on the external level, but they cease to be separate once they are linked in the consciousness of the knowing subject. This does not mean that they become one, because in that case it would no longer be a relation: there is unification in separation (bhedābheda).

159

II 2.5 configuration of cow and by the configuration of a particular man seen as a whole, and the multiplicity of manifestations typical of the individual realities (svalakṣaṇa°) and the various parts12. - 5 -

6. The connection existing between the factors of the action (kārakānām) is based on the awareness of the action (kriyā°); the notions of space etc. rest on the connection between limit and limited.

• The notion of action is based on unity-multiplicity, insofar as the various factors that contribute to carrying out the action - wood, saucepan, Devadatta, rice - are internally linked to each other by the verb 'he cooks' and externally differentiated. Also spatial and temporal succession - given the mutual involvement of the objects which act as the limiting and limited element - is merely a particular case of relation characterized in the same way by unity and multiplicity. The notions of universal, individual substance, action, number etc. all rest on inherence (°samavāya°), which is a particular type of relation. - 6 -

7. Thus the knowing subject who aspires to the production of determinate results may obtain them with an entity that is at the same time both differentiated and undifferentiated. Thus one may not speak of error (bhrāntiḥ) as regards them.

• With the notions that have been discussed so far, that is, action and so on, which are both one and many, the knowing subject has the capacity to obtain the determinate effects he desired (which demonstrates their congruence with reality), accompanied by the continuity and homogeneity (°ānugunya°) of their manifestations. Therefore such notions are not erroneous13. - 7 -

12 The unity-multiplicity tension, which characterizes the series of concepts examined in this āhnika, is not reduced, as it might have seemed so far, to that of internal-external. In actual fact, for the concepts of jāti (universal) and of dravya (understood as individual entity constituted by a complex of parts) the double pole of unity-multiplicity is also to be found at the level of external reality, as the kārikā and vṛtti indicate.

13 The notions of action etc., do not, that is, become subsequently contradicted by other cognitions: thus one has the requisite of sthairya which, together with arthakriyāsamvāda (already mentioned), determines the empirical validity of a cognition.

II 3.1-2

Chapter III

1-2. The means of knowledge (pramāṇam) is that thanks to which the object is situated within its own confines (vyavatiṣṭhate)1 ‘this thing, with these characteristics’. This means of knowledge is an ever freshly arising light related to a subject (svābhāso ’bhinavodayah)2. This light, whose essence is the inner reflective awareness of that which is thus manifested, becomes - as regards the object without spatio-temporal differentiations etc. and expressed by a single name - knowledge (matiḥ), [provided it is] not invalidated.

• The means of knowledge is that on which the setting of the object within its confines depends (vyavasthāpyamānatā), both as regards its mere being and the various possible qualifications such as permanence and so on. It consists precisely in the manifestation to a knowing subject of a new object (apūrvavastu°) as related to him (svatvena). And this very manifestation ‘this’ or ‘permanent’3, which thus appears to the subject as separate and new, becomes the object of reflective awareness and finally valid knowledge (pramitiḥ), whose stability is not undermined by

1 In the interpretation of the notion of vyavasthā- vyavasthāp- (which refers to the definitive establishment of something, both as regards its intrinsic nature and the other realities in whose context it is situated; ‘establish separately’) - so pregnant with meaning and untranslatable in a single word - I have kept in mind Abh.’s gloss (ĪPV II p. 68 niyatām prakāśamaryādāṃ na ativartate).

2 Svābhāsa (together with viṣayābhāsa) characterizes the pramāṇa also in the opinion of Dignāga and his school (cf. PSV 1.9a, PS 1.11ab and vrtti, PS 1.10 in particular). However in Dignāga, and in the Yogācāra view in general, sva° refers to knowledge itself seen as the subject, whereas in the ĪPK it refers to the knowing subject and its centrality (cf. ĪPV II p. 73 svatvena ābhāsamāno ya ābhāso ...; Bh ibid. svatvena - grāhakāśrayamatāvisayatvena; Utp.’s vrtti pramātuḥ svatvenāpūrva-vastvavabhāsah). On pramāṇa as revealing a previously unknown object, cf. the second (and probably complementary) definition of pramāṇa in PV II.5c ajñātārthaprakāśo vā (Katsura 1984: 220, Dreyfus 1991: 25 ff., Franco 1991, van Bijlert 1989: 150-53).

3 According to Abh. (see text note 55), Utp. chooses to make these two terms masculine (in the kārikā they were neuter), thus making them agree with ābhāsa, to state implicitly the single nature of the means of knowledge and of its objects (both being ābhāsas).

161

II 3.1-2 another means of knowledge4. The object of such knowledge - which is an activity of the subject (pramātrvyāpārah) - is that entity which is

4 According to the Pratyabhijñā, therefore, there is no differentiation between the means of knowledge and its result (pramāṇa-pramā), as also the Buddhists - but not only them (cf. Bandyopadhyay 1979) - maintain (see for example PS 1.8b-11 and vrtti, PV III. 301-319, NBT pp. 79-86, 90-1, TS 1343-60). The Buddhist position on this point is substantially unitary, with a distinction between Sautrāntika and Yogācāra resulting from the admission or non-admission of the reality of the external object. Thus, according to the Sautrāntikas, prameya is the external object, pramāṇa the assumption by knowledge of the form of the object (visayākāratā, sāriipya), pramā the obtained knowledge of the object (visayadhigati). According to the Yogācāra, prameya is the appearance of the object to consciousness understood as the appearance of consciousness itself in the form of object (visayābhāsa, visayākāratā), pramāṇa knowledge in the form of subject (svābhāsa) or, in other words, the capacity (yogyatā) of knowledge to know itself; pramā, finally, the self-awareness of knowledge (svasamvitti; on svasamvitti as the final outcome the two doctrines would agree); see e.g. PV III. 366-7, TP 1346 with the relative pañcikā and especially PS 1.8cd-10 and vrtti, with the notes of Hattori; see also Iwata 1991: 1-20. The distinction between pramāṇa and pramā - the Buddhists say - is only the outcome of the analytic consideration of a reality, cognition, which is in itself one. The two terms which are thus foregrounded cannot in any case represent a relationship of cause and effect - because this would require the actual otherness of the two terms - but at most a relation of vyavasthāpya-vyavasthāpaka, with a division of roles within the same reality (NBT p. 82 na cātra janyajanakabhāvanibandhanah, yenaikasmin vastuni virodhah syāt api tu vyavasthāpyavyavasthāpanabhāvena tata ekasya vastunah kimcid rūpam pramāṇam kimcit pramānaphalam na virudhyate). And up to this point Śaivas and Buddhists are mainly in agreement (except that the Śaivas admit a relation of the type hetu-phala). A proof of this is the fact that Abh. in concluding his argumentation (ĪPV II pp. 73-5, ĪPVV III pp. 71-2) - which develops Utp.'s above-mentioned position - quotes part of a verse from Dharmakīrti (PV III.308c tadvaśāt tadvvavasthānāt). On the contrary, the two positions differ on the concept of 'function, activity' (vyāpāra), performed by the elements occurring in cognition. Vyāpāra is completely denied by the Buddhists, who consider every distinction on this basis completely imaginary (utprekṣitu), so much so that, for example, an act such as piercing with an arrow may be analyzed in various ways, all equally legitimate, attributing to the bow the function (vyāpāra) either of kartr, or karana, or apadāna (TS 1346 ata utprekṣito bhed o vidyate dhanurādivat / utpādyotpā-dakatvena vyavasthcyam tu nesyate // and relative Pañjikā). But, above all, the impossibility of vyāpāra is the direct consequence of the acceptance of the doctrine of momentariness (TSP I p. 488 kṣanikatvena nirvyāpāratvāt sarvadharmāṇām). Knowledge, therefore, only 'appears' to be endowed with a function (PV III.308a savyāpāram ivābhāti). The reply of Abh. - who, as we have seen, had just quoted with approval the third pāda of the same verse - implicitly refers to this statement by Dharmakīrti (who in turn depends on PS I.8cd savyāpāra-prapatitatvāt pramāṇam savyāpāram ivābhāti): vyāpārā-tmakam eva hi karaṇasya prathamam ādhyatvāt phalam /

phalam eva sat): vyāpārā-tmakam eva hi karaṇasya prathamasādhyatvāt phalam /

manifested to the consciousness, taken separately5, as a universal6, II 3.1-2 devoid of spatial differentiations etc., denoted by a single word in conformity with the specific reflective awareness (vimarśabhedānusāri°) and endowed with its own causal efficiency (svakārthakriyāprāpte). - 1-2 -

3. In an object, though its unity is established by the unifying power of the mind (anusamdhāna°) , various manifestations may be distinguished according to the inclinations, practical requirements and specific experience [of the subject].

• For the object, though it is unitary, having been assumed as such thanks to a single [unifying] reflective awareness (ekapratyavamarśa°)7,

vyāpāraś ca vyāpriyamānāt kartur vyāpāryamānāt vā karanād ananyātmakā eva yukta iti mukhyatvena pramānaphalatayor abheda eva nyāyyah ĪPVV III p. 72; see also ĪPV II p. 75). Vyāpāra not only exists but constitutes the very essence of pramā, and according to the Śaivas the non-differentiation of pramāṇa and pramā hinges on this: vyāpāra is not a different reality from the subject that acts and from the instrument that is set in action. All this had already been essentially contained in the laconic pramātṛvyāpārah that follows pramitiḥ in the vṛtti. But Abh. (ĪPV II p. 75) goes on to say that this does not mean that pramāṇa and pramā are simply two ways of saying the same thing (paryāyatvam); the cognitive light (bodhātmā ābhāsah) which is the essential nature of both is turned towards the external object in the pramāṇa, whereas in the pramā it is turned inwards as pure determinate awareness, contracted due to the influence of the object assumed in it, having as its essence the word.

5 Prthak pṛthak, in the sense that the pramāṇa works only on the single ābhāsa (in the form of sāmānya). The knowledge of the particular is the result of the subsequent unification of the group of single ābhāsas, among which those of time and space have a particular individualizing power. As Abh. notes, the same vimarśa, which is prakāśapramāṇa, sometimes rests on the single ābhāsa - and we have a sāmānya -, sometimes combines several ābhāsas - and we have a svalakṣaṇa; in the latter case it takes the name of anusamdhāna.

6 On the basis of the interconnection - and in the final analysis of the identity - between vimarśa and śabdana, Utp. is led to state the sāmānya character of the object of the pramāṇa. in that it is precisely the sāmānya that is expressed by the word (obviously in the Śaiva view).

7 The particular realities, in themselves completely different from one another, form an apparent unity on the strength of the same judgment that they are able to determine because of their very nature. This concept, even in its formulation, refers back to Dharmakīrti (see e.g. PV I. 109, already quoted, and the vṛtti tad api pratidṛśyam bhidyamānam api prakṛtyaikapratyavamarśasyābhedāvaskandino hetur bhavad abhinnam khyāti). This may be applied, mutatis mutandis, both to account for the formation of the idea of class or universal which embraces a plurality of dis-

163

II 3.3 there are different manifestations depending on the will, practical requirement, and experience of the subject8. - 3 -

To elaborate:

4-5. Just as the various manifestations are differentiated as 'long', 'round', 'tall', 'man', 'smoke', 'made of sandalwood' and so on, without this entailing a spatial-temporal differentiation, so one also has various distinct manifestations such as 'being', 'jar', 'individual substance', 'made of gold', 'shining' and so on; each has its own separate efficiency. They are the object of the word.

• Given an object unitarity in itself9, for example a jar, depending on how a subject regards it (avadhi tsāvaśāt) what appears to him may be the length, or the triangular, circular etc. shape. What appears of a man may only be his 'being erected' if the subject is only looking for a reference point, or a shelter, or a shade. 'Man' as such, however, appears to those who regard him entirely (nyakṣena)9a, seeking the services that

crete realities that may not be strictly reduced to anything other than temselves (as in the context of Dharmakīrti's argumentation), and to account for the synthesization of a group of distinct ābhāsas in the apparent unity of the object which confronts us in practical reality (as in the context of Utp.'s argumentation). The expression ekapratyavamarśa may be understood either as 'single, same reflective awareness' concerning different things, which, owing to this, are assumed as being one (cf. PVSV p. 41 ekam ekākāram pratyabhijñānam commenting on ekapratyavamarśa° v. 73a); and as 'reflective awareness of things as being one' (ekatvena pratyavamarśah). The meaning is basically the same. Utp. and Abh. use this expression in both senses.

8 Here, too, there is an evident reference to a remark made by Dharmakīrti (PV 1.58 and vrtti), according to whom a thing that appears to perception as an undifferentiated entity is in fact grasped at the moment of determinate knowledge, depending on which of its components the subject is inclined to foreground (a female body may be seen as an object of desire, a corpse, something to eat; a person may at times be seen as a father and at times as a teacher). A similar remark had already been made by Bharthari (VP III.8.64 ācāryo mātulaś reti yathaiko yapadiśyate / sambandhibhedād arthātmā sa vidhih paktibhedāyoh //). See also PVSV p. 41 evam śimśapādayo'pi bhedāh [...] janayanti anyām vā yathāpratyayam dahanagrhādikām kāṣṭhasādhyām arthakriyām.

9 I.e. a svalakṣaṇa (see the previous kārikā). The multiplicity of manifestations does not entail the loss of the object's individuality, the object being a collection of ābhāsas, distinguished by a particular collocation in space and time and by a specific form (ākāra).

9a On the meaning of this uncommon word cf. Vṛṣabhadēva's Paddhati on VP I.3 (p. 20) nyakṣeṇa iti ābhimukhyena kārtasnyena vā (Abhyankar-Limaye 1965: VII n. 4).

164

are peculiar to him. To some [smoke] appears as merely smoke, in its II 3.4-5

general form (dhūmamātram eva); on the contrary, those who are familiar

with it (tadvidās tu) grasp its specific aspect, for example, that it is the

smoke from burning leaves, just as an expert is able to grasp the par-

ticular features of stones, silver etc. (manirūpyādiviśeṣa) and so on10.

However, the object that is manifested with these distinctions is not dif-

ferentiated as regards space and time. Thus for the object 'jar' there is

the manifestation 'being' - which is in itself different [from the specific

manifestation 'jar' etc.] - common to all the countless other objects such

as cloth etc. and the cause of specific effects such as the determinate

awareness 'is, exists' which is produced by the mere fact of existing; still,

as regards the jar, there may be the manifestation 'jar' known by the inner

and outer senses and common to every object with a broad base and bul-

bous shape, and different from the other manifestations, i.e. absent in

cloth etc.; or yet another, 'made of gold', absent in clay jars etc. This

manifestation becomes the object of sensory experience depending on the

subject's practical requirement and so on11. Each of these manifestations

is referred to for the practical purposes of communication, by those who

aim at obtaining this or that determinate result, with a single word: the

word 'jar' refers to the manifestation 'jar' and not to 'being' or 'made of

gold'12. The manifestation 'Caitra' common to the different stages of

10 This is an explicit reference to the ābhyāsikapratyakṣa in Bhartrhari's for-

mulation (VP I. 35 pareṣām asamākhyeyam abhyāsād eva jāyate / manirūpyādi-

vijñānam tadvidām nānumānikam //).

11 For instance, Abh. explains (ĪPV II p. 98), he who on seeing a jar thinks of

its price will be aware of the manifestation 'made of gold'. This does not occur at

a later stage, as a result of mental elaboration, but at the moment of the direct

perception itself (aindriyaka eva); see also below vrtti on II.3.7.

12 Cf. the position of Bhartrhari, according to whom a thing, in its intrinsic unity,

cannot be expressed by a word (except the pronoun); it may be the object of many

words, each referring to one aspect or power (śakti) among the many it is composed

of (Vṛtti p. 131 apara āha - yady api vastu na bhinnam śabdārthas tu bhinnah /

śabdo hy ekavastuviṣayānām api śaktīnām avacchedonapagrahe vartate; on the

meaning of apara in Bhartrhari's work see Prakīrṇakaprakāśa II p. 267, quoted in

Aklujkar 1970: 116 n. 5.). This theme is taken up again and developed by Dignāga

and Dharmakīrti. The thing, because of its many aspects, cannot be fully expressed

by one word (PS V.12 bahudhāpy abhidheyasya na śabdāt sarvathā gatiḥ /

svasambandhānurūpyena vyavacchedārthakāry asau //; cf. Hattori 1980: 70 n. 3,

where a relevant passage from Jinendrabuddhi's PSṬīkā is also referred to; see also

the parallel passage, not extant in Sanskrit, concerning cognition through logical mark,

in Frauwallner 1959: 102). The function of the word is only to denote a certain part

of the thing through the exclusion of what is other than that (arthāntaravyāvṛttya

165

II 3.4-5 childhood etc. is devoid of place and so on; and what has been said above applies to this, too. Thus causal efficiency is differentiated [for every ābhāsa]. – 4–5 –

6. Things possess a determinate causal efficiency (niyatārthakriyā) depending on the variety of the manifestations they are composed of; and, on the contrary (punah), [a different] one based on their appearing as unitary realities owing to a common substratum (sāmānādhikaraṇyena)13.

– In the unitary particular entity there is a specific effect for every single manifestation determined by the power of Necessity (niyatiśaktyā) and [equally determined by Necessity] a single substratum of the different manifestations, which produce – as we have said – manifold effects. The concrete entity (vastu) is such because of the manifestation of this sameness of substratum (sāmānādhikaraṇyābhāsavaśāt). In fact, this sameness of substratum constitutes the unity of what is manifold (anekasyaikatā)14. – 6 –

tasya vastunah kaścid bhāgo gamyate, quoted in PVSV p. 62). As Dharmakīrti points out in a subsequent passage, this does not mean that a thing has 'parts' (PV I.135ab ekatvād vasturūpasya bhinnarūpā matih kutah), but only that it appears as associated with various causes of error concerning its nature; a word serves to remove one of these causes (PVSV p. 64 tatraiva ca te śabās tais tair bhrāntikāraṇair ih samsrṣṭarūpa ivābhāti yathāsamketam vicchedāya vyāpriyante; cf. PV I.50-51). Dignāga's statement quoted above is closely related, also in its formulation, to the one found at the beginning of the PS (I.5ab): dharmino 'nekarūpasya nendriyāt sarvathā gatiḥ. They are like the two faces of a coin. And similarly Dharmakīrti explains the right way in which this assumed multiplicity of aspects and properties is to be understood (PV III.231; cf. Hattori 1977: 52). See also Torella 1992: 333-35.

13 The kārikā hinges on the ambiguity of the expression niyatārthakriyā, which belongs simultaneously to both parts of the sentence, with a different orientation of meaning. At least this is the interpretation that Abh. gives in ĪPV and ĪPVV: things have a multiplicity of arthakriyās, each one of which is inherent in every ābhāsa arthakriyā), due to the predominance of one ābhāsa over the others and therefore to their appearing according to a sameness of substratum.

14 The ultimate intention of this stanza is therefore, as I understand it, to restate the reality of the unity of the thing which the analysis carried out in the previous stanzas had, in a sense, denied. The meaning of the sentence seems to coincide with that of Abh.'s avatāranikā to the stanza (ĪPV II p. 101 nanu evam pratyābhāsaṃ eva vastutve eko ghaṭātmā na vastu syāt; ĪPVV III p. 113 nanu evam eko 'riho gha- tatvohitatvādyātmā na vastu bhavet, bhavati).

7. As in the case of the distinct rays of light in a torch and of the II 3.7 currents in the sea, so in manifestations that are not in contrast with each other the notion of unity (aikyadhīḥ) is produced by their appearing as unitary entities (ekakāryā).

– The individual rays of light in a torch appear as an undifferentiated whole and the same holds for the currents of rivers in the notion of the sea and for the various flavours in the nānaka. Similarly, manifestations such as 'white', 'big', 'cloth', capable as they are of interpenetrating, lend themselves to the manifestation, in the direct experience itself, of the various unitary things having a single, unitary, effect (tattadekakāryaikadravyā)15.

15 The very elliptical expression used in the kārikā (ekakāryā, literally 'produced by one') is analyzed by Abh. (ĪPV II pp. 106-7, ĪPVV III pp. 116-7) in its various possible implications, which are substantially analogous to those succinctly indicated in the vrtti. When one sees a unitary effect that cannot be traced back to the individual components of a given aggregate, this proves that they form a new complex unity (ekam vastu, svalakṣaṇa). The vrtti says that certain manifestations, characterized by mutual compatibility, may give rise to ekakārya (= ekakriyā) – namely to a unitary effect - and to ekadravya (= ekavastu, svalakṣaṇa) - namely to their appearing as a unitary and individual reality; the notion of their unity is produced by this. One must, however, remember that the individual ābhāsas continue to keep their own specific identity and causal efficiency intact, ready to come to the fore again depending on the attitude of the knowing subject. Ekakārya° may also be understood as internal bahuvrīhi referring to °ekadravya°: 'of a unitary substance with a unitary effect'. In this way the expression would be used in the same sense in which it is understood, in an analogous context, by the Buddhist logicians (to attribute unity to different entities on the grounds that they produce the same single effect). Dharmakīrti, too, - as we have seen - makes the (illusory) non-difference of determinate entities derive from their producing a single effect - hence the concept of sāmānya (PVSV p. 57 ekakāryataiva bhāvānām abhedaḥ). (A somewhat parallel question is to be found in the capacity of a causal complex (hetusāmagrī) to produce a single effect though being composed of many different causes to which Dharmakīrti does not acknowledge a common svabhāva; see Steinkellner 1971: 184-88). Cf. also TS 201ab ekakāryopayogitvād ekaśabdasya gocarāḥ. The reference to TSP p. 59 seems even more pertinent. Here Kamalaśīla (probably bearing in mind PVSV p. 68) states that ordinary experience groups together data that are completely heterogeneous and separate, such as a certain shape, colour etc., on the basis of their concurring to carry out a single function (e.g. that of containing water), constructing in this way the unity of a 'thing' (e.g. the jar): ta eva rūpādayaḥ salilādidhāraṇārthakriyākārinaḥ samudāyo ghata iti vyapadiśyante. Returning to our text, it may be added that ekakārya° in the vrtti definitely has a different meaning from ekakāryā in the kārikā: the expression is to be understood as a partical comment on eka° alone.

II 3.7 manifestations such as 'blue', 'yellow' etc. This is what we call 'having the same substratum'. - 7 -

8. Faced with a non-specific manifestation of 'fire' etc., a single means of knowledge knows what the outcome or cause of it is, its being hot, its being able to be denoted by this or that word and so on16.

• On a particular occasion (kadācit)17, when faced with the bare manifestation 'fire' in which only the universal 'fire' is present, without its particular features consisting in the single individuality (°svalakṣaṇa°) delimited by the association with a particular space etc.18, there is through the means of knowledge operating at that moment - and through that means only - and in a way that embraces the three worlds and the three times, the establishment of the various specific features of its own nature, such as its effect, its cause, its heat, the fact that it goes upwards and is denoted by the word 'fire' etc.19. - 8 -

9. On the contrary20, it happens differently in the case of the activity

16 Cf. Dharmakīrti's analogous statement, in a different context: (PV I. 43, 45ab) ekasyārthasvabhāvasya pratyakṣāsya satah svayam / ko 'nyo na drṣṭo bhāgah syād yaḥ pramānaiḥ parīkṣyate // ... // tasmād drṣṭasya bhāvasya drṣṭa evākhilo guṇaḥ ...

17 I.e., following ĪPVV III p. 121 (kadācit iti jijñāsāvasare), 'at the moment in which the desire to know appears'.

18 The pramāṇa, in fact, works on the individual ābhāsa, which, as we have seen, is a sāmānya.

19 The statement that a means of knowledge (pramāṇa, which may be generically understood here as a 'cognitive act') acts on the individual manifestation may give rise to some difficulties: e.g. (cf. ĪPV II p. 107) it becomes on these grounds impossible to determine the relation of cause and effect, whose elements being the object of distinct pramāṇas would remain isolated from one another in knowledge. This stanza and the vrtti reply to this objection as follows: this same pramāṇa (in this case a visual perception) that knows the individual ābhāsa - for example the bright form of fire - at the same time implicitly knows all those other ābhāsas which are set by the power of Necessity (niyatiśakti) in invariable concomitance with it: there is therefore no need for another pramāṇa to ascertain the concomitance, valid for every space and time, of that particular bright form with heat etc. This implicit knowledge regards not only intrinsic properties, such as heat, the fact of having as its cause wood and as its effect smoke and so on but also those derived from convention, such as the connection with a certain word and so on.

20 The difference from what was said earlier lies in the fact that the activity set in motion in the subject by the perception of a determinate, particular object requires the combined operation of several means of knowledge (pramāṇasamūha), whereas

that starts at that moment21 in the subject who aims at producing certain effects, as regards a particular, individual object, differentiated by various, specific sensations of place etc. (deśādikādhyakṣāntarabhinne)22, and also [in the case of that activity promoted] by inference.

• On the contrary, the bodily activity of a person (kāyapravṛttih)23, that may concern only the particular object in association with a whole variety of perceptions (of place, time etc.), in that he aims at producing certain effects, necessarily depends on the concourse of several cognitive acts (pramāṇasamūhād eva). This activity may also be promoted by inference, qualified, of course, by the direct perception of the property-possessor (dharmin). -9-

10-11. Though the particular forms of the manifestation of the objects may be differentiated (bhinna¯vabhāsacchāyānām) - in that they are far or near, directly or indirectly perceived by the senses, external or internal, faulty in what is needed to make them known or for other reasons -, their unity is not affected by this thanks to the main manifestation

for the knowledge of the same svalaksana a single means (with svasamvedana) is sufficient. This distinction corresponds to that between mere knowing and vyavahāra. (Pramāṇa, as I have already pointed out, may be translated here as 'cognitive act'; it would be misleading to say 'various means of knowing', as they may also be various operations of the same means - even if every pramāṇa is essentially new -). The apparent contradiction is pointed out in ĪPV II p. 114, where three different ways of explaining it are proposed. As regards inference in particular, Abh. (ĪPVV III p. 148) notes that the difference of the object, introducing a temporal differentiation, necessarily also brings about a difference in the nature of the pramāṇas; indeed, it is not possible to say that the direct perception through which the dharmin is known is one with the inference itself, which knows the fire, in the classical example, only indirectly through the linga.

21 At the moment when the cognition of the svalaksana occurs.

22 Abh. puts forward three distinct interpretations of the compound. In the first (following the order in ĪPVV) taking bhinne as a neuter substantivized adjective (= bhede), one gets 'there being a variety of perceptions', where the locative is taken to mean cause (nimittasaptamī); the need for a plurality of pramāṇas stated in the commentaries depends on this. As Abh. notes, the second 'the svalaksana being differentiated by perceptions of time etc.' coincides basically with the first; this is also the meaning attributed to it by the vṛtti. As regards the third interpretation 'the svalaksana constituting - though combined with perceptions of space etc. - an undifferentiated reality within the knowing subject' (adhyaksa-antar-antar-abhinne) - this seems traceable only to Abh..

23 And, one understands, also the other two types of activity, that of the mind and of the word. The translation of arthitayā had been omitted in the first edition (I thank J. Bronkhorst for pointing it out).

169

II 3.10-11 (mukhyāvabhāsatah) consisting in a single reflective awareness (eka-pratyavamarśākhyāt)24.

– Even if there may be a differentiation in the particular forms of the manifestation (i.e. as regards whether the objects are near or far etc., evident or not evident etc., external or internal), the unity of the objects is not contradicted, because there is a single reflective awareness forming the primary nature of the various particular manifestations25. – 10-11 –

12. Causal efficiency itself is not intrinsic to things, since it is determined by the will of the Lord. Thus a thing may not be said to be different because it does not possess a certain efficiency.

– Objects, such as the jar etc., when imagined do not lose their nature as a jar etc., even if they lack the ability to produce external effects, because this capacity is not intrinsic to their being (asvābhāvikatvāt), since it is determined by the Lord for every single manifestation26. – 12 –

24 Cf. again PV I. 109 ekapratyavamarśasya hetutvād dhīr abhedinī / eka-dhīhetubhāvena yakinām apy abhinnatā // (on this peculiar statement by Dharmakīrti see the critical remarks in Prabhācandra's Prameyakamalamārtānda pp. 469-70; Shah 1967: 143). Abh. interprets °ākhya° in the compound ekapratyavamarśākhyād as prathana, but the general meaning does not change much.

25 Thus the avabhāsacchāyās, or the various direct perceptions, the pratibhāsas of the Buddhists - which cannot but differ from one another even if they concern the same object, depending on whether it is near or far, frontal or lateral etc. are contrasted with the mukhyāvabhāsa - or the manifestation not divided up and discontinuous, seen, as it were, from a more elevated vantage point, and of a more general character (sāmānyarūpa) having as its essence the ekapratyavamarśa, that is a single reflective awareness, which recognizes and sanctions the basic unity of the thing through its manifestations; ekapratyavamarśa, for its part, is contrasted with the individual vimarśas corresponding to the individual pratibhāsas (or avabhāsacchāyās). This point is dealt with by Abh. in his two commentaries.

26 Utp. is aware that the previous kārikā is bound to arouse the Buddhists' opposition; what is particularly unacceptable is that the nature of an external and internal manifestation of an object is basically identical. The Buddhists would say that the reality of an object consists in its efficiency and only the external manifestation is efficient; therefore the two cannot be put on the same plane. In his reply Utp. denies, as he has done earlier, that efficiency constitutes the essential nature of a reality; it pertains to it only under certain conditions - primarily externality (see above I. 8.5ff.) - depending solely on the Lord's power of Necessity. In an earlier passage of the ĪPVV (III pp. 30-31) Abh., following in the steps of the tīkā, had denied that arthakriyā might be either svarūpa, lakṣana, or upalaksana of the real thing.

170

13. Even if the reflective awareness of silver is one, this awareness 'silver' referring to mother-of-pearl is not valid, since there is incongruence27 as regards the place, which has a qualifying function [in the cognition] (upādhideśāsaṃvādāt); also in the case of the double moon the ether appears differently.

• Even if the cognitions as 'silver' of real silver and of mother-of-pearl are in themselves equally real, insofar as in them the reflective awareness 'silver' [on which its validity is based] is the same, however the cognition 'this is (here there is) silver' referring to mother-of-pearl is to be considered erroneous because of its impermanence (asthairyāt), since it is not congruent as regards the accessory quality - place -, as its connection with the place occupied by mother-of-pearl is later found to be contradicted. Also in the case of seeing a double moon there is no congruence as regards the place, occupied in this instance by the ether28.

• 13 -

and 'acceptīon' in Dharmakīrti's work (it is apparently absent in Dignāga) see Nāgatomi 1967-1968, Mikogami 1979, Kano 1991.

27 Abh., as he did for ākhya in the previous kārikā, makes samvāda derive from a root vad- 'to shine'. Thus samvāda is taken to mean 'to be correctly and completely- (samyak) manifest' or 'to be manifest combined with a lasting determinate awareness' (vimarśānuvṛtti) (cf. ĪPV II p. 126). In translating samvāda in the vṛtti I have, however, kept to the current meaning.

28 The extremely laconic style of the vṛtti, which I have tried as far as possible to preserve in translation, makes a few additional comments necessary. Firstly, the relationship this return to the subject of error, already dealt with above, has with the preceding stanzas. Abh. says it is a reply to those who contest the role of the vimarśa in the establishment of reality, objecting that thus the notion of erroneous knowledge ends up by being denied. If a thing is such in that it is recognized as such by an act of determinate reflective awareness (vimarśa) - remember the maxim so frequently cited by Abh. bhāvarāśir yathā vimṛśyate tathā asti - which follows its appearing to consciousnessness (avabhāsa, prakāśa), then the mother-of-pearl that is mistaken for silver is no less silver than real silver. The stanza in question, containing Utp.'s reply, is inserted at this point. It is true that a vimarśa is in itself valid, but it is meant to be something momentary, it tends to last, unless that tendency to last is blocked by a further vimarśa. And this is precisely what happens in the cognition examined here 'this is silver, there is silver here'. A later vimarśa ascertains that the place that qualified the object of the earlier vimarśa (the 'here') is different from the one that had been understood: the place that ought to contain the silver is occupied by the mother-of-pearl, the place - in the case of seeing two moons - of the second moon is occupied by the ether. This reasoning rests on the assumption that every concrete object of knowledge (svalakṣaṇa) is mixed with space and time and that therefore the vimarśa must embrace them, too.

II 3.14

14. Thus, both the differentiation of things - due to qualities such as sound and so on - and their non-differentiation - due to universals etc. (jātyādibhih)29 - are possible on the basis of the unity of the knowing subject.

• It is a fact of direct experience that qualities such as sound, form, configuration and so on, suggest30 the differentiation of things and universals etc. their unity: this is admissible only if the unity of the subject is assumed. The determination of mutual differentiation also depends on the unifying function [of the knowing subject]. - 14 -

15-16. What function can the means of knowledge perform - being a new light (navābhāsah) - as regards the knowing subject, he who exists in the absolute sense, the Lord - like a smooth wall that is the substratum for the painting of the multiform universe - , whose association with non-being would be a contradiction in terms, the Ancient One31, whose nature is perennially manifest (sarvadābhāta°), inherent in every cognition?

• The means of knowledge - since it is an ever-new light which establishes the still not established reality of the object - can do nothing as regards the knowing subject, who is uninterruptedly luminous. Since it is the various ever-new manifestations that constitute the various means of knowledge, this light (manifestation) will be, of course, the means of knowledge for what is other, but not the self, which is always established; the Lord is, on the contrary, the knowing subject because he acts freely as regards knowledge. All the various multiform manifestations are pos-

29 The doctrine of differentiation and non-differentiation is restated yet again. Objects share both conditions: they are differentiated if one looks at the distinctive qualities of each of them, non-differentiated if one looks at the universal, similarity etc. (°ādibhih refers to sādrśya and bhedāgrahana according to ĪPV II p. 132). These two aspects may co-exist without excluding one another only on the basis of the unity of consciousness.

30 ĪPVV III p. 160, following in the steps of the tika, considers at lenght why the word āksepa 'allusion, suggestion' is used; and concludes yatah sphutam avabhāsanam nāsty alakṣanam, bhavati ca bhāsanam, tata ity arthah.

31 Purāna (literally 'the Ancient One') to be understood here as meaning 'he who existed before (purā)', before every other thing, beyond time, the eternal etc. This is also how Abh. explains the term (ĪPVV III p. 162), relating it to the impossibility of attributing any form of non-existence to the subject, which (non-existence) is in effect the soul of time.

sible in their multiformity insofar as the subject is stable. His volition II 3.15-16

makes the various objects that are manifest in reality, without lapsing

from their own nature of which he is the ultimate essence (tatsāra-

svarūpābhramśe), appear and disappear by virtue of the power of māyā.

In reality he alone possesses perpetual existence (sattā) independently of

every other thing, since it would be contradictory to associate any form

of non-existence (previous non-existence, non-existence due to destruc-

tion etc.) with him. Even if he is taught in terms of being and non-be-

ing, nonetheless his nature as subject remains unaltered, since the teach-

ing would be impossible there not being he who wishes to teach32. - 15-

16 -

17. [What in reality happens] is only that the use in practical reality

of notions and expressions (°vyavahārah) like 'Lord' etc., which due to

the condition of obfuscation was not brought about earlier (apravartitapūrvah), is promoted through the bringing to light of the powers33.

32 I.e. there not being the subject.

33 Utp. says yet again here that what is the true reality, the sovereignty of con-

sciousness, is not something that has to be proved (sādhya), brought to light; it is

constantly luminous, it is, in fact, light itself, it is unfading. If anything, it is rather

a question of removing that state of obfuscation which prevents the limited subject

from recognizing that he and the Lord are basically one and the same. Or, in other

words, to call consciousness by its real name, thus making it enter the sphere of

everyday existence; discourse and everyday reality are both included in the seman-

tic field of the term vyavahāra. (It has even been supposed that this extension of

meaning derives from the coalescence of two different words vyāhāra 'noun, verbal

designation' and vyavahāra 'practical transactions', both being represented in Pali by

the same word vohāra; cf. Bapat 1975. It is also to be noted that vyāhāra and

vyavahāra are usually both translated into Tibetcan with the same word tha sñad). It

is therefore a question not of demonstrating something new, but of pointing out

(pradarśana; cf. ĪPVV I p. 87) a reality that has always existed, of calling man's

attention to this and thus inducing him to fill the void that followed awareness and

to use this latent notion in practical life (vyavahārah pravartyate). And this is in-

tended to exclude, in the case of such vyavahārasādhana referring to the sovereignty

of the I, any nuance of apūrva, which is, on the contrary, admitted as being still

subtly present in the vyavahārasādhana referring to an insentient object (cf. ibid).

Recognition, which is in fact vyavahārasādhanarūpam (ibid. p. 39), concerns an en-

tity which, invested from the outset with a valid means of knowledge, is already

manifest (maulikadrdhapramānaprabhāvaprakāśamānatā yogini vastuni); the process

that leads to the identification of the I with Śiva - which is the specific function of

the śāstra - is therefore essentially an inference based on the svabhāvahetu, which

(as Abh. says, ibid. p. 106) vyavahārasādhaka eva abhidhīyate na tv aprasiddhapra-

173

II 3.17 - Only, the use in practical reality of notions and expressions like 'Śiva', 'Lord' etc. referring to our self which is the knowing subject - which was not brought about due to the total obfuscation caused by māyā - is established (sādhyate) for the purpose of teaching, meditation etc. through bringing to light the reasons that justify it, i.e. the presence in the I of the pure power of freedom etc. - 17 -

sādhanarūpah, thus leaving the basic self-illuminating quality of his object intact. This is what the sacred texts on the knowledge of the Self - the Vedānta, the Siddhānta and the secret tantras - aim at (ĪPVV III p. 179 yāvanti hi ātmajñānaśāstrāṇi śrutyantasiddhāntarahasyatantrarūpāṇi, tāni ātmani naiva siddhim apūrvararūpām racayanti / tasmāt vyavahāramātrasādhanaphalāni eva tānīti bhāvah).

II 4.1

Chapter IV

1. And thus, his power being infinite, he makes those things1 manifest thanks to his volition; and this constitutes his activity, his being creator (nirmātṛtā).

– The knowing subject, that is essentially consciousness, having infinite power, the Lord, by means of his volition makes those entities manifest in this way. And it is precisely in this power of volition (icchāśaktiḥ) that his activity, that is, his being creator, consists. – 1 –

2. But2 an insentient reality does not have this power – namely, to confer existence on something that is not3. Therefore, the relation of cause and effect (kāryakaranatā) is essentially reduced to that of agent and object of the action (kartrkarmatva°).

– But an insentient reality – whether it is primordial matter (pradhāna°) or atoms or seed – is incapable of producing something that does not exist. Being cause and effect, therefore, is merely being agent and object of the action and nothing more. – 2 –

3-4. What is non-existent is non-existent and that is that. One cannot predicate the existence of what is non-existent, nor, on the other hand,

1 By ‘those things’ Utp., according to Abh. (IPV II p. 152), means that the Lord makes manifest what he had already previously manifested. His work of manifestation is therefore continuous: it does not take place only once and it is uninterrupted (avicchinnena prabandhena).

2 The examination of the relation of cause and effect begins with the statement of the Śaiva position and the whole chapter is devoted to justifying this.

3 In the ĪPV (and similarly in the ĪPVV) Abh. proposes three distinct interpretations of asatah satah. According to the one at the beginning of the Vimarsini (ĪPV II p. 153) we have ankurāsya sato ‘sato vā, with reference therefore to the tārkika (Vaiśeṣika) and Sāṃkhya positions regarding the existence of the effect in the cause, which is denied by the former and accepted by the latter (in the ĪPVV III p. 186 bijāsyaiva seems to have to be corrected to ankurāsyaiva). The second is asadrūpasya satah (ankurāsya), which may therefore be traced to the interpretation in the vrtti. The third is asato ‘nkurāsya sato vā bijāsya.

175

II 4.3-4 would the existent gain any advantage from obtaining existence [which it already possesses]. If one then points out that in practical reality there is general consensus on the existence of the relation of cause and effect, [then one replies that] it consists in the fact that a thing, that is internally present4 (antarviparivartinah), becomes, by virtue of the power of that indefinable Being (tasya kasyāpi), the object of knowledge for both senses.

• Attributing the nature of existent to what does not exist is contradictory, and it is already established in what exists. [The relation of cause and effect consists in this:] a thing, already present within [the I], is 'created' by the Lord, or in other words, is caused by him to become the object of knowledge for the internal and external senses. - 3-4 -

5. Thus, action - which has so far been the subject of discussion (saiṣā) -, characterized by succession and manifesting itself both internally and externally (antarbahihsthitih), being related to an entity that is unitary (ckasya)5 and capable of showing both aspects, is proved to be one.

• This action, though taking place in succession, being both internal and external6, is connected with a unitary reality, the agent, established as such through inner awareness, and with an object7 that is also unitary,

4 To be more precise: 'that moves about internally'. Viparivrt- often recurs in Bhartrhari's work.

5 In saying this, Utp. is referring both to the subject and the object of the action (this is the interpretation in the vrtti which all the other commentaries also accept, cf. ĪPVV III p. 189 ekasya iti kartuh karmaś ca). Each is unitary in itself and unitary in that it is inserted in a superior unity, that of the cognizer that assumes these two forms (cf. ĪPVV III p. 258 ātmānam eva pramātrsamjñam karmabhāvena kartrbhāvena ca avalambamānā kriyā karmasthatvena kartrsthatvena vā vivicyamānā ekāśrayaiiva bhavantī vastutā ekaiiva).

6 I.e. the two internal and external levels of the action. Both the subject and the object, Abh. explains (ibid. p. 189), can be involved in these two dimensions without losing their unity (ekas ca asāv āśrayah samvidrūpo bhaṅavān svacchandatvavasvacak-tvābhyām sahate eva ubhayam antarbahirūpan: ... karmāṇo 'pi antarbahirubhayākārasahatvam parāmarśābhedāt).

7 All actions primarily rest upon the agent and are therefore called kartrstha: however, some actions may be particularly connected with the object and they are therefore called karmastha (cf. ibid. p. 190). This division goes back (but from a different angle) to the fourfold one - kartrsthakriyā, kartrsthabhāvaka, karmastha-kriyā and karmasthabhāvaka - mentioned in the MBh (on Pānini III. 1. 87, vārtt. 3).

since it is recognized as being one by reflective awareness (aikyena II 4.5 pratyavamṛśyasya), despite the manifold changeable forms of the manifestation. And it is precisely on the basis of this fact - i.e. that its substratum is unitary - that its unity is proved. - 5 -

6. The entity which is [first] internal with respect to another, once it has become external is to be understood as the effect of that other. The two planes of externality and internality are such in relation to the knowing subject.

• For an object the condition of effect is merely its having been made external (bāhyatāpādanam); thus both being external and being an effect are such only in relation to a single reality, and, indeed, it is depending on the knowing subject that one may speak in practical reality of internal and external (antarbahirvyavahārah). [Therefore effect is produced by the knowing subject and none other]7a. - 6 -

7. Cause, therefore, is only the knowing subject, and he in the two modes of manifestation of the effect remains unchanged in his unity8. It is therefore appropriate to say that action belongs to a unitary entity.

• Thus cause is solely the knowing subject, who remains one even if there is succession in the external and internal manifestation of the effect. On the basis of what has been said action is therefore proved to belong to a unitary (and single)9 subject. - 7 -

8. Precisely for this reason Parameśvara is taken as efficient cause as regards the shoot; [however] it is not possible to also consider entities different from him - such as the seed etc. - as causes.

• Considering the fact that producing something (utpādanam) means making it manifest externally (bāhyābhāsatāpādanam), the status of

The action of creating (nirmānakriyā) - Abh. explains (ĪPVV III p. 190) - concerning an object which has the cognizer as its ultimate essence, is at the same time kartrstha and karmastha.

7a See text note 170.

8 Abh. gives a different interpretation (ĪPV II p. 162): «… since the knowing subject is the cause of the twofold, external and internal, manifestation of the effect …».

9 Here and elsewhere eka means both 'unitary' and 'one, single'.

177

II 4.8 cause may be exclusively attribued to the conscious being. For this reason some consider the Lord efficient cause10, but in so doing they become inconsistent in considering inert and insentient realities - such as the seed etc. - which have no intentionality (niranusamdhānāsya)11 also causes. - 8 -

9. In fact, the potter produces the jar through a whole series of operations to which he subjects the clay etc., following the rule determined by the Lord (aiśvaryaiva vyavasthayā)12.

• Even when it has been ascertained that the cause is a knowing subject, i.e. in this case the potter, nevertheless, for the jar to be produced it is necessary to carry out determined operations on the materials, such as clay, following a criterion that does not automatically derive from the nature of the thing (na svabhāvena) but which is dictated by the restrictive order (maryādayā) created by the Lord, called 'necessity'13. - 9 -

10 The allusion is to the followers of the Nyāya-Vaiśeṣika, as Abh. and Bhāskara confirm (ĪPV II p. 163). They consider the Lord solely as efficient cause, which would be in itself admissible if it did not necessarily also entail the admission of a series of other causes from whose cooperation only effect finally derives. Material cause, auxiliary causes etc. - i.e. in the case of the shoot, seed, water, earth - are all insentient realities from which, in the Śaiva view, the causal function is precluded.

11 The term anusamdhāna is used in this school with various meanings. Sometimes it means 'unification, connection', often referring to the unifying function of thought which establishes relations between things (it is then glossed with ekīkarana), pervading them with its own dynamism. In other cases - as in the whole of this chapter - it has the meaning of 'intentionality, straining towards a particular objective'. These two levels of meaning may also coexist simultaneously.

12 In the ĪPV Abh. devotes quite a long commentary to this stanza, putting forward different possible interpretations; these reflect the doubts that the establishment of causality as outlined may give rise to. In the case of the potter, who is here used as an example: there is a knowing subject who as such may legitimately assume the role of cause, but how can one then say that cause is always and everywhere the supreme Lord, and, on the other hand, if one accepts this thesis at the outset, what room is there in it for the potter? The stanza may be interpreted in both of these ways, depending on the meaning one chooses to give to the expression aiśvaryaiva vyavasthayā; and also in a third (or even fourth) way, if one takes the optative janayet as being 'interrogative' (sampraśna) and not in the sense of 'obligation' (niyoga).

13 This means that the potter's kartrtā is not absolute, but made possible by the higher kartrtā of the Lord, on whom the order of things - the operations the potter

10. The various things (tattat), jar etc., - lasting and able to fulfil II 4.10 the functions that are natural to them (sthirasvārthakriyākaram)14 - may also be produced by virtue of the volition of the yogins alone15, without the need for clay or seed.

• By virtue of the volition of the yogins alone, without recourse to clay etc., things such as the jar may be produced, that endure and are capable of fulfilling the functions of the jar and so on. - 10 -

11. It is precisely because of this that an effect (kāryam) or an essential property (svabhāvah) - which comes into being from that root which is the arising [from a cause] (utpatimūlajah)16 - may legitimately constitute the logical reason of an inference only provided that another valid means of knowledge has established that it is not the creation of a yogin.

• Precisely because of this, an effect or an essential property - that implicitly contains a relationship of causality17 - are only illusory has to carry out etc. - depends within the creativity of the Lord, otherwise one would not understand how clay etc., insentient as they are, can obey the craftsman's will. See also ŚD I. 44-45ab and vrtti.

14 My translation conforms to the vrtti (ĪPVV III pp. 194-5 sthiram ca arthakriyākaram ceti yajanā vrttikrtā krtā) takes tattat sthirasvārthakriyākaram as a single compound (which thus comes to mean 'able to carry out lastingly the various forms of efficiency that are natural to them'). claiming that his interpretation must also be considered admissible (... iti samāse 'pi na kaścid doṣah); we do not know if the ṭikā authorized this type of construction. On the contrary, according to the vrtti, tattat is separate and refers to ghaṭādi and sthirārthakriyākaram is to be understood as a karmadhāraya expressing two qualifications of ghaṭādi.

15 The example of the yogin [which occurs frequently in Indian philosophical literature] was already to be found in the ŚD (see III. 34b-36a and Utp.'s vrtti, etc.). See also above I. 5. 7.

16 The translation goes back to Abh.'s gloss (ĪPVV III p. 197 svabhāvo 'pi sva-bhāvāntaraniyato yo jātas tatra utpattir eva kāryakāranabhāvātmikā mūlam. Cf. the well-known passage from Dharmakīrti (PVSV p. 17) siddhas tu kāryakāranabhāvah svabhāvam niyamayati.

17 Dharmakīrti, as we know, admits only two types of 'essential connection' (svabhāvapratibandha; see Steinkellner 1984) between things - or rather between concepts -: identity (tādātmya) and causality (tadutpatti). Therefore the only inference he considers valid is the one based on svabhāva (Steinkellner 1974, 1992) - anupalabdhi also enters into its ambit (cf. Iwata 1991: 86-88) -, e.g. 'this is a tree, because it is a śimśapā' where the concepts are co-extensive, in that the one im-

II 4.11 reasons, if the certainty that this is not the creation of a yogin is lacking. However, even when this certainty exists, the status of logical reason (hetutā) still depends on the necessity instituted by the Lord. - 11 -

plicitly embraces the other and refers to the same thing; or on kārya, 'there is fire there, because there is smoke'. Utp. points out that the svabhāvahetu, too, in the final analysis emerges as being based on causality (tadutpattigarbha), given that the nature of a thing is that which is due to the causes that produced it (cf. above note). Dharmakīrti had already expressed himself in similar terms (PVSV p. 99 śālibījadinām api sa svabhāvah svabetor iti yo na taddhetuh so 'tatsvabhāvah syāt; PV I. 38 anvayavyatirekād yo yasya drṣṭo 'nuvartakah // svabhāvas taysa taddhetur ato bhinnān na sambhavah //; cf. Steinkellner 1971: 188 notes 35 and 36). The causal implication, as Abh. observes following the tīkā (IPVV III pp. 197 ff.), may be evident to a greater or lesser degree. It is obvious when by virtue of the sva-bhāvahetu 'to be endowed with smoke' one goes back to 'to be endowed with fire'. It is less evident, but equally present, in the case of 'this is a tree, because it is a śimśapā': here it is, in fact, the cause of the śimśapā - i.e. its seed - which infuses it with the nature of tree (ibid. p. 202 śimśapābījasyā viśiṣṭasya yā śimśapām prati bījată śimśapākarantā, saiva vrkṣatvam tatra niveśayati; see also IPV II p. 175 svabetuta eva hi śimśapā vrkṣasvabhāvāvyabhicārinī jātā; and this cause that produces the śimśapā as always united with vrkṣatva depends in turn on Parameśvara's power of Necessity, IPVV III p. 200). It is therefore essential in each case to establish at the outset that the smoke or the śimśapā in question are not yogin creations. The inference based on svabhāvahetu as conceived by Buddhists and Śaivas differs in that for the former it does not give rise, strictly speaking, to a new and real knowledge. According to the Buddhist pramāṇa tradition, the very relationship of gamya-gamaka between two essential properties is ultimately unreal, just as is the distinction of a plurality of properties within one thing. 'All this question of inference and object to be inferred' - Dignāga says in a famous passage often quoted, in slightly different forms, by later authors (Randle 19262: 51-54; see also PVSV pp. 2-3, IPVV III p. 200) - 'depends on a differentiation between property and property-possessor imposed by the mind, and does not concern the actual external existence'. The difference between the two kinds of inference - besides the operating of two concepts belonging to the same thing in the one case, and to two distinct things in the other - would seem to lie in the fact that the svabhāvahetu can establish only vyavahāra and not vastu (NBT p. 106 tad ayam arthah - vrksyavahāro 'yam, śimśapāvyavahārayogyatvāt). Abh., commenting on the tīkā (IPVV III p. 213), distinguishes between vastu°, vastuvyavahāra° and śabdavyavahārāsādhana. In the śaiva view vrkṣatva and śimśapātva are two different and in themselves completely autonomous ābhāsas, set by the Lord in a relationship of sāmānādhikaraṇya but not necessarily present to consciousness at the same time. Thus the status of the svabhāvahetu, moving from a real entity towards an equally real entity, as it were, enhanced; as Abh. says (IPV II p. 180) referring to the Buddhists, tasmāt sarvesu svabhāvahetusv ābhāsabhedam vinā vyavahāramātrasādhanam (vyavahāramātra, tnat is, śabdavyavahāra; IPVV III p. 213).

180

12-13. 18 A new (nūtanam) manifestation of smoke etc. derives from a II 4.12-13 manifestation of fire etc. common to various other knowing subjects, which, though not directly perceived at that moment, is its determining condition (adhipateh) 19. The effect, which is invariably concomitant with

18 These two stanzas respond to a twofold need. The first is the need to succeed in explaining causality whilst remaining faithful to the equation ābhāsa=vastu (how can one through inference determine the reality of a fire that is anābhāta being the cause of smoke present here and now?). The second need is to obtain this result by distancing oneself from the Vijñānavādins, who, in maintaining that the ābhāsas of the cause and effect must belong to the same individual santāna, risk removing every general value and therefore all practical importance from causality. The solution proposed by Utp., according to how Abh. develops it, is the following. Some subjects present in a certain place (e.g. the kitchen) grasp the relation of invariable concomitance (vyāptigrahaṇa) between fire and smoke. in general, but through the observation of particular cases. Namely, they become one as regards these two ābhāsas, or, in other words, there is a single ābhāsa 'fire' and a single ābhāsa 'smoke' for all the subjects present. This identification is provided by the action of the Lord himself through the power of Necessity (niyatiśakti), and here lies the fundamental difference from the Vijñānavādins. Later, one or more of these subjects finds himself faced with a particular ābhāsa 'smoke' - thus present on his own mental continuum (svasantānavartin) - and, remembering the previously established vyāpti, goes back to a generic ābhāsa 'fire', which is immediately present to other subjects (parasantānavartin): therefore at the moment of the inference the subjects involved are found to be, by the very will of the Lord, unified as regards a particular smoke and a generic fire. In the kārikā bhūyah, understood in the vrtti as an adjective referring to °pramātr° (cf. vrtti °pramātrantarā°), may also be taken separately as an adverb. In this case it would mean 'also later on', i.e. also at a time subsequent to the grasping of the invariable concomitance between smoke and fire, that is, at the time when the inference is made (Bh II p. 185 bhūyo 'pi na kevalam vyāptigrahaṇakāla eva api tu anumānakāle 'pi). This double interpretation of bhūyah goes back to the ṭīkā itself (cf. ĪPVV III p. 216).

19 Of the four pratyayas 'causal conditions' in the Sautrāntika conception (as outlined in SDS p. 85 te catvārah pratyayāḥ prasiddhā ālambanasamanantarasaḥa-kāryadhipati rūpāḥ ...) the adhipatipratyaya is the one most directly responsible for the production and the specific nature of the effect (cf. Madhyamakavrtti p. 33 yas-min sati yad bhavati tat tasyādhipateyam iti adhipatipratyayalakṣaṇam); for in-stance, it is the organ of sight that is the adhipatipratyaya of the visual sensation, in the presence of a determinate object also able to arouse the sensations of taste, smell etc. (SDS p. 86 cakṣuṣo 'dhipatipratyayād viṣayagrahanapratiniyamah / uditaśya jñānásya rasādisādharānye prāpte niyamakasyādhipatir bhavitum arhasi loke niyamakasyādhipatitvopalambhāt). In the AK the adhipatipratyaya is included in a system (see II. 49-55a. 61b-73) formed by six hetus and four pratyayas - the other pratyayas are there hetu°, samanantara° and ālambana° - (cf. de La Vallée Poussin 1923-31: I pp. 299-308; for different classifications inside the

II 4.12-13 the cause, is indicative (lingam) of the latter being there. On the contrary, a manifestation of smoke different from the one we have spoken of derives solely from the manifestation 'smoke' perceived by other subjects, which takes on in this case the role of determining condition.

• The manifestation 'smoke' (in the distance etc.), which, though not preceded by the direct presence of the manifestation 'fire', arises from the manifestation 'fire' common to the various other knowing subjects in that place and endowed with such causal power by the power of Necessity; this manifestation 'smoke' is that which invariably makes known the existence of this manifestation 'fire'. On the contrary, that smoke which derives from other smoke that already exists20 - the manifestation 'fire' now being completely extinct -, arises precisely from the manifestation 'smoke' present to the consciousness of various other subjects, which, though not being directly perceived at that moment21 [by the subject who makes the inference], is invested [by the Lord] with the role of determining condition22, as we saw in the previous case. - 12-13 -

Buddhist tradition, id.: I p. 299 n. 1; see also id. 1928-49: 448 ff. for the hetupratyaya theory in the Vijñaptimātratā). The first hetu in the series (kārana°) is identified with the adhipatipratyaya. All dharmas can be kāranhetus (but not with regard to themselves) in the general sense, in that they do not make obstacle to the causal process; but there is a principal (pradhāna) kāranhetu which is that responsible for the effective production of the effect - for example, sight or form for the visual sensation or food for the body (AKBh p. 83: see also de La Vallée Poussin 1923-31: I pp. 247-248). The adhipatipratyaya in the kārikā therefore would correspond to the pradhānakāranhetu. Actually, neither Utp. nor Abh. explain what they exactly mean by adhipatipratyaya. In particular Abh. limits himself to pointing out the character of 'being remote, out of sight' (ĪPV II p. 185 tasmāt kāryāt so 'numīyate yatah parokṣo 'sāv adhipatitvād eva, ĪPVV III p. 215 tac ca vahnyābhāsādi yato 'numātuh parokṣam, tato 'dhipatipratyaya ity ucyate, etc.). See also n. 21.

20 The reference is to the smoke the shepherd puts into a leather bottle and lets out later. The example, known as the gopālaghatikā, already occurred in ŚD V.67: see also Karnakagomin's commentary on the PVSV p. 99.

21 The adhipatipratyaya must necessarily belong to a different (earlier) time from that of the effect (cf. AKBh p. 84, where there is an exposition of the Sautrāntika criticism of the Sarvāstivādins, supporters of the relation of causality between co-existent dharmas: see also PV III.246). And, on the other hand, if it were not parokṣa, it would be the object of direct perception and there would be no need to resort to inference.

22 Each of the elements that enter into the relation of causality plays a role, that does not directly derive from or coincide with its own nature (as in the Buddhist

14. Also the relation of cause and effect conceived as 'there being this, this other is produced (asmin satīdam asti)23 is not admissible for realities that are insentient and as such incapable of 'requiring' (apekṣā0)24.

– A fixed temporal succession of antecedent-consequent, which is expressed in the formulation 'there being this, this other is produced', also occurs between things that have no causal connections, such as for instance the rising in the firmament of the constellations of the Kṛttikās and Rohiṇī25. The relation of cause and effect should therefore be formulated thus: 'there being the causal power of the thing that precedes, there is the existence of the one that follows'. But this is not possible for insentient realities incapable of requiring26. And even if we understand the meaning of the formula as 'the causal power of the precedent constitutes the existence of the consequent', in this way, too, virtually nothing is said; neither can the own nature, characterized by the causal power, of the preceding element appear as the existence of the following one. – 14 –

conception), but is, so to speak, superimposed upon it by the power of Necessity of the Lord, who remains the only true causal agent, the agent par excellence. This is also generally valid for the whole sphere of practical reality.

23 The Buddhists in general are referred to here. The relation of causality is resolved, in the final analysis, in a fixed temporal concomitance as being the only ascertainable thing (cf. Torella 1979: 397-98. n. 32).

24 To the same objection put forward in TS 521 and relative Pañjikā – «If causality is explained in this way, namely without the intervention of an operation, how can one explain expressions such as 'the effect depends on (requires) a cause'?» – the Buddhist replies that it is only a question of turns of phrase: what is called 'dependence, need' (apekṣā) is merely an invariable immediate succession (ya ānantaryanīyamah saivāpekṣābhidhīyate).

25 The Pleiades and Taurus. This argument is commonly used by the various critics of the Buddhist conception of causality. The invariable concomitance between the rising of these two constellations, a Buddhist would reply (cf. TS 1423-4), rests anyway upon a causal connection, which links the two terms not directly but through their common dependence on a single complex of causes (cf. PV I. 8); in this case, in fact, it would be the same wind (prabhañjanaviśeṣa) that pushes both constellations consecutively (santatyā).

26 I.e. in the sense that the existence of one thing 'requires' (apekṣā) the existence of another. The concept of apekṣā – which is the meaning of the locative case here – contains on the one hand this intentional dynamism (anusamdhāna in the following kārikā: cf. SSVT p. 2 jadasya virabhisamdheh ...), on the other the resting of one thing upon another: IPVV III p. 228 paraniṣthārūpā anusamdhānarūpā vā saptamy utthāpitā (directly commenting upon abhiprāya and anyanusañgitā in the țīkā), and yet again ibid. p. 229 anyonyalagnatātmikā apekṣā and abhiprāyakṛtā.

183

II 4.15 15. In fact the meaning of the locative case may not be applied to self-contained entities, incapable of intentionality (anusamdhānā°), whether [cause and effect] are considered existent or non-existent.

• Effect and also cause may be understood as existent or as non-existent; but, whatever the case, they still remain insentient realities, self-contained and lacking intentionality, whose being is independent of other entities. And it is for this reason that the meaning expressed by the endings of the subordinate cases27, whose essence is precisely its dependence on a predominant element, is not applicable to them. - 15 -

16. On the basis of what has been said the only logically admissible relation between things is to be found in the meaning of the case endings, which consists in a relation between the action (kriyākāraka°) and has as its only foundation the knowing subject28.

27 All the cases are subordinate - excluding, of course, the nominative which is by definition svatantra - (ĪPVV III p. 231 dvitīyādayo); the kartṛ presides over and promotes the carrying out of the individual functions of the various kārakas and they all bring about the main action expressed by the verb. But subordination itself presupposes a conscious dynamism, an intentionality which everything that is conceived as jada, as confined in itself, cannot possess. as well as a capacity for unification, for establishing relations, which is all expressed in the single word, anusamdhāna, in all its concentration of meanings. However one wishes to conceive this dependence (we have seen that, following the tīkā, Abh. has distinguished two types, anyonyalagnatārūpā and abhiprayārūpā), it is precluded from inert and insentient realities. Utp. returns to this subject and develops it further in the SSVṛ (p. 2): na hi jadayoh sambandhinor apeksārtho ghatate pāratantryārtho vā / athāpy ankura eva hījam apeksate hīje vā paratantro bhavati, tadāvatryādes tadātmalābhassyety etad api na samyak; upacāro hy ayam apeksata iva paratantra īvati syāt, na tu jadasya mukhyaivākaṅkṣā prārthanā cecchaiva viśiṣṭā apeksā pāratantryam vā parāpravan- atā parakīyaviniyogākaṅkṣā svecchācarananirodhalakṣaṇah samkalpaviśeṣa eva saṃgacchate. alabdhātmaṇaś cāitmalābhārtham apeksādi kathyate. If the Buddhist premises are accepted - Abh. concludes (ĪPVV III p. 231) - the relation of cause and effect therefore remains inexplicable, seeing that however one puts it - hīje sati aṅkuro, hījād aṅkuro 'nikurasyā hījāṃ etc. - no formulation stands up to a logical examination, as there is no room for assuming the role itself of kāraka.

28 This concept is already formulated in substantially the same terms in ŚD IV. 32ab vinaikatvam ca na bhavet kārakaṅvam kadācana; Utp. gives a very succinct commentary on this, referring the reader to the ĪPK (and commentaries) for a more extensive treatment of the subject: ghatādeś ca yad etat svakāryakaranam bhavadbhir isvate tac cidekarūpatvam vinā na syāt / jadasya nirabhisamdheh karanāyogād ity etad api īśvarapratyabhijñāyām evoktam.

• The connection between earth, seed, water and so on is correctly understood as the meaning of the case endings consisting in a relation between verbal action and the factors of the action, depending on a single subject, and is not to be identified with some other type of 'dry' (śuṣka)29 connection such as that of cause and effect. - 16 -

17. Even if it is maintained that cause and effect have one another's very nature, merely the unity of the two is obtained, for if there were any distinction it could not be said that they have the same nature30.

• Even if it is maintained that cause and effect have one another's form, the only result would be that they are a single thing and not that they constitute a relation. - 17 -

18. The differentiation of a unitary entity is action, occurring in temporal succession. In this way we necessarily return to our thesis of an agent subject, as being that which becomes modified in the various forms31.

• The presentation of a single reality in differentiated ways, change, impelled (or 'measured') by time, is precisely action. Therefore for an entity free to modify itself, endowed with energy, being a cause means being the agent of the action. - 18 -

19. But this is not possible for an insentient reality, because differentiation would conflict with unity, given the manifold forms of manifestation

29 The meaning of śuṣka is explained in ĪPVV III p. 234 śuṣkah kartrtātma-svātantryalaksanobhayamelanāimakaraśaśūnyah; i.e. without the 'juice' that is represented by conscious dynamism.

30 The previous kārikā concluded the argumentations put forward by Utp. against the Buddhist conception of causality, based on the absolute otherness and isolation of single entities. A different argument must be addressed to the followers of Sāṃkhya who, in admitting the pre-existence of the effect in the cause, maintain the continuity and substantial unity of things. But the thesis of the Sāṃkhya (also criticized in the ĪS), as the following kārikā points out, cannot figure as a real alternative to the Śaiva thesis.

31 The Sāṃkhya conception of causality as the continual modification of a primordial nature (prakrti, pradhāna) to forms that are constantly new and already implicitly contained in it, in the end, enters into the Śaiva scheme of the kartrtā: the modification of the prakrti is precisely its being the agent of the action of modification. The Śaivas would have no objection were it not for the fact that the prakrti as conceived by the Sāṃkhya - i.e. jada - has no right to assume this role, which is the exclusive prerogative of the conscious being (cf. the following kārikā).

II 4.19 tion. On the contrary, it is possible in the case of a conscious unitary reality.

• This is not possible for an insentient reality, because its nature which is single would conflict with its appearing in differentiated forms. On the contrary, it is possible for an absolutely limpid (svacche), unitary, conscious reality, because there is no conflict here between its unity and its capacity to receive manifold reflections32. - 19 -

20. Even if the unity of consciousness is maintained33 to be the only ultimate reality, there cannot be action, for two entities divided as regards the nature of their manifestation (ābhāsabhinnayoh), without a preliminary act of thought which grasps and establishes the unity (ekatvaparāmarśam), characterized by the desire to act.

• Even if one posits the conscious principle as the only reality, a differentiation of the manifestations conceived of as random (i.e. without a cause) is not admissible34, and on the other hand there is no action in

32 The I, according to a simile which is very dear to these schools, is like a clear mirror serving as substratum for the manifestation of objective reality. Owing to its absolute transparency it remains unmodified by the apparition of these forms, which are in a sense different from it. It is this very purity that causes what is actually not separate from it to appear, instead, as separate (ĪPVV III p. 243 etad eva hi tannirmalatvam yat svātmānatiriktasya api atiriktasya iva avabhāsanam). See also SPr I.54-55 atyantācchasvabhāvāt sphatikasyā yathā svakam / rūpam paropraktasya nityam naivopalabhyate // tathā bhāvasamāyuktam bhagavams tāvakam vapuh / atyantairmalatayā prthak tair nopalabhyate // «Just as due to its absolute transparency, the intrinsic nature of crystal, that is continually coloured by other things, is not perceived, so indeed, O Blessed One, your body, that is united with the various beings, due to its absolute limpidity is not perceived without them».

33 Utp. now turns his criticism to the śāntabrahmavāda, which might avail itself of what is stated in the preceding kārikā to propose itself as the definitive truth, able to overcome all the contradictions pointed out in the theories of causality examined so far.

34 The multiplicity and diversity of manifestations cannot be conceived of as being purely random, because the order and regularity of reality would remain inexplicable. The multiplicity must therefore have a cause and this cause cannot be anything but action. However, in the conscious principle as understood by the Vedāntins (śāntabrahmana) action is by definition absent, and - as the vrtti explains - every insentient reality is incapable of action. Action, as the bridge between the one and the many, must necessarily contain as a precondition an awareness of the not absolute otherness of the two terms, in order that these may play the roles - which are distinct yet intimately coordinated -, for example, of kartr and karma. As Abh.

186

this conscious principle. If, however, it renders externally manifest II 4.20

through an act of determinate thought combined with a desire to act in

this way, then action is possible35. On the other hand, an insentient reality

cannot even be the agent of the action of being - 'it exists, is' - since it

does not possess the freedom that is manifested through 'wanting to be'

(bubhūṣāyogena)36. Thus the ultimate truth in this regard is that the

knowing subject, and he alone, 'causes' the insentient reality 'to be'

(bhāvayati), or, in other words, appears in various forms such as mount

Himācala and so on. - 20 -

21. Therefore causality, agency, action are nothing but the will of Him

who wishes to appear in the form of the universe, in the various

manifestations of jar, cloth and so on.

• The very will of Him who is free and has consciousness as His

nature to appear as universe constitutes His being cause as regards the

universe, in the form of agency; and this is the power of Action. In this

way action in the primary sense (mukhyā) - i.e. the desire to act -

belongs uniquely to the agent, one, whose essence is consciousness.

says (ĪPV II p. 205), action, which is single, could not be the svabhāva of two things,

if these were utterly different form one another. This moment of unity occurs in the

first attempt towards action (cikīrṣā), when the object is still completely immersed

in the I. It must be pointed out that Abh.'s commentaries give three or four different

interpretations of the kārikā, according to whether ābhāsabhinnayoh is made to

depend on cikīrṣā, parāmarśa or kriyā respectively; however, these interpretations

are not incompatible and, on the contrary, broaden the implications of the meaning

as a whole.

35 Action, Abh. explains, following the țikā (see text note 207), is the 'rendering

manifest' (ābhāsanā), whose essential feature is constituted by the very first impulse

of thought and will to act (parāmarśa - cikīrṣā or icchā) which already contains in

embryo the future manifestation and externalization (cf. ŚD I. 19-20a and vrtti).

36 A certain action may be attributed to an insentient reality (ŚDVṛ p. 159 ghato

'rthakriyām udakāharanādikām karoti), only in a figurative sense; strictly speaking

it would not be permissible even to say that it 'is', unless one takes into account that

this sattā that practical experience ascribes to it is in reality śivatākhyā; see above,

Introduction pp. XV-XVI. This point is developed by Utp. in the vrtti on ŚD IV.

32b-33a, where Somānanda solves in the same terms the problem of the attribution

in discourse of the qualification kāraka (and thus of case endings) to entities that

are by definition non-existent, such as the antelope's horn etc. Bhartrhari would have

faced the problem by referring to the distinction between mukhyasattā and vastvartha,

on the one hand, and upacārasattā and śabdārtha, on the other. On the related

question whether (kārakas') question presupposes animation see Cardona 1974.

187

II 4.21 There is no object of action without an agent; action, in fact, is attributed to the object etc. only in a figurative sense, through the agent37. - 21 -

37 All the other kārakas may even be absent from the sentence, but the presence of the subject is enóugh to ensure the carrying out of the verbal action, and, vice versa, even if all the other kārakas are present the absence of the subject alone makes the action impossible (ĪPVV III p. 253; cf. VP III. 7. 101-2 and Helarāja's commentary). Cf. SPr II. 54 prakāśate samvid ekā tadanyat tu prakāśyate / prakāśyam ca bhavet karma tac ca kartrā vinā katham // «Consciousness alone shines; that which is other from it is illuminated. What is illuminated is the object, and how can the object subsist without a subject?»

III 1.1

SECTION III. REVELATION

CHAPTER I

1. On the basis of the preceding argumentations (evam)1, action, which consists of an internal and external level, subject to temporal succession, pertains to the knowing subject alone2: therefore cognition and action are inseparable from one another.

1 This is the sense that is expressly indicated by Abh. (IPVV III p. 257).

2 This knowing subject is, first and foremost, the supreme knowing subject, Parameśvara, but also every limited subject, insofar as Śiva is his model and ultimate essence. What the kārikā expounds is therefore the śivatattva (IPVV III p. 258 śivatattvalakṣaṇasya parāpramātṛ); all the tattvas, from Sadāśiva (see the following kārikā) are his modes of being, or so-to-speak internal modifications, he is the sole reality (ibid. p. 257 śivatattvam hi sarvapadārthānām vapuh, p. 263 śivatattvam eva kam vastut). Utp. does not place this supreme reality (called īśitṛ in the following kārikā) beyond the scale of the tattvas, but at the top of it, as the union between śiva and śakti. His powers are absolutely pure (cf. above I.8.11) in that they are without object. If at the pati level the subject sees the world as his own body (III.2.3, IV.1.4), at the paramapati level we cannot even speak of the knowable (IPVV III p. 367 paramapatau tu bhagavati prameyakathaiva na asti, ibid. p. 266 na ca śiva pade viśvam kiñcit). All the universe is completely dissolved into the I; śivata is precisely the plane where only the “I” exists. At various points in the ŚDVr a (not constant) distinction is made between Śiva on the one hand and Paramaśiva (or Parameśvara) on the other seen as the all-inclusive form (p. 3); elsewhere (p. 27) the God is expressly said to assume the form of the thirty-six cciples, all indistinctly understood as kārya (... tattvarūpam sattrimśattattvasamkhyam kāryam rūpena bibharti). It must be borne in mind, however, that what Utp. primarily intends to do in this work is to render explicit the teaching of his master Somānanda. Bhāskara (Bh II p. 211 ff.) distinguishes between prakriyāśāstra (IPK is one of these), whose aim is to describe the nature of the tattvas (tattvasvarūpakathana), and sāraśāstra (e.g. the Mālinīvijaya), devoted to the supreme non-duality. According to the latter, within Sadāśiva - since it has the nature of śāntabrahmant - there is not a full contact with Śakti; thus Śakti is posited as a separate principle (the thirty-fifth) above Sadāśiva and is to be united with Śiva (the thirty-sixth) distinct from Sadāśiva. Though in reality Śiva and Paramaśiva are one single thing, Anāśritaśiva refers to his specific nature (svarūpanirdeśa) and Paraśiva refers to his pervading the whole scale of principles. However it is not

189

III 1.1 - Action, subject to temporal succession, is nothing but externalization; it pertains to the knowing subject, the self, which resides within. Thus cognition and action are mutually inseparable in him. - 1 -

2. But by virtue of the pre-eminence of the internal state there is, at the beginning, the Sāākhya principle3; then, with the predominance of externality, the Pārameśvara principle.

• But when in the Lord (īśituḥ), between the planes of internality and externality, internality has pre-eminence, there is the Sādākhya4 principle; then, with the increase in externality, there is the principle of Īśvara5. - 2 -

3. Īśvara is opening outwards (unmeṣaḥ), Sadāśiva is closing inwards (nimeṣaḥ)6; True Science (sadvidyā) is the state of the notions of 'I and 'this' having an identical substratum.

legitimate - Bhāskara continues - to make it into a thirty-seventh principle, because vyāpra and vyāpaka cannot be placed on the same plane, namely, vyāpaka cannot be collocated in the same series as the vyāpya, though higher up (the same applies to those who posit a further thirty-eight principle). The fact that the description given by the prakriyāśāstra stops at the śivatattva may be explained, again according to Bhāskara, in two ways: either they were only aiming at giving a differentiated description of the tattvas, without intending to proclaim the supreme non-duality (to which in any case they constitute a way of access); or they do not mention a reality beyond the tattvas in order not to expose it to objectivization.

3 'Principle' (tattva), according to Abh.'s definition (ĪPVV III p. 264, ĪPV II P. 219), is a reality that extends, as a unifying element, to a set of distinct entities, such as earth for hills, mountains, countries etc., or water for lakes, rivers, seas. See as regards this MPĀ, vidyāpāda, V. 3-4 and Rāmakanṭha's commentary. Cf. also the (implicit) etymological interpretation that Utp. gives of it (ŚDVṛ p. 27) viśvaprapañcatananaksama 'capable of extending the deployment of everything'; there is yet another definition in TP 73. The tattva refers to the objective dimension of a certain plane of reality, that is, to the content and quality of the experience of the subjects who pertain to those levels (ĪPVV III p. 274 vedyabhāvaniṣṭhā daśā tattva-svarūpā tadavabhāsayitṛsuddhamātrvedyavastusādhanā).

4 Abh. (ĪPV II pp. 217-8) gives more than one explanation of the word Sādākhya: the derivative with vṛddhi has here the meaning of 'that is in sadākhya' (cf. Pāṇini IV. 3. 53). where sadākhya means the plane on which the notion of 'being' (sat) first appears; or of 'what is the meaning of Sadākhya (understood as a synonym of Sadāśiva)'. The word Sadāśiva, again according to Abh. (ĪPVV III p. 264), means that the nature of Śiva does not cease to be present, even if the object begins to arise.

5 Parameśvara, as the vṛtti makes clear, stands for Īśvara here.

6 Nimeṣa refers to the moment in which the 'this', which in Śuddhavidyā becomes fully evident and faces the I (still, of course, in the sphere of consciousness

• Īśvara and Sadāśiva are respectively opening outwards and closing inwards, the condition of externality and internality. Pure Science (śuddhavidyā) is the plane of him who, having all things as his essence, thinks 'I am this universe', where the two terms have the same substratum (sāmānādhikaranyena)7, without any differentiation between external and internal, cognizable object and cognizing subject, both resting on the sole reality which is pure consciousness. - 3 -

4. [The latter principle is called sadvidyā] for in this things, having reached the plane of object of cognition and the level of 'this', are known by what is their true reality, as their essence is consciousness.

• Things, though they have become cognizable object in the form of 'this' on the plane of cognizable reality differentiated [from the subject], are known by their true reality in the terms of 'this is I', insofar as their essence is none other than consciousness; this is the sense conveyed by the adjective 'pure' [in Pure Science]8. - 4 -

and non-differentiation), is still evanescent and veiled by the I which completely overcomes it, as at the dawn of creation or the twilight of cosmic dissolution. The I-this relation is reversed in the Īśvara principle; the 'this' in Sadāśiva and in Īśvara are related to one another in the same way as a sketch is related to a finished painting. There is no real differentiation between the Sadāśiva, Īśvara and Śuddhavidyā principles; the 'identical substratum' which, according to Utp., characterizes Śuddhavidyā is also an essential element for the first two, which show the disequilibrium between subject and object though they have unity as their common basis (in the one case - cf. IPVV III p. 266 - it is idam that is projected in aham and vice versa). The texts seem to be quite in agreement on this, though there are certain discrepancies in the description (IPVV III p. 266 asau vyāpāro vidyābhyah / ubhāv api ca tadvyāpāramayau paramaśivanāthasya ekaghanam aiśvaryam; IPV II p. 222 tada-dhiṣṭhātrdvayagatam karanam vidyātattvam āha; PTV p. 224 bhagavatsadāśiveśadaśā śuddhavidyāmayī; etc.).

7 Sāmānādhikaranya is, in grammatical terms, 'co-referentiality', having the same case, such as the subject and nominal predicate of a proposition. The plane of māyā is characterized by the opposite condition, prthagdhikaranya: subject and object appear as two realities unrelated to one another: 'they have a different substratum'.

8 Abh. (IPVV III p. 273) refers to the MPĀ to define the nature of Śuddhavidyā. Indeed, the tenth patala of MPĀ's vidyāpada (with Rāmakantḥa's vrtti) is devoted to the vidyātattva. Vidyā is the basic component of human knowledge, which takes place through the various forms assumed by the buddhi (determination, intuition and so on) (vrtti p. 311 adhyavasāyasmrtipratibhādipratyayabhedabhinnā buddhir api yayā vedyate sā vidyā tat param karanam); Vidyā embraces it entirely,

III 1.5 5. Here9 there is imperfection (aparatvam) on the one hand - because things are manifested as other than the self -, perfection (paratā) on the other - because they are veiled by the I. This is in fact the perfect-imperfect state (parāparadaśā).

• Here there is imperfection because there is the notion of 'this', perfection because all the cognizable is veiled by the I10; this is therefore the perfect-imperfect condition. - 5 -

6. Others11 maintain that Vidyā is the idea of differentiation with respect to things, present in the one who is, however, for his part, [fully]

in the sense that it 'knows', is aware of, all forms of cognition (X.5 sā sannātīva satatam sarvāṅgāliṅgane kṣamā / na tathā prākṛtī buddhir bahiraṅgā yatas tu sā //). However, Vidyā is bound to the limited human condition (X.20a pauruṣam bhāvam āśritya), and, as such, it illfuminates the ordinary cognizable reality (X.15b vedyārthan pratipādikā) but it cannot reveal the highest principles (X.20d na śivārthan pradāyikā). This task pertains to Śuddhavidyā, which transcends the plane of māyā akala subject and still less in the mukta -, in his consciousness, that the whole universal deployment, from the lowest plane, is the object of knowledge (X.23 śivādyavaniparyanto yo 'yam adhvātivistṛtah / sa sarvaś ca citau jñeyo nāśivatvād vimucyate //).

9 Or in other words, as Abh. comments (ĪPV II p. 227), 'in the two principles (Bh: Sadāśiva and Īśvara)'.

10 Cf. ĪPVV III p. 274 aparatvam apūrnatā anyākān kṣitvam idam iti, paratvam pūrṇatvam aham iti.

11 According to a conception that Utp. does not present as his own (Abh., ĪPV II p. 228, attributes it to the Rauravāgama), Śuddhavidyā is the principle in which differentiation appears, restricted, however, only to the sphere of the object: the subject, in other words, without declining from its own total absorption in pure consciousness (therefore no identification with a mind, body etc.) regards the object as other than himself. It is the condition of the Vidyeśvaras - which we will return to later - who, omniscient and omnipotent, face a reality considered other than themselves. It may be added, again on the basis of Abh., that what differentiates this conception from that of the preceding kārikā is that there Vidyā represents in the sphere of the consciousness 'I-this' the predominance of the I which veils the 'this'; whereas here it is the predominance of 'this' veiled by the I. The introduction of differentiation determines a certain assimilation with māyā (the Raurava and other texts in fact call Śuddhavidyā also by the name of Mahāmāyā); cf. Rauravāgama, vidyāpāda. IV. 28b māyopari mahāmāyā sarvakāraṇakāraṇam, a māyā, however, which is not fully developed (aprarūḍha) because the 'this' in it is incapable of developing in the sense of full duality, veiled as it is by the I (the 'purity' of Śuddhavidyā consists precisely in this). Mahāmāyā

an agent having consciousness as his essence, as in the case of the III 1.6 Vidyeśvaras: this idea of differentiation is similar to that caused by the power of māyā.

• Some consider Vidyā as the view of everything as differentiated [from the I], pertaining however to a subject [fully] endowed with consciousness and agency. This Vidyā is also the power of māyā. Being beyond the flow of existence, the Mantresvaras and the Vidyeśvaras reside there12. - 6 -

7. The power of Vidyā reveals in the state of 'beast' (paśubhāve) the true nature as Sovereignty; the power of māyā, on the contrary, conceals it13.

is sometimes (cf. e.g. PTV p. 225) considered a distinct tattva - between Māyā and Śuddhavidyā - in the need to assign an abode to the vijñānakevalas, which on the hierarchical scale of subjects occupy an intermediate position between the pralayākalas and the vidyeśvaras etc., located respectively in Māyā and Śud-dhavidyā; in other texts it is, on the contrary, included in Śuddhavidyā or in Māyā, depending on which of its aspects - mentioned above - one wishes to refer to (see Śataratnasamgraha II. 24-27 and comm., ŚDVṛ p. 25; see also Dvivedi 1983: Upodghāta, 139-141).

12 ĪPVV III p. 277 explains, following the tīkā, the co-presence in this principle of the Mantresvaras and Vidyeśvaras on the basis of the essential unity of Mantras and Vidyās (śuddhānādiparamarśasāraśabdarāśikalāśarīratvena); the difference is that in the former mukti, samvedana and śaktimat predominate, while in the latter siddhi, vācya and śakti predominate.

13 The conception expressed in this kārikā (attributed by Abh., ĪPV II p. 231, to the Sadardhasāra, a text of the Trika school) seems to be that favoured by Utp., since it is not introduced as pertaining to 'others' (ibid. anye ity anukteh). Abh. formulates it in the avataranikā to the ĪPV as follows: in Sadāśiva the differentia-tion is not evident, icchāśakti is in action; in Īśvara the differentiation is evident, the śakti is jñāna; in Śuddhavidyā (literally 'in the Vidyeśvaras') the differentiation is not only evident, but also fully developed (prarūdha) - but only as regards the plane of the knowable, subjectivity is untouched - kriyāśakti is in action. After that, with the differentiation which also embraces the sphere of the I, one enters the sam-saric world, the realm of māyāśakti. Vidyāśakti is also in action on this plane, having the opposite function from that of māyā, that is, it discloses true reality to the yogins and the jñānins in whom the error of differentiation has been overcome and survives only in the state of karmic impulse. The parallel avataranikā in ĪPVV contains some variants and seems more in line with Utp.: jñānaśakti in Sadāśiva and kriyāśakti in Īśvara (as in ŚD II. 1); a residual trace of kriyāśakti constitutes, on the contrary, the power of Śuddhavidyā, also called Mahāmāyā.

193

III 1.7 - In beings in the flow of existence there is the recognition of the self as Sovereignty thanks to Vidyā14 and the state of 'beast' through the action of māyā. - 7 -

8. On the contrary, when absolute differentiation appears, and the void, the intellect or the body - which are other than the self - are considered as 'I', it is then that the power of māyā displays itself.

• When things appear exclusively as 'this', as differentiated [from the subject], and entities such as the body and so on appear as the cognizing subject - the 'I' -, then the power of māyā called 'obfuscating'15, which is the cause of these two errors, displays itself in the Lord. - 8 -

9. That cognizing subject identified with the void etc. who faces a cognizable reality assumed to be separate [from himself], such a subject - who is in fact himself an object16 - is enveloped by the pentad constituted by Time etc.

14 On vidyāśakti and its degrees cf. ŚDVṛ pp. 31-32, 25; see also below III. 2. 3.

15 Abh. (ĪPVV III p. 285) quotes a passage from the Rudrayāmalasāra (māyā vimohinī nāma kalāyāḥ kalanam sthitam), to which the vṛtti probably alludes.

16 All the various planes - from the body to the mind and finally the void - with which a person believes he can identify his own subjectivity and which he opposes to an objective reality considered 'other', are on the contrary, in the final analysis, part of it, as they too are objects compared to the authentic subjectivity constituted by consciousness (cit). But if in one way they are not identical to consciousness they are however an expression of it, there not being anything substantially other than consciousness and that does not ultimately merge with it. The error is therefore a double error. paradoxical like the nature of māyā which lies at its basis: identifying consciousness, the I, with what is not consciousness is māyā, but it is māyā too which causes what in reality is not different from consciousness to appear as other than it. In order that the reality of the object may be conceived as being differentiated from the I a degradation of the I is necessary. Without this identity crisis the I, in its fullness, could only embrace the whole as his own self and the word 'this' could not be uttered.

17 The manifestation of the present, characterized by vividness and clarity, is the point of reference for conceiving a past and a future. The seat of the experience of time, Abh. adds (ĪPVV III p. 287), is primarily the empirical subject and only secondarily things, which receive their temporal colour from the time of the subject with whom they are associated.

18 Utp. himself, according to what Abh. reports (ibid. pp. 290-1, see text note 31), acknowledges two interpretations of the expression karmaphala: 'the action and

• That cognizing subject identified with the void etc. who faces the whole of cognizable reality conceived as separate [from the self], is, in fact, himself the object of cognition as 'this' and is conditioned by the five priciples, Time and so on. From Time he derives, through the manifestation of the present, the existence of the past and of the future17; from Necessity the strict concatenation of [cause with] effect and of action with its fruit (karmaphalaniyamah)18, from Attachment the yearning for fruition; from Science and Force this subject - who, however, is not identified with consciousness nor is he free - derives a partial knowledge and activity, essentially constituted by consciousness and freedom. - 9 -

10-11. Cognizable reality, distinct in products and instruments, appears in twenty-three forms; it has a single, undivided, radical cause, the pradhāna. The series of internal and external instruments has thirteen aspects; the class of products, divided into gross and subtle, has ten.

• The products are divided into two groups. Five are constituted by sound, tangible shape, colour, taste and smell and are called tanmātra because of their subtle nature19; by combining them one has the gross elements (bhūta, mahābhūta) is referred to in both the Sāmkhya and Āgama

its result (i.e. the karmic impulse that derives from it)' or 'the karma and its fruit (i.e. the heaven etc.)'. The principle of Necessity is therefore that which ensures in the subject the apparent validity and fixity of the cause-effect relation in everyday experience, so that what is essentially the Lord's creativity may appear as the capacity of one thing invariably to produce another.

19 The relation between the series of subtle elements (tanmātra) and that of the gross elements (bhūta, mahābhūta) is referred to in both the Sāmkhya and Āgama texts as the relation between universal and particular. The tanmātra represents the archetypal, quintessential form of the relative mahābhūta of which it constitutes the primary quality (sound-ether, tactile sensation-air, etc.), though not the only one, as all the schools are forced to admit. In fact experience shows that - with the exception of ether - all the gross clements also have some of the qualities that are not considered co-essential to them; sound, for example - says the MrA, vidyāpāda, XII. 17 and relative vrtti - is present not only in ether, but also in air, fire, water and earth. Thus two distinct conceptions emerge. According to the first every gross element possesses only one single quality and, if we perceive others, this is due to its combining or connecting (samparka, sanniveśa) with other elements. Abh. (ĪPVV III p. 299) quotes, with refcrence to this, a line from the MPĀ, ksubhitāt sparśatan-mātrād dviguno vyūhato marut (not found in the published text; but cf. vidyāpāda, XX. lcd-2ab yat tat prāk sparśatanmātram tāmasam samudāhrtam / tatkṣobhād abhavad vāyur dviguno vyūhalakṣaṇaḥ). which he glosses with pūrvabhūtantara-vyūhayogāt. This is the thesis of the ancient Sāmkhya, which will be later supplanted

195

III 1.10-11 ments called earth etc., of which there are also five, and they are essentially no different from the former (tāny eva)20. The ear, the skin, the eye, the tongue and the nose are the five instruments of cognition; the voice, the hands, the feet, the excretory and genital organs are the five instruments of action. The internal organ is threefold: the mind, the intellect and the sense of the I. This cognizable reality, composed of products and instruments and divided into twenty-three forms, has a single and undivided plane called pradhāna, which constitutes its radical foundation. -10-11-

in the classical Sāṃkhya by the thesis of the accumulation of qualities (ether has sound, air has sound and tactile sensation, and so on, until one reaches earth which possesses all five), which is that generally accepted in the theoretical section of the Āgamas (e.g. Kālottara, Pauskara etc.) and presupposed by ritual. Furthermore this accumulation is generally explained as deriving from the fact that the individual tanmātra generates the relative mahābhūta not in isolation, but by associating itself with the other tanmātras in various ways (see Pauskarāgama, pumstattvapatala, II. 243-4, cit. in Śaivaparibhāṣā p. 125; MPĀ, vidyāpāda, XVIII. 102-104ab, etc.).

20 This seems to mean that the gross elements do not have a different reality from the essential qualities that compose them, combined and separate (cf. the fragmentary quotations from the ṭīkā on this point in the vrtti, cited by ĪPVV III pp. 299-300, ‘tāny eva’ iti vrttim vyācaste ‘tāny eva ca’ ityādinā ‘natu vastvantaram’). This principle is clearly formulated in TĀ IX. 289: «In this regard the earth etc., as is attested by direct perception, are nothing but an aggregate of qualities, neither separate or different from them, does any subject appear that may be called earth and so on».

III 2.1

Chapter II

1. Such being the structure of reality1, Rudra is the deity that presides over the state in which this2 subjectivity exclusively subsists; Brahma and Viṣṇu3 reside in the flowing of the differentiated cognizable reality.

• Rudra is he who presides over the condition of pure [limited] subjectivity represented by the void or by an extremely subtle body formed by the puryaṣṭaka4 - in other words, over the condition in which there is a reabsorption of every other cognizable reality. Brahma and Viṣṇu preside over the manifestation of differentiated cognizable reality, the former causing creation and the latter continuity. - 1 -

2. This cognizer blinded by māyā, bound by the karma, is immersed in the saṃsāra; but once Science (vidyā) has made him recognize his own nature as Lord, then, his essence being solely consciousness, he is called 'liberated'.

• The māyic cognizer who is identified with the void and so on, depending on the karma because of the law of necessity5, is in the power

1 This is how I understand tatra - not glossed by Utp. - in conformity with Abh.'s interpretation.

2 'This', i.e. the limited subjectivity described in III. I. 9.

3 Brahma, Viṣṇu and Rudra along with Īśvara, Sadāśiva and Anāśrita are the so-called Kāraṇas or Kāraṇeśvaras, the hypostases of Śiva situated on the various planes of reality in an order of increasing ontological dignity. Each of these deities presides over (adhiṣṭhā-) a plane of reality, brings his devotees to this plane and, vice versa, becomes directly present to those who are absorbed into the contemplation of that level of reality (IPVV III p. 305). Abh., commenting on the tīkā, connects them with forms of knowledge, levels of the word and locations in the human body. Brahma is associated with the first moment of cognition and manifestation (prātha-makalpikī srṣṭih), with the first effusion of paśyantī in the form of pratibhā and his locus is the heart; Viṣṇu is associated with the continuing of cognition, as in memory and dream, with madhyamā and vaikharī, his locus is the throat; Rudra with the reabsorption of all discourses, the first step into the supreme word and his locus is the palate.

4 On the puryaṣṭaka see below pp. 204-205.

5 That the principle of 'necessity' is at the root of the karma is stated in the tīkā

197

III 2.2 of the samsāra. With the recognition of the true reality of the self attained thanks to Science he is free. - 2 -

3. The cognizer is called 'lord' (patiḥ) when things appear to him as constituting his own body. When they appear to him differentiated due to māyā, the subject, polluted by the various contaminations, by the karma etc., is called 'beast' (paśuḥ).

• On the plane of sovereignty (aiśvarya°) the cognizer who sees the universe as his body is 'lord' (patiḥ). On the plane of limited individuality (pumstva°) the subject, enveloped by contaminations - attachment and so on⁶ -, by actions, by the ripening of the fruits that derive from them and the latent impulses determined by them, is a 'beast' (paśuḥ). - 3 -

4. Consciousness that is devoid of freedom and, vice versa, freedom devoid of consciousness: these are the two forms of maculation - so called because it obliterates one's own true nature - called āṇava7.

• The absolute reality is free consciousness; only insofar as this is its essence is it in perfect fullness. Since it thus wills, in virtue of its very freedom [there is the āṇava maculation which is twofold]: 'maculation' insofar as it involves error as regards the self's real nature and 'twofold atomicity' insofar as consciousness becomes limited with the loss of its fullness, creating a subjectivity in which only the light of consciousness subsists without freedom, or only freedom without consciousness. - 4 -

5. There being the āṇava8 maculation (atraiva), the apparition of a cognizable reality differentiated [from the subject] constitutes the maculation called māyā; that, then, which in a subject endowed with agency, but lacking the full light of consciousness, determines births and ex-

on the following kārikā, as may be gathered from ĪPVV III p. 312 āha 'niyatyā' iti niyatir yataḥ karmano mūlabhūmiḥ; see also vrtti on III. 1. 9.

6 All these elements, i.e. the contaminations (kleśa) - avidyā, rāga, asmitā, dveṣa and abhiniveśa - and so on, are listed in Yogasūtra II. 3 ff.

7 The fundamental maculation, 'concerning the anu' (āṇava), or the reduction of consciousness to a minimal, 'atomic', state (anu).

8 The maculation āṇava, which is often simply called 'maculation', is necessarily the basis of the other two and is, in the course of recovering consciousness in its fullness, the last to disappear.

periences is the karmic maculation. All three are solely the work of the III 2.5

power of māyā.

• When, there being this twofold ānava maculation, cognizable

reality, though in reality not separated from the subject, appears differen-

tiated from him, there is, on the basis of this other error, the maculation

known as māyic. The karmic maculation, which pertains to the agent

devoid of the light of consciousness, is due to the error of considering

the cause of births and so on the actions, which on the contrary - as has

been established - are not causes9. The Lord's will to create these three

maculations is called the power of māyā. - 5 -

6. Those who do possess pure consciousness but not the agency to

the highest degree (uttamakartrtā) are created by the Lord as separate,

distinct from the self, due to their being devoid of agency10.

• Some subjects, though having a consciousness not stained by the

cognizable, are created by the Lord as different from what is their truc

nature, because they are devoid of the full agency. - 6 -

7. Though having the same characteristics - consciousness etc. -

these subjects are differentiated from one another because of a particular

will of the Lord: they are the Vijñānakevalas11.

9 The error on which the karmic maculation is based is therefore that of attribut-

ing to the actions and karmic impulses determined by them the status of cause as

regards birth, length of life, quality of fruitions etc. Actions, being insentient realities

(jada), cannot, according to the conception expounded in the Kriyādhikāra, be the

cause of any thing whatsoever; the sole real cause is the agent, the Lord.

10 This is, in particular, a reference to the Sāṃkhya conception of the puruṣa

in the state of kaivalya (cf. ĪPVV III p. 319). The use of the expression 'are created

(nirmita) by the Lord' points out how all these various maculations and levels of

subjectivity are merely a 'construct' created by the free dynamism of Consciousness.

11 Here begins the description of the various categories of subjects. Utp. ex-

pressly mentions the Vijñānakevalins, the Pralayakevalins, the Vidyeśvaras and the

Sakalas, implicitly referring to the more classical division (see MV I. 15 ff.) which

lists seven levels of subjectivity - Śiva, Mantramaheśvara, Mantresvara, Mantra,

Vijñānākala, Pralayākala and Sakala - (see e.g. TP, MrĀ, TĀ, PH etc.). There are

variations in the texts as regards the attribution and location in the scale of the prin-

ciples; in the latter case this ambiguity is often due to the confusion between on-

tological plane and field of action. Besides this there are other divisions, based on

different criteria (e.g. the well-known one of the Svāyambhuvāgama in kevala, sakala

and amala; see Sataratnasamgraha 1. 34); cf. Dvivedi 1983: Unodghāta. 134-35. As

199

III 2.7 - These subjects characterized by the pure light of consciousness are differentiated from one another, though there is no differentiation as regards their attributes of consciousness, permanence etc. This occurs because they are created thus by the Lord and for no other reason. This class of subjects, corresponding more or less to the purusas of the Sāṃkhya, has the name of Vijñānakevala12. - 7 -

8. The agents that are identified with realities other than consciousness, such as the void etc., are the Pralayākalas13. Karmic maculation also pertains to them; on the contrary, māyic maculation may or may not be present (vikalpitah).

• The Pralayakevalins assume as I the void, the prāṇa etc., not grasping their having consciousness as true nature. In these limited souls the karmic impulses are also present; on the contrary, the māyic maculation regards the ŚD, first (I. 31) vidyā, vidyeśa, mantra and mantreśvara are mentioned, then (I. 41c-43) paśu, śānta, kevalaśa kevalaśambhu, pralayakevalin and vijñānakevalin. The division of the subjects - mostly, as has been said, into seven classes - may be considered the common heritage of Shaivism in the broad sense, whilst keeping in mind the remark of Jayaratha (TĀV vol. VII pp. 7-8), according to which this differentiation between subjects is particularly appropriate to the dualist schools such as the Siddhānta (siddhāntadarśanādisamucitam).

12 The state of Vijñānakevala (or Vijñānākalaś), Abh. explains (ĪPVV III p. 322), may be attained through a special initiation aimed only at going beyond the path of māyā without there being any conjunction with the supreme reality, or through the creative meditation (bhāvanā) of the puruṣa in the state of isolation, without the karmic bond, according to what is expounded in the Gītā; or through concentration (dhāraṇā), contemplation (dhyāna) and absorption (samādhi) that have as their object the path of māyā as far as this point of deployment. The attainment of this state is therefore a special kind of experience (bhogaviśeṣa). It is for this reason that one may speak of its being similar to the state of the Sāṃkhya puruṣas but not absolutely identical to it.

13 'Devoid of kalā because of dissolution' where 'dissolution' is generally the state of total absorption pertaining to deep sleep, and 'kalā' stands for the various principles that constitute the faculties and products (ĪPV II p. 152), listed above, III. 1. 10-11. The māyic maculation is present in the state of savedyasausupta, in the state of deep sleep in which the traces of objective reality still remain; this is the level of identification with the mind (buddhi). In the state of apavedyasausupta, to which the identification with the void and the prāṇa corresponds, every cognizable reality is completely swallowed up and with it the māyic maculation, which just consists in the apparition of a cognizable object differentiated from the subject. These states are analyzed in the following kārikās.

may or may not be present, depending on whether there is connection III 2.8

with cognizable reality or not. - 8 -

9. A class of subjects identified with consciousness and endowed with

agency experiences cognizable reality as differentiated; the karmic mac-

ulation having disappeared, the māyic maculation is present in them.

They are the Vidyeśvaras14.

• There are subjects who have transcended the karma and are iden-

tified with consciousness, being also endowed with agency15. These,

though having the nature of Vidyeśvara16, continue to be affected by the

maculation of māyā, being in relation with a differentiated cognizable

14 The Vidyeśvaras are a group of eight deities headed by Ananta, whom Śiva

invests with determinate functions; in particular, they are called to coopcrate in the

liberation of the limited souls, acting as intermediaries in the revelation of the teach-

ing of Śiva, etc. They employ another class of subjects as their instruments. the

seventy million Mantras; they are identified. as some texts indicate explicitly (see

MrV, vidyāpāda, p. 46, TĀV vol. VI p. 55; cf. also ĪPVV II p. 405), with the class

of the Mantramaheśvaras. It may be useful to mention the clear description given

by the Mokṣakārikā of Sadyojyotiḥ and the relative vrtti of Rāmakantḥa (particularly

ly II 73-94), obviously without wanting to impose it in every detail as the norm for

the whole śaiva tradition. In this the terms vidyā and mantra - and consequently

Vidyeśvara and Mantresvara - are interchangeable; the Vidyeśvara-Mantresvaras are

however divided into a higher group of eight headed by Ananta (which we can there-

fore call Mantramaheśvaras) and a lower group (apradhāna) of a hundred and

eighteen, composed of the hundred Rudras etc. The higher Mantresvaras are situated

ontologically below kriyāśakti. After them, but still above māyā, come the seventy

million Mantras, they too are divided into two parts, one half impelled directly by

Śiva and the other by the Mantresvaras. Then follow the hundred and eighteen

Mantresvaras which are mantraprāyojakas but have an inferior status, as they are

situated within māyā - and are consequently sakalas. Regarding their functions,

the higher Mantresvaras are distinguished from

the lower ones since only they perform the five activities of creation etc. in the

māyic world. All the Mantresvaras and the Mantras together attain supreme libera-

tion once Ananta ceases to exist, which happens after he has completed the five ac-

tivities, at the beginning of the mahāpralaya. As regards the discrepancies between

the texts it may be remembered, for example, that the PH places the Mantrama-

heśvaras in the sadāśivatattva, SvT X. 1060 and TP 1. 10 vrtti place the Vidyeśvaras

in the īśvaratattva, etc.

15 Unlike the Vijñānākalas.

16 The term Vidyeśvara expresses, according to Abh. (ĪPVV III p. 325), the con-

tact with knowledge and sovereignty (vidyā-bodha and īśvaratva-kartrtva). These

subjects however - unlike the nature of Sadāśiva and Īśvara, on which level dif-

201

III 2.9 reality. They do not possess the full faculty of action and are therefore distinct from the Lord and one another17, as has been previously seen; therefore, they too must be considered limited souls (anutvam). - 9 -

10. All the beings that are immersed in the flowing of existence, starting from the gods, are affected by the three maculations: but of these it is precisely the karmic maculation that constitutes the sole direct cause of the samsāra.

• The beings in the power of the samsāra, from the gods to immobile things, have all three maculations: but it is only the karmic maculation that is responsible for setting in motion the samsāra. - 10 -

11. And this consciousness-principle constituted by agency (kartrtā°), reinforced by kalā, being limited, comes to be - in the individual who is identified not with consciousness but with the void etc. - in a subordinate position18.

• In the beings immersed in the flowing of samsara this consciousness-principle, constituted by agency, in the sphere of insentient realities [assumed as subject] such as the void, the prāna and so on, corroborated

ferentiation is not yet manifested - conceive the objects of their knowledge and action as different from themselves due to the māyic maculation; they are, rather, similar to the Lord of the dualist doctrines.

17 See MrĀ, vidyāpada, IV. 5.

18 The consciousness-principle (cittattvam), supreme consciousness, in its fullness is constituted by the interpenetration of bodha (luminous, cognitive principle) and of kartrtā or vimarśa (dynamic principle which is expressed in the activity of the I). The maculation called ānava is responsible for the fragmentation of this unity by suppressing either one or the other of these two components (cf. above kārikā 4). In the samsāric world cittattvam is found devoid of the bodha aspect and reduced to only kartrtā, but no longer a full kartrtā (and not even an 'almost' full one, as it was for example in the Vidyeśvaras). It has now become a shadow of its former self and is only partially revitalized by the contribution of the kalā principle, 'the partial agency'. The consciousness-principle, thus limited, in the individual is therefore found in a decidedly secondary position, superseded by the whole concretion of inert realities with which the I is erroneously identified. In the verse (kalodbalitam ... cittvam) there is an obvious echo of passages form the scriptures; cf. v. I. 10 of the Svāyambhuvāgama (Government Oriental MSS Library, Madras. MS. 16797) kalod-balitacaitanyo vidyādarśitagocarah / rāgena rañjitaś cāpi buddhyādikaranair tataḥ // and the substantially identical one of the Tantrasadbhāva cit. in ŚSV p. 37 (ad III. 3); cf. also Kiraṇāgama I. 16. On the kañcukas (kalā etc.) see Torella 1998.

202

by the power of kalā pertaining to the Lord, is limited, being placed in III 2.11 a subordinate state. And therefore the status of limited soul occupies here a pre-eminent position. - 11 -

12. The pre-eminence, in the conscious self, of agency and consciousness, with the consequent subordination of insentient realities such as the void, constitutes knowledge whose distinctive feature is the absorption in it (tatsamāveśa9)19.

• On the contrary, when this same void etc. comes to be in a state of subordination as regards the consciousness whose essence is agency, then in the subject thus substantiated by this consciousness there is a knowledge characterized by the absorption in the power of that. - 12 -

13. When the seat of agency - 'I - is the void, understood as the absence of the mind etc., where only the karmic impulses exist (˚samskāramātrini) without any evidence or form, there is the state of the absence of the knowable20.

• When there is a subjectivity characterized by the identification of the I solely with the negation of the mind, the prāṇa and so on21, then in the state of deep sleep, which means residing exclusively in the above-

19 On samāveśa sec Introduction pp. XXXII-XXXIV.

20 This verse describes the state of apavedyasauṣupta 'deep sleep devoid of the knowable' which represents the plane of the identification of the I with the void. The void is here the negation of every reality that is the object of knowledge including the mind and the prāṇa, which is experienced, apart from in deep sleep in the strict sense of the word, also in the state of cosmic dissolution, stupefaction, or in the deepest states of samādhi. However, one can only speak metaphorically of there not being the knowable: in fact it does not entirely disappear (the śaiva texts often remind us that never be the complete annihilation of what exists) but it is rather reduced to the shadowy and vacillating state of samskārat.

21 Abh. explains (ĪPVV III p. 333) that the negation (niṣedha) already appears on the plane of aham-idam which extends as far as Īśvara (śivatattve īśvarānte); it is caused by the introduction of the idam, which precisely entails negation as differentiation (bhedan rūpam) from the consciousness and between objects. While consciousness remains predominant and negation performs only a subordinate role, we have as subjects the Mantramaheśas and the Vijñānākalas, which are differentiated on the grounds of possessing or not possessing vimarśa. On the contrary, when negation predominates there is the śūnyapramātr, which the present śloka is referring to; finally, when the objects predominate, there is the subject Sakala.

III 2.13 mentioned state, there is the absence of the knowable, in the sense that the karmic impulses, though they exist, are imperceptible since they are devoid of form. The difference from the Sāṃkhya puruṣa lies precisely in this. - 13 -

14. It is the inner activity (āntarī vrttih)22 of the sensory faculties, it is that which sets in motion the various breaths, the prāṇa and so on; it is called 'life', whose essence is the eightfold body. Or the subjectivity identified with the breath of life (prāṇe) has as its essence the eightfold body (puryastakātmikā).

• Precisely this knowing subject identified with the void - in which there exists an inner dynamism, the driving force of all the various breaths, prāṇa and so on, which is called 'life' and is common to all the sensory powers - is called 'living' (jīva°)23. These sensory powers combined with the I constitute the eightfold body. Or it is formed by the subject inasmuch as it is identified with the prāṇa24. - 14 -

22 The kārikā identifies āntarī vrtti with śūnyapramātr (ĪPVV explains that the identification is only metaphorical, in actual fact the real relationship is one of cause and effect); the vrtti places āntarī vrtti (antarvrtti) in a bahuvrīhi referring to śūnyapramātā. The tīkā and ĪPVV explain that this bahuvrīhi may be understood in two ways. In the first, antarvrtti is the very dynamism of subjectivity - present in the śūnyapramātr - that sets in motion the vital dynamism (jīvana) common to the sensory powers, a dynamism which is that of the five functions of the breaths of life of whose differentiation it is the cause. In the second, with a subtle distinction, it directly denotes the vital energy that is common to all the sensory powers and sets in motion the various breaths. The mention of antaryrtti in bahuvrīhi (depending on śūnyapramātā) is particularly pregnant; Abh. remarks that it aims at removing the role of cause from the sensory powers and leading them back to their total dependence on the agent who is the sole true source of their dynamism. The classical passage on this in SK I. 6, which the ĪPVV does not fail to quote.

23 Cf. ĪPVV III p. 334 tāṃ vrttim jīvanalaksanāṃ ... utthāpayan jīva ity ucyate.

24 The limited individuality, the 'living' (jīva), the transmigrating nucleus, therefore corresponds to the plane of the identification of the I with the void (śūnyāhantā) or the prāṇa (prāṇāhantā). It consists in a body called 'eightfold' (puryastaka), or, according to another traditional etymology, the 'ogdoade that is in the body', a kind of subtle body that forms the deep structure of the person and determines his continuity as the vehicle of the I in transmigration and the depository of the karma etc. There are various opinions on its composition; from eight elements - five tanmātras, manas, buddhi and ahamkāra (see Kālottarāgama XVII 4cd-5ab, SK III.17; see also Sāṃkhyakārikā 40 and Gauḍapādabhāṣya) -, which is probably the most widely

204

15. When the subject rests exclusively on these planes, this is the state of deep sleep - similar to cosmic dissolution - which may or may not be affected by the maculation of māyā, depending on whether the knowable is present or absent (savedyam apavedyaṃ ca).

• On the plane of the identification with the void, since there is no knowable reality, the maculation of māyā is absent. On the plane of the identification with the prāṇa etc. the state of deep sleep, which is similar to cosmic dissolution25, is affected by the maculation of māyā, since in

credited conception, to thirty (Rāmakanthha's Kālottaravrtti p. 126; see Torella 1979: 376, notes 35, 36); some texts are collected in Dvivedi 1983: Upodghāta, 132-133, note 3. In Utp.'s opinion, as gathered from ĪPVV III pp. 334-336, the puryaṣṭaka referring to the subjectivity identified with the void (śūnyāhantā) is constituted by the 'pentad' - the inner group of powers common to all the senses, which gives rise to the five particular breaths and makes the body's machine function; it therefore includes the class of breaths (as what is impelled), the class of the sensory faculties (as what impels) and the śūnyāhantā (as that presiding over the whole) (sa preryaṃ prānādivargaṃ prerakaṃ ca indriyavargaṃ adhiṣṭhātrīṃ ca śūnyāhantāṃ ākṣipya vartamānaḥ puryaṣṭakaḥ ity uktaḥ). But, as is noted, the current conception is that of the puryaṣṭaka as connected with the prānāhantā (prānāhantāpakse sarvasya abhimatam puryaṣṭakāvastānam). In this case one has: the pentad of breaths, combined with the three classes of faculties (those of knowledge, of action and the antahkarana) impelled by the ahaṃkāra which, however, is not included in the number since it presides over it (prānādipañcakaṃ ahaṃkārapāraṃ ahaṃkāreṇa gatrayeṇa sahitaṃ pañcakaṃ samvedanopadhānatvena adhiṣṭheyam prānādirūpaṃ aham ity evam pramātṛbhāvena adhiṣṭhitaṃ puryaṣṭakam). However, the picture that emerges from Abh.'s remarks is not completely clear. In conclusion Abh., still guided by the ṭikā, says that there is a 'most subtle' body (sūkṣmatama; is it the one connected with śūnyāhantā?), which is merely the rough sketch of the subtle body (pelavaśarīrasya hi tad āsūtranamātraṃ); and a body sthūlatama (the physical body).

The transmigrating body is only the intermediary one between the two (madhyama), in which there are also the tanmātras. Nevertheless, there is no real difference between the above-mentioned conception and the conception of the puryaṣṭaka composed of five tanmātras plus the three components of the antahkarana, since the sparśatanmātra also embraces the five breaths (sparśatanmātreṇa prānādisvīkārāti).

25 The similarity consists in the fact that both in the state of cosmic dissolution and in deep sleep the constituent elements of the person (the body and so on), like those of external reality, no longer exist. But this non-existence is real in the case of cosmic dissolution, whereas in deep sleep there is merely the loss of the consciousness of them, which may come about through an effort of will in the case of the samādhi, through ingesting a certain substance as in drunkenness, through exhaustion as in sleep. The state of deep sleep is also distinguished from dissolution by its brevity (cf. ĪPV II p. 265).

III 2.15 this state there is relation with the knowable, such as the contact with pleasure and so on26. - 15 -

16. The clearly manifested (spaṣṭāvabhāsā) creation of things in the mental sphere alone which are mistaken for objects of the senses, is called the dream state.

• The dream state for the limited soul is the creation of objects - which appear in the sphere of consciousness as if they were really being seen, etc. -, a creation that occurs through the power of the mind alone without the action of the powers of the senses such as sight etc.; this creation is the work of the Lord27. And this is an illusion inasmuch as this perception, e.g. of a form, common to other subjects, after having lasted as such [for the length of the dream], does not continue later28. - 16 -

17. The creation which is stable, in that it is the object of all the senses and external, common to all knowing subjects, is the waking state.

• The creation which occurs thanks to the powers of all the internal and external senses is the waking state. Here too, as in the dream state, illusion may occur, in the form of seeing two moons etc.29 - 17 -

18. This triad is to be abandoned, since, as the prāṇa etc. predominate and, consequently, [authentic, free] agency becomes subordinated in it, there is union with pleasure and pain, essentially consisting in the attenuation or intensification of this.

• These waking, sleeping and dream states are to be abandoned since in them the prāṇa etc. - presumed to be the self - are pre-eminent and

26 As the memory, for example, of having slept 'well' shows.

27 It is not the limited subject who is responsible for these images, since some are unwelcome and others, that would be welcome, are linked with other times and places (see ĪPV II p. 226).

28 In line with the conception of illusion as apūṛṇakhyāti the illusion of the dream does not lie in the perception itself but in its lack of permanence.

29 According to Abh. (ĪPVV III p. 340, ata eva āgameṣu avasthānām api anyonyānupraveśo bahuśākha uktaḥ), this is a hint at the conception according to which there exist within every single state various planes and aspects corresponding to its merging with other states. In the waking state as perception the dream state as illusion is present etc. (cf. ŚSV pp. 11-12, MV II; in particular see TĀ X.227bc ff. and TS IX where both IPK III. 2. 16-17 and MV II are commented on).

freedom is reduced. A slight (mātrayā) increase in freedom means III 2.18

pleasure, joy etc., and thus a decrease in freedom is pain, aversion and

so on. The saṃsāra constituted by all the fruitions is the bond. - 18 -

19. The vital breath, constituted by rising and falling breath, is

present in everyone in the dreaming and waking state. In deep sleep the

breath called samāna is present, consisting in the interruption of those,

in a way similar to what occurs during the equinoxes.

• And that self identified with the vital breath, in the form of in-

halation and exhalation, or rising and falling breath (prāṇāpāna°), is

present in the waking and sleeping states30. When these two breaths

become equal (sāmye)31, like day and night at the equinox (viṣuvatī-

30 The prāṇa in the broad sense of the word (breath, vital energy, life) is a func-

tion, a mode of consciousness itself (cidvṛtti; prāk samvit prāṇe parinatā, said the

Tattvārthacintāmani, cit in ŚSV p. 68 etc.), capable of transmitting life to realities

such as the body, which are in themselves inert and insentient (jada). This undif-

ferentiated vibration (sāmānyaparispanda) is manifested in five fundamental forms

which include the whole range of states of consciousness. Rising and falling breath

(prāṇa in the narrow sense of the word and apāna) characterize the state of ordi-

nary consciousness (waking and sleeping). They incessantly travel from the heart to

the dvādaśānta in opposite directions: the prāṇa, flowing towards the exterior (is

'day', 'sun'), represents the tension towards the object, the apāna ('night', 'moon')

represents the tension towards the subject (cf. ĪPVV III pp. 342-343). The dualistic

opposition between prāṇa and apāna is the same as that which pervades ordinary

consciousness. Now one now the other predominates, just as in the days of the year

night prevails over day and vice versa. This perpetual play of opposition is

suspended in the state of deep sleep: in the savedyasausupta when in the interval

between prāṇa and apāna there is a momentary rest in the place of the heart, in the

apavedyasausupta throughout the duration of deep states of obnubilation, torpor etc.

If this suspension of the prāṇa-apāna flow is immediately plausible for the

apavedyasausupta where by definition the pole of objective reality disappears, it is

legitimate, however, to ask oneself - as does Abh. (ĪPVV III p. 346), following the

tīkā - how this happens for the savedyasausupta, too. The flowing of the prāṇa -

is the reply - does not depend on the mere presence of the cognizable (vedyamātra)

but on its appearing as separate from the subject, and the cognizable that is present

in this state (sensations of pleasure etc.) is such that it appears, so to speak, to be

one with the subject (grāhakād avicchinnam iva).

31 In the kārikā cheda therefore has the double meaning of 'interruption' (vic-

cheda) - of the breaths as opposites - and 'equality'. At this point - which corresponds

to the plane of the Pralayākalas - the breath is precisely referred to as 'equal'

(samāna). It does not, however, represent the definitive eclipse of prāṇa and apāna

• which would entail going beyond phenomenal existence - but only a momentary

207

III 2.19 va)32, due to their increasing and decreasing being blocked, the breath is called samāna ("equal"). - 19 -

20. Flowing upwards through the middle path, the breath is called udāna; it is in the fourth state and is made of fire, it corresponds to the Vijñānākalas, the Mantras and the Lord33. The supreme breath is the vyāna, whose essence is all.

stasis in which they remain in the state of saṃskāra, ready to manifest themselves once more. They still have to be truly surpassed.

32 In his commentaries Abh. expounds two distinct traditional etymologies of visuval: ‘capable (vat) of pervading (visu)’ and ‘that sets in motion (suvat) the distinction (vi=viśeṣa) or the imbalance of duration between day and night’. Bh. finds this second interpretation obviously contradictory, seeing that he insists against all likelihood on glossing prerayati (with which Abh. explains suvalti) with nivārayati. The difficulty may perhaps be overcome by understanding that visuval sets in motion the imbalance only insofar as, there being a momentary balance, the imbalance must necessarily follow. The passage from the Kālapāda (another name for the Kālotlara) cit. in IPVV III p. 343 may be understood in this sense: sāmyam vaiṣamyajanakaṃ visuval santatam smrtam (this ardhaśloka, not included in the Sārdhatriśatikālottara, probably comes from another of its many versions). The image of the visuval is not merely incidental but is taken from the Śaiva scriptures, where the systematic projection of external time, with all its subdivisions (including the visuval) in the individual dimension, is frequently found, with the consequent identification of cosmic time with ‘prāṇic’ time; in particular see Kālottarāgama (sārdhatriśati) XI. 1-10 and X. 22-23 with the relative vrtti, SvT VII. 168 ff.; for another use of visuval see for example SvT II. 316-334a; cf. SSP III pp. 358 ff., Śaivāgamaparibhāṣamañjarī (transl.) p. 276. Similar conceptions are to be found in the Buddhist Kālacakra school: Vimalaprabhā, passim, Nadapāda's Sekoddeśatikā pp. 42-44; on visuval – but understood as taking place lagnodayābhisamdhau – see Vimalaprabhā p. 178.

33 It appears from the fragmentary indications in the IPVV that the tīkā suggests two interpretations of the compound vijñānākalamantreśo – a karmadhāraya, as is explicitly said, and not a dvandva (which would have required the plural). The most obvious one is ‘the Vijñānākalas and the Lords of the Mantras’; the other, adopted by Abh., is ‘the Vijñānākalas, the Mantras in the broad sense of the word (i.e. Mantras, Lords of the Mantras and Great Lords of the Mantras) and the Lord (Abh.: the Lord Sadāśiva)’. From the tīkā quotations one would obtain, more precisely, ‘Vijñānākalas, the Mantras, the Lords of the Mantras and Īśvara’, where Mantreśvara (Lord of the Mantras) derives, according to the tantra method, from mantra and īśa, to which, however, an individual meaning is also assigned (cf. IPVV III p. 352, nanu atra vyākhyāne mantreśvarāḥ kena samgṛhītāḥ / mantrapadena tajjñīyavād īśapadena samānavācyatākrtena tantranyāyeneti).

208

• [The breath is referred to as] udāna because - after accomplishing III 2.20 the unification of the prāṇa and the apāna through the interruption of their flow along the two sides and transversely - it rises (udgamanāt) flowing upwards through the middle path34. This breath - which flows in the fourth state -35, similar to the middle eye, has the quality of fire because it rises upwards. The breath ceases to flow in the state beyond the fourth - this being the state of fullness and as such devoid of spatio-temporal delimitations -36 and entering a state of absolute fullness takes the name of vyāna. These two states are to be attained since in them the energy of the prāṇa appears as essentially constituted by the supreme Lord37. - 20 -

34 The middle path is the susumnā; the two side paths are idā and pingalā.

35 The breath udāna rises, therefore, along the suṣumnā burning every duality (ĪPVV III pp. 350-1 cites some of them: creation and reabsorption, prāṇa and apāna, day and night, birth and death, aversion and attachment). The fourth state - together with the state beyond the fourth - constitutes going beyond ordinary existence. In this state differentiation dissolves like melting butter, the idantā is veiled by the I as on the Sadāśiva and Īśvara levels. The objective realities with which the I had identified himself are themselves transformed on contact with the I which is finally recognized as endowed with the divine qualities of sovereignty, permanence etc., so that they continue to subsist, but as though they have ceased to be objects; they are compared to copper which on contact with mercury is transformed into gold. In the state beyond the fourth state the difference is only one of intensity; the quality leap has already been achieved. The state beyond the fourth state, in which only the 'pervading' breath (vyāna) subsists, is the stabilization of the fourth, the differentiation is now completely dissolved, the idantā dispelled; remaining within the terms of the simile suggested by Abh., even the gold into which the various levels of subjectivity have been transformed - from the body to the void - as it is increasingly permeated by the mercury wears away and finally dissolves, these realities only surviving in the residual form of samskāra (cf. ĪPVV III pp. 327-8, 331, 350; ĪPV II pp. 257-8, 270-1, 275-6).

36 In fact the flowing presupposes spatial and temporal change, whereas the transfourth is the state of absolute simultaneous fullness.

37 Literally 'similar to the supreme Lord'. But my translation is in conformity with Abh.'s gloss (see 'cxt note 108), which in turn appears to follow Utp.'s indications.

IV 1

SECTION IV. SUMMARY OF THE ESSENTIAL PRINCIPLES

1. The very Self of all creatures is Maheśvara, one, whose form is all, full of the undivided consciousness 'I-this'.

– The one, full of the 'savouring' (°camatkāra°) of the undivided perceiving subject, of the undivided perceptible object and of the fusion of the two2, in the fourth state which is to be sought first3, the Self common to all living beings, whose form is all, is Maheśvara. - 1 -

2. There (tatra), realities such as the mind etc., which are included in the sphere of the 'this' freely created [by the Lord] (svasṛ̣tedambhāge), are caused by him to be considered as the fictitious I (ahamkāra°) as being the perceiving subject.

– This universe springing forth as constituted by the free expansion (jṛmbhā°) of the supreme Lord4, realities that are created as the object

1 The translation refers to the most immediate and general meaning of the text of the kārikā. The far more complex explanation given by the vṛtti depends on considering the two terms of the expression aham idam first separately and then together.

2 The above note has already mentioned how this interpretation has been arrived at. As Abh. explains (see text note 2), we have here the reflective awareness of the perceiving subject as not distinct from the other subjects, i.e. of the fundamental unity of the subject beneath the apparent fragmentation; the same applies to the object; and finally the awareness of the ultimate non-otherness (melanā) of subject and object.

3 Because starting from this the maculation begins to dissolve (see text note 1).

4 The text, which remains doubtful, seems to contain a double interpretation (tantra) of tatra in the kārikā (which I have translated in a deliberately vague manner): maheśvarasya jṛmbhāmaye 'smin nirgate and tasminn idantāparāmarśe. This would coincide with the brief references in IPVV III p. 358 to the ṭīkā, which evidently gave this type of interpretation: tatra iti vaiyadhikaranyena vivṛtya sāmā-nādhikaranyena vivṛnoti (i.e. tatra taken alone or agreeing with °bhāge). Abh. in his two commentaries also follows the line of the simultaneous double function of tatra, which he refers in both cases to Maheśvara: 'Maheśvara being the self, the part 'this' is created by him upon a substratum constituted by himself' (thus in the first case

210

of perception - the mind, the breath, the void - in the sphere of the IV 2

reflective awareness 'this', precisely these realities, which represent a

part of the knowable, are caused to be considered as the fictitious I and

are, on the strength of this, transformed into the limited perceiving sub-

ject. - 2 -

3. The individual subjects considered as being many are the fruit of

the non-recognition of their true nature. In them action and beatitude are

created, that is, the fruition characterized by pleasure and pain5.

• It is precisely the rendering limited of that whose essence is all

which is called 'non-recognition'. And thus the awareness of a limited

subjectivity attributed to multiple and fragmented realities, such as the

mind, the breath and so on6, is called absence of the perfect knowledge

of the supreme subjectivity. The individual selves are many; in them -

who are manifested as knowing subjects - Maheśvara creates his own

beatitude and activity, which are themselves inherent in the only real

agency. This is precisely the fruition which is called pleasure, since it

consists in a particle of [divine] beatitude, and pain, since it is an in-

finitesimal fragment of [divine] activity. The fact that activity is pain will

be explained later. - 3 -

4. In the limited subject sattva, rajas and tamas correspond to

knowledge, activity and, thirdly, māyā7, in the Lord - for whom things

represent his own body (svāñgarūpeṣu).

tatra is a locative absolute with sati understood, in the second, it is a locative depending on svasrṣṭa°).

5 The capacity for fruition that the Lord instils in the various limited souls is therefore a reflection created by him of the powers of beatitude and activity, which are, on the contrary, innate in him. They are therefore kalpitā. Beatitude and activity are not a different thing from jñāna° and kriyāśakti, prakāśa and vimarśa (ĪPVV III p. 359).

6 Abh. explains this point (ibid. p. 360) by using the example of a long serpent (gonāsa), whose body is pervaded by a single life (jīvatva), capable, however, of being divided into many distinct partial vitalities which animate the individual parts of the same serpent when it is cut into pieces; and also capable of becoming one again.

7 The power of māyā is that which on the Sadāśiva-Īśvara plane determines the appearance of idantā within the consciousness (ĪPVV III p. 361). It is a further expansion of kriyāśakti; indeed, in Abh.'s words (ibid. p. 366), it is the kriyāśakti itself in which, however, instead of samharana predominating (srṣṭi predominates in jñānaśakti) śūnyatā or nisedha predominates, i.e. the mutual delimitation of subject

IV 4 - The knowledgc and activity of the Lord, together with māyā, become sattva, rajas and tamas in the limited subject. - 4 -

5. Owing to the state of differentiation, these qualities (gunānām), which are transformed in the faculties and in the products, cannot be considered as powers inherent in a subject who is the possessor of them8.

• Since sattva, rajas and tamas are manifested as distinct [from the subject], the term powers may not be applied to them, in that there is no differentiation between the power and the possessor of the power. In actual fact everything is an expansion of the power(s)9. - 5 -

6. In the Lord there is being, beatitude, activity. In the limited subject there is that, the absence of that and the two together10. The latter constitute rajas, which is pain and is composed of sattva and tamas inter-mixed.

• In the Lord, the infinite agency, whose essence is 'savouring' (ca-matkāra°), is called activity and consists of supreme light and beatitude11.

and object. This power does belong to Śiva but, as the formulation of the kārikā points out, it is not on the same plane as the powers of knowledge and action, since it is connected with differentiation (cf. ibid. p. 361).

8 This kārikā is intended to be a reply to those who want to take the parallel between knowledge, activity, and māyā, on the one hand, and sattva, rajas, and tamas on the other, too far, by considering the latter 'powers' of the limited subject just as the former are powers of Śiva. Utp. replies that this cannot be so, because the limited subject looks on them - and on all the deployment of phenomena that evolves from them and is imbued with them - as other than himself, whereas the power is by definition not separate from the subject who possesses it. In the limited subject knowledge (prakāśa) etc. are not spontancous (naisargikya) and autonomous functions, but require the contribution of other factors - e.g. the senses - in order to be performed, which are all penetrated by and derived from the gunas. This is why they are called gunas, i.e. 'qualities, components', but also 'subordinate, auxiliary elements' (cf. ĪPVV III p. 362).

9 What has been said so far is only valid, however, within the limits of the world of differentiation. Once the subject recognizes his own true nature, the objective world ceases to appear in the guise of otherness and it is then that all the principles become what they effectively are, his 'powers'.

10 What is referred to here, as the vrtti makes clear, is the guna rajas, constituted by the presence of sattā as prakāśa-ānanda (the guna sattva) and by its (their) absence (the guna tamas).

11 These qualities of the Lord are involved with one another and in the final analysis are identified with one another (cf. ĪPV II p. 289 yā sattā bhavanakartṛtā

On the contrary, in the individual self this light and beatitude, which have become knowable realities, constitute sattva. Their absence is tamas. Sattva and tamas - or light-beatitude and their absence - though two distinct entities, are mutually united to form rajas. In the limited subject activity and pain are called rajas. - 6 -

7. Those realities which in the Lord are the object of an unconventional reflective awareness in terms of 'this', those, mixed and separated, are thus manifested in various ways12.

• The things that constitute this universe (ye caite), which are to Maheśvara - as in the case of the child - indicated by the meaning of the word 'this' (reflective awareness being the very essence of the light), without the connection with [or: the consideration of] the conventional linguistic expression 'this' coming into play13, those very things appear - since they were created thus by God - as particular realities (svalakṣaṇa°) thanks to several manifestations having the same substratum (°sāmānādhikaranyena) and in many forms insofar as they are universals each taken singly14. - 7 -

8. On the contrary, in the individual subject15, those entities that are manifested separately become the object of mental elaboration in the

sphurattārūpā ... saiva ... camatkārarūpā satī kriyāśaktir ucyate ... svātmaviśrāntirūpatvāc ca saiva ānandah). In the paśu the guṇa sattva corresponds to the sattā and ānanda aspect in the Lord.

12 'On the plane of the limited subject' is understood.

13 Some form of parāmarśa is inseparable from the pure light of cognition, as reflective awareness reactive to it, an act of appropriation on the part of the I. On this level, on which the 'this' is conceived purely and simply as a negation of the I (ĪPVV p. 367), things are the object of parāmarśa only in terms of 'this', without any other differentiations. And furthermore it is specified that this parāmarśa is also free from any contact with the conventional linguistic expression 'this', since the word - which is, as we know, inseparable from it - is still at a level transcending every articulated language, similar rather to a silent nod of the head. An example that clarifies this is the dim and undifferentiated awareness of (and reaction to) what is other than itself and that surrounds it, that the newborn child is presumed to have.

14 See the following kārikā.

15 Kārikā 7 has defined the status of the object in relation to the Lord. But Abh. (ĪPVV III p. 367) makes it clear that it is not the supreme Lord in this case, since at the level of Paramaśiva one cannot speak of any objectivity whatsoever, this only beginning to emerge when he assumes the nature of Sadāśiva and Īśvara (indeed the

IV 8 sphere of memory, imagination and so on, and have a variety of distinct names.

– The entities that are manifested separately, that is, the universals16 are shown by the limited subjects – thanks to the power of mental elaboration (vikalpanaśaktyā)17 – as the object of inner reflective awareness through various names such as 'jar', 'silvery', 'white', 'cloth', 'cart' and so on. The limited subjects, conditioned by the experience of these objects, then also make themselves the object of various denotations such as 'I am thin', 'I am happy or unhappy'. In memory this function of mental elaboration takes place in the wake of the previous direct experience; in the various forms of imagination etc. it is, on the contrary, autonomous18. It is precisely this manifestation of a differentiation between perceiving subject and object perceived, substantiated by the word19, which constitutes the bond of the saṃsāra in the limited soul. – 8 –

9-10. The creation that pertains to him is not also common [to other subjects] and is dependent on the creation of the Lord. It occurs in the limited subject – essentially identical to the Lord – in virtue of the very

plane of the Lord (pati°, īśvaradaśā) corresponds to these). Kārikā 8 expounds how the object appears on the plane of limited subjectivity.

16 Of the two kinds of object, particular (svalaksana) and universal (sāmānya), only the latter (cf. text note 16) is taken into consideration here, because the word, which in this kārikā is shown as that which accompanies and renders possible discursive thought, has as its content precisely the sāmānya, i.e. the ābhāsa taken separately (see above II. 3. 2).

17 Vikalpana is, as has been seen, that characteristic function of discursive thought that proceeds to define the object through the evocation and negation of what is other than it. In this the essential role is played by the word, as the Buddhists show by the conception of the apoha.

18 See above I. 8. 1 ff.

19 Many śaiva texts insist on the fundamental ambivalence of the word, which permeates every level of human activity. The powers of the word, often conceived as the powers that preside over the various alphabetical classes, may lead man to freedom or enmesh him even more in the saṃsāra. To remain within the terms of the discourse of the vṛtti, they may also, if correctly polarized, gradually re-absorb the vikalpa in pure reflective awareness – pure insofar as it is not dualizing (the two possible ways of overcoming the vikalpa will be dealt with at some length later). IPVV III pp. 379-380 refers to some of the classical passages of these speculations with quotations from the SK, the Timirodghātā (the explicit attribution to this text is contained in ŚSV p. 7) and the MV, to which the ŚS, at least, can be added.

214

power of the Lord, ignored, however, as such20, which aims at preventing the resting on the self, unstable, whose nature is vital energy, diversified in the multiplicity of the various phonemes, coinciding with the activity of mental elaboration (vikalpakriyā).

– The individual subject - whose nature is identical to that of the Lord - creates, thanks to the power of the Lord not recognized as such, objects in the form of the various universals21, not perceptible on the part of the other subjects and modelled on the direct experience of objects that may be experienced by everyone. And this power of the Lord, aimed at obstructing the resting on authentic subjectivity, takes the name of māyā; it contains the infinite number of all the various phonemes - K etc. -, it is continuously in motion in the form of vital energy, it carries out the function called 'mental elaboration'. In this way it creates the objects internally. - 9-10 -

11. The creation of the Lord may be common or not common to all subjects (sādhāraṇo 'nyathā ca), manifesting itself [in both cases] in all clarity. With the suppression of the mental constructs, resulting from concentration on a single point, the plane of the Lord is gradually reached22.

20 As we saw earlier, also the limited subject possesses an innate power of creation which is in essence no different from that of the Lord, since the individual himself is essentially identical to the Lord although he ignores or has forgotten this identity. This power of creation, limited by non-awareness, remains restricted to the individual sphere: what is created are images, feelings, ideas etc., which depend more (as in the case of memory) or less (as in fantasy) on the objects in the phenomenal world created by the Lord. These two kinds of objects and of creations are distinguished by being the last 'common to all subjects', the others are 'not common' i.e. not able to be experienced except by the limited subject who creates them. But the power of creation is essentially the same.

21 The object of the mental elaboration (vikalpa etc.) is not the svalakṣaṇa, but the sāmānya, which is not attached to a definite time and place.

22 Or: '... becomes the object of the reflective awareness ['I am this universe'] peculiar to the condition of Īśvara'. The creation of the Lord is therefore twofold. On the one hand it constitutes the world of objective existence. regarding which all subjects are coordinated: on the other - restricted to a single individual - it constitutes the sphere of dream and error, such as the mistaken sight of two moons, which as far as clarity and immediacy are concerned is on the same plane as the other (cf. III. 2. 16); cf. ĪPVV III p. 386 aīśvarasya sargasyeva spaṣṭavasatībhāsatvam, ata eva avikalpyateti rūpam. This interpretation of the first part of the kārikā and in particular of the expression sādhāraṇo 'nyathā ca is that suggested by the vrtti, but not the only possible one. For instance, one may understand, as appears from the

IV 11 - On the contrary, the creation of the Lord is common to all cognizing subjects - since they are born within it23 - or also limited to only one subject like that which is manifested in dream, in error and so on, where it concerns a single subject. Permeated only by the reflective awareness of 'I am this universe', this creation of the Lord is free from mental constructs (vikalpojjhitā) - since no differentiation arises within it24 - and it is manifested in all clarity (spaṣṭāvabhāsā ca)25. On this plane, by applying oneself and intensely cultivating those moments when the mental construct becomes attenuated, which occur sporadically (antarāntarā) while the activity26 pertaining to the limited subject is taking place, the beings in the power of the saṃsāra gradually attain, through

tīkā, 'the creation of the Lord, common or not common. is different (from the creation of the limited subject)', in which asādhārana ('not common') is implied by the conjunction ca. To this Abh. adds an interpretation of his own (ĪPVV III p. 386 and ĪPV II pp. 301-2): if the creation pertaining to the limited subject, or the creation of the vikalpas, assumes full vividness being experienced differently - i.e. recognizing the power of the Lord in it -, then it too becomes common, i.e. able to be experienced by everyone, as in the case of he who, being imbued with the energy of the mantras, becomes able to transform his representations and intentions into reality; in this case the pāśavah sargah, too, can lead to the condition of the Lord (īśvaratām asya dadāti). As he himself explains, the aiśah sargah of the kārikā may be understood as pāśavah sargah, insofar as also this latter form of creation rests, in its ultimate essence, on the former; if this interpretation is not directly referred to in the vrtti and tīkā - Abh. adds - it is only because the ultimate identity of the pāśava and aiśvara creation is too obvious to be stated overtly.

23 In the sense that the creation of the object is parallel to that of the subject. Reality, in effect, is constituted by the interaction of determinate levels of subjectivity with the corresponding levels of objectivity (see e.g. PH pp. 6-7).

24 The absence of the split between subject and objective reality constitutes the cardinal difference between Īśvara's creation and the paśu's creation (ĪPVV. III p. 388 'bhedānudayāt' iti / anena aiśvarasya sargasya pāśavāt sargāt vailakṣaṇyaṃ darśitam; see also ĪP II p. 329); and it is preciscly for this reason that the creation of the Lord does not represent a bond. However, it must be considered (ibid.), that in the paśu's creation two levels may be distinguished, and whereas the first only shows the emergence of differentiation, with the māyic and āṇava maculation, it is the second, presided over by Brahmā (cf. III. 2. 1) and consisting in the various psychic and mental activities (pratyaya), that directly causes the karmic maculation and binds to the saṃsāra.

25 Dharmakīrti defines the type of knowledge that pertains to the yogins in similar terms: PV III. 281 prāguktam yoginām jñānam teṣām tadbhāvanāmayam / vidhūtakalpanājālam spaṣṭam evāvabhāsate; see also PVin I. 28 and the following passage in prose (Steinkellner 1972: 203):

26 I.e. the vikalpa.

the emergence of the state of the Lord in all its fullness, the dissolution IV 11 of the state of limited individuality. - 11 -

12. He who, having all as his essence, thus knows: «All this multiform deployment is mine», he, even in the flow of mental constructs, attains the state of Maheśa27.

• Also the individual subject who - though continuing to produce mental constructs deriving precisely from the power of the Lord - has become fully aware of his own nature as Lord - 'Mine is this multiform deployment of the saṃsāra' - this subject, too, who permeates the universe without any differentiation and transforms all the mental constructs into pure reflective awareness, attains the state of Maheśvara. - 12 -

13. The liberated soul looks at the 'common' cognizable reality as being undifferentiated from himself, like Maheśvara; the bound soul, on the contrary, looks at it as absolutely differentiated.

• Cognizable reality28 is the same for the bound and the liberated soul; however, the bound soul knows it as absolutely differentiated and the liberated soul as the body of his very self. - 13 -

14. But when he is completely full of the infinite series of principles dissolved in him, he is Śiva, solely consciousness and bliss, having as his body the supreme syllable29.

• But when the cognizable is entirely dissolved within him and there is the full consciousness of the I, the state of Śiva is attained. - 14 -

27 This is an alternative path to that expounded in the previous kārikā; both are directed to the same aim which is to free from the bond of the vikalpas. In the first the vikalpas were gradually suppressed by fixing the nirvikalpa (cf. PH pp. 39-40; ĪPVV III p. 386 avikalpakaparigrhītaspastavastuvisayaikāgrātāvalambanena); here, on the contrary, their inclusion - and transfiguration - in the sphere of the totality of consciousness is taught.

28 I.e. the so-called objective reality (cf. IV.11) facing both the liberated soul and the bound soul.

29 Unlike the situation expounded in the previous kārikā, where the cognizable subsisted though in a state of non-differentiation with the I - a state that Abh. compares to that of the plane of Sadāśiva-Īśvara -, here one cannot even speak of cognizable reality, as it has now been dissolved and absorbed in the I without a trace. This is the state of the supreme Śiva.

217

IV 15 15. Thus seeing the true reality of what is his own self and the knowledge and action that pertain to him, he knows and creates objects as he wishes.

• Having thus recognized as Lord his own self full of the infinite powers of knowledge and action once the signs of recognition represented by the powers have come to light, he sees and creates all things at will. - 15 -

16. Thus this new, easy path has been explained by me as the great master expounded it in the Sivadrsti. Thus he who, putting his feet on it, brings to light in the self the nature of creator of the universe whose essence is the nature of Siva, and is uninterruptedly absorbed in it, attains perfection.

• Within the course of ordinary reality, in its various forms, solely in virtue of an act of recognition (pratyabhijñātrāt) is the state of Siva attained. This new path, devoid of tortuosity, was indicated in the treatise that bears the name of Sivadrsti by the venerable Somānanda, who had direct experience of the form of the blessed Parameśvara; I have here furnished a logical justification of this path, thus causing it to enter the hearts of men. He who by applying himself intensely to this enters into the nature of Siva, becomes in this very life a liberated soul. - 16 -

17. Just like the beloved, who after much insistence finally stands in the presence of the maid in love, though he is there before her he does not give her any pleasure until she recognizes who he is - as he seems just like other men until that moment -, so for mankind the self, who is yet the Lord of the world, cannot manifest his own glory until his qualities have been brought to light. For this reason the doctrine of the recognition of the Lord has here been expounded.

• The example of the beloved shows that the recognition of the Lord must be awakened30 and that its fruit is extraordinary. - 17 -

30 Utp. uses the causative form of the root upad-, as he has done before, in his oscillation between the meanings of 'awaken', 'render possible', and 'justify logically' (cf. also the following kārikā).

18. In order that the ordinary man too can attain perfection (°siddhi°) IV 18 effortlessly, Utpala, son of Udayākara, has explained with argumentations (upapāditā) this recognition of the Lord.

• This easy path to perfection has thus been shown. - 18 -

Here ends the Īśvarapratyabhijñāvrtti.

ABBREVIATIONS

Abh. Abhinavagupta

AK Abhidharmakośa

AKBh Abhidharmakośabhāsya

APS Ajadapramātrsiddhi

Bh Bhāskarī

HB Hetubindu

ĪPK Īśvarapratyabhijñākārikā

ĪPV Īśvarapratyabhijñāvimarśinī

ĪPVV Īśvarapratyabhijñāvivrtivimarśinī

ĪS Īśvarasiddhi

KSTS Kashmir Series of Texts and Studies

MMP Mahārthamañjarīparimala

MPĀ Mataṅgapārameśvarāgama

MṛA Mrgendrāgama

MṛV Mrgendravṛtti

MV Mālinīvijaya

MVV Mālinīvijayavarttika

NB Nyāyabindu

NBṬ Nyāyabinduṭīkā

NM Nyāyamañjarī

PH Pratyabhijñāhrdaya

PS Pramāṇasamuccaya

PTLV Parātrimśikalaghuvaṛṭṭi

PTV Parātrimśikāvivaraṇa

PV Pramāṇavārttika (PV I = Svārthānumāna, PV II = Pramāṇasiddhi, PV III = Pratyakṣa. PV IV = Parārthānumāna)

PVSV Pramāṇavārttikasvavrtti

PVin Pramāṇaviniścaya

ŚD Śivadṛṣṭi

ŚDVṛ Śivadṛṣṭtivivrtti

ŚS Śivasūtra

ŚSV Śivasūtravimarśinī

ŚSĀ Śivastotrāvalī

ŚV Ślokavārttika

SK Spandakārikā

SN Spandanirnaya

SS Sambandhasiddhi

SSVṛ Sambandhasiddhivrtti

SP Sambandhaparikṣā

SPr Samvitprakāśa

SDS Sarvadarśanasamgraha

221

SŚP Somaśambhupaddhati

SvT Svacchandatantra

TĀ Tantrāloka

TĀV Tantrālokaviveka

TBh Tarkabhāṣā

TP Tattvaprakāśa

TS Tattvasaṃgraha

TSP Tattvasaṃgrahapañjikā

TSā Tāṇtrasāra

Utp. Utpaladeva

VBh Vijñānabhairava

VP Vākyapadīya (the kārikās are quoted according to Rau's cdition)

222

BIBLIOGRAPHY

TEXTS

Abhinavagupta, Īśvarapratyabhijñāvimarśinī (see Bhāskarakanṭha, Bhāskarī).

Abhinavagupta, Īśvarapratyabhijñāvivrtivimarśinī, edited by Madhusudan Kaul Shastri, vols. I-III, KSTS LXII LXII LXV, Bombay 1938-43.

Abhinavagupta, Tantrāloka with Commentary by Rājānaka Jayaratha, edited with notes by Madhusudan Kaul Shastri, vols. I-XII, KSTS XXIII, XXVIII, XXX, XXXVI, XXXV, XXIX, XLI, XLVII, LIX, LII, LVII, LVIII, Allahabad-Srinagar-Bombay 1918-1938.

Abhinavagupta, Mālinīvijayavārttikam, edited with notes by Madhusudan Kaul Shastri, KSTS XXXI, Srinagar 1921.

Abhinavagupta, Parātrīṃśikāvivaraṇa (see Gnoli 1985).

Abhinavagupta, Tantrasāra, edited by Mukunda Ram Shastri, KSTS, XVII, 1918.

Abhinavagupta, Tantroccaya , edited by R. Gnoli and R. Torcella, in Indo-Sino-Tibetica. Studi in onore di Luciano Petech, a cura di P. Daffinà, Università di Roma 'La Sapienza', Studi Orientali IX, Roma 1990.

(Bhaṭṭa) Arcaṭa, Hetubinduṭīkā with the Sub-Commentary entitled Āloka of Durveka Miśra, edited by Pandit Sukhlalji Sanghavi and Muni Shri Jinavijayaji, Baroda Oriental Institute, 1949.

Bhartrhari, Mahābhāsyadīpikā, Āhnika I (see Bronkhorst 1987).

Bhartrhari, Vākyapadīya with the Commentaries Vrtti and Paddhati of Vṛṣabhadeva, Kāṇḍa I, edited by K.A.S. Iyer, Deccan College, Poona 1966.

Bhartrhari, Vākyapadīya, containing the Ṭīkā of Punyarāja and the Ancient Vrtti, Kāṇḍa II, edited by K.A.S. Iyer, with a Foreword by A. Aklujkar, Delhi 1983.

Bhartrhari, Vākyapadīya with the Commentary of Helārāja, Kāṇḍa III part I, edited by K.A.S. Iyer, Deccan College, Poona 1963.

Bhartrhari, Vākyapadīya with the Prakīrṇakaprakāśa of Helārāja, Kāṇḍa III part II, edited by K.A.S. Iyer, Deccan College, Poona 1973.

Bhartrhari, Vākyapadīya (mūlakārikās), Bhartrharis Vākyapadīya, edited by W. Rau, Abhandlungen für die Kunde des Morgenlandes, 42, Wiesbaden 1977.

Bhāsarvajña, Nyāyabhūṣaṇam, edited by Svāmī Yogīndrānanda, Ṣaddarśanaprakāśa-napratiṣṭhāna-granthamālā No.1, Varanasi 1968.

Bhāskarakanṭha, Bhāskarī. A Commentary on the Īśvarapratyabhijñāvimarśinī of Abhinavagupta, vols. I-II, edited by K.A.S. Iyer and K.C. Pandey, The Princess of Wales Sarasvati Bhavana Texts Nos. 70 and 83, Allahabad 1938-1950.

Dharmakīrti, Vādanyāyaprakaraṇa with the Commentary Vipañcitārthā of Āchārya Śāntarakṣita and Sambandhaparīkṣā with the Commentary of Āchārya Prabhācandra, Dharmakīrtinibandhāvalīḥ (2), Bauddhabhāratīgranthamālā No. 8, Varanasi 1972.

223

Dharmakīrti, Pramāṇavārttika-kārikā (Sanskrit and Tibetan), edited by Y. Miyasaka, Acta Indologica, 2, 1971-72.

Dharmakīrti, Pramāṇavārttika with the Commentary ‘Vṛtti’ of Acharya Manorathanandin, critically edited by Swami Dwarikadas Shastri, Varanasi 1968.

Dharmakīrti, Pramāṇavārttikam. The First Chapter with the Autocommentary, text and critical notes by R.Gnoli, Serie Orientale Roma XXIII, IsMEO, Roma 1960.

Dharmakīrti, Hetubindu (see Steinkellner 1967).

Dharmottara, Nyāyabindutīkā (see Durveka Miśra, Dharmottarapradīpa).

Dignāga, Nyāyamukha (see Tucci 1930).

Dignāga, Pramāṇasamuccaya (see Hattori 1968).

Durveka Miśra, Dharmottarapradīpa [being a sub-commentary on Dharmottara's Nyāyabindutīkā], edited by Paṇḍita Dalsukhbhai Malvania, Kashiparasad Jayaswal Research Institute, Revised II Ed., Patna 1971.

Helarāja, Prakīrṇakaprakāśa (see Bharṭhari, Vākyapadīya. Kāṇḍa III parts I-II ).

Jayanta, Nyāyamañjarī, edited by Pt. Surya Nārāyaṇa Śukla, parts I-II, Kashi Sanskrit Series 106, Varanasi 1969-71.

Jñānaśrī, Apohaprakaraṇa, in Jñānaśrimitranibandhāvalī, edited by Anantalal Thakur, Tibetan Sanskrit Work Series 5, Patna 1959.

Kaunda Bhaṭṭa, Vyākaraṇabhūṣaṇasāra with the Darpana Commentary by Śrīharivalabha, the Parīkṣā Commentary by Bhairava Miśra [...], edited by Pt. Śrī Sadāśiva Śāstri Joshi, Haridās Sanskrit Granthamālā 133, Benares 1939.

Kiraṇāgama (Vidyāpāda) (see Vivanti 1985)

Kṣemarāja, Pratyabhijñāhrdayam, edited by Mukunda Ram Shastri, KSTS III, Srinagar 1918.

Kṣemarāja, Spandanirnaya, edited by Madhusudan Kaul Shastri, KSTS XLII, Srinagar 1925.

Kṣemarāja, Śivasūtravimarśinī, edited by J.C.Chatterji, KSTS I, Srinagar 1911.

Kṣemarāja, Spandasamdōha, edited by Mukunda Ram Shastri, KSTS XVI, Bombay 1917.

Kṣemarāja, Sāmbapañcāśikāṭīkā. In: Tantrasamgraha [prathamō bhāgah] edited by Gopiṇāth Kāvyālj, Yogatantra-Granthamālā 3, Varanasi 1970.

Kumāriḷa, Ślokavārttikam with the commentary Nyāyaratnākara of Śrī Pārthasāratimiśra, edited and revised by Swami Dwarikadas Shastri, Prācyabhārati Series-10, Varanasi 1978.

Mādhavācārya, Sarvadarśanasamgraha, edited by Uma Shankar Sharma, Vidyabhavan Sanskrit Granthamala 113, Varanasi 1978.

Madhyāntavibhāgaśāstra, containing the Kārikās of Maitreya, Bhāṣya of Vasubandhu and Ṭīkā by Sthiramati, critically edited by R.C. Pandey, Delhi 1972.

Maheśvarānanda, Mahārthamañjarī with the Autocommentary Parimala, edited by Pt. V.V. Dvivedi, Yogatantra-Ratnamālā 5, Varanasi 1972.

Mālinīvijayottaratantraṃ, edited by Pt. Madhusūdan Kaul Shastri, KSTS No.37, Bombay 1922.

Mataṅgapārameśvarāgama (Vidyāpāda), avec le commentaire de Bhaṭṭa Rāmakaṇṭha, édition critique par N.R. Bhatt, Publications de l'Institut Français d'Indologie No. 56, Pondichéry 1977.

Matangapārameśvarāgama (Kriyāpāda, Yogapāda et Caryāpāda), avec le commentaire de Bhatta Rāmakantḥa, édition critique par N.R. Bhatt, Publications de l'Institut Français d'Indologie No. 65, Pondichéry 1982.

Mokṣākaragupta, Tarkabhāṣā, edited by Embar Krishnamacharya, Gaekwad Oriental Series 94, Oriental Institute, Baroda 1942.

Mrgendratantra (vidyāpāda and yogapāda) with commentary of Nārāyanakanṭha, edited by Madhusudan Kaul Shastri, KSTS L, Bombay 1930.

Nadapāda (Nāropā), Sekoddeśatīkā, being a Commentary on the Sekoddeśa Section of the Kālakakra Tantra, edited by Mario E. Carelli, Gaekwad Oriental Series 90, Oriental Institute, Baroda 1941.

Nāgārjuna, Madhyamakasāstra with the Commentary Prasannapadā by Candrakīrti, edited by Swami Dwarika Das Shastri, Baudddha Bharati Series 16, Varanasi 1983.

Nāgeśabhatta, Paramalaghumañjūsā, with the commentary Jyotsnā by Pt. Kālikāprasād Shukla, Baroda Sanskrit Mahavidyalaya, M.S. University of Baroda, Baroda 1961.

Nārāyana Bhaṭṭa, Stavacintāmani with vivrti by Kṣemarāja, edited by Mukunda Ram Shastri, KSTS X, Srinagar 1918.

Patañjali, Vyākaraṇa-mahābhāṣya, edited by F. Kielhorn, III ed., revised and furnished with additional readings, references and select critical notes by K.V. Abhyankar, vols. I-III, Poona 1962-1972.

Patañjali, Vyākaraṇa-Mahābhāṣya with Kaiyaṭa's Pradīpa and Nāgeśa's Uddyota, edited with footnotes [...] by Raghunāth Kāshināth Shāstrī and Śivadatta D. Kudāla, vols. I-V, Bombay 1932-45.

Paribhāṣāsamgraha, edited critically by K.V. Abhyankar, Bhandarkar Oriental Research Institute, Poona 1967.

Rāmakantḥha, Spandakārikāvivrti, edited by J.C. Chatterji, KSTS VI, Srinagar 1913.

Ratnakīrti, Apohasiddhi, In: Ratnakīrtinibandhāvalī, edited by Anantalal Thakur, Tibetan Sanskrit Work Series 3, Patna 1957.

Rauravāgama, vol. I, édition critique par N.R.Bhatt, Publications de l'Institut Français d'Indologie No.18, Pondichéry 1961.

Sadyojyotis, Mokṣakārikā with the vrtti of Rāmakantḥha. In: Astaprakaranam, edited by Vrajavallabha Dvivedī, Yogatantra-Granthamālā 12, Varanasi 1988.

Sadyojyotis, Nareśvaraparīkṣā with Commentary by Rāmakantḥha, edited by Mukunda Ram Shastri, KSTS XLV, Srinagar 1926.

Śāntarakṣita, Tattvasamgraha with the Commentary 'Pañjikā' of Kamalaśīla, vols. I-II, critically edited by Swami Dwarikadas Shastri, Bauddha Bharati Series 1, Varanasi 1981.

Sārdhatriśatikālottarāgama, edizione critica a cura di R. Torella, Rivista degli Studi Orientali, vol. 50, Fasc. III-IV, 1976.

Sārdhatriśatikālottarāgama, avec le commentaire de Bhatta Rāmakantḥha, édition critique par N.R. Bhatt, Publications de l'Institut Français d'Indologie No. 61, Pondichéry 1979.

Sarvasiddhāntasamgraha, Chapters III-VII edited as Appendice II in La Vallée Poussin 1902.

Sataratnasamgraha with Sataratnollekhinī, edited by Pandit Panchanan Sastri, repr. Delhi 1984.

Śivāgrayogin, Saivaparibhāṣā (see Sastri 1982)

225

Somānanda, Śivadrṣṭi with the vrtti by Utpaladeva, edited by Pandit Madhusudan Kaul Shastri, KSTS No. LIV, Srinagar 1934.

Somaśambhupaddhati (see Brunner 1963-1977).

Svacchandatantra, with commentary ‘Uddyota’ by Kṣemarājācārya, edited by V.V. Dvivedi, Vols. I-II, Parimal Sanskrit Series No. 16, Delhi 1985.

Utpaladeva, The Siddhitrayī and the Pratyabhijñā-kārikā-vrtti, edited with notes by Pandit Madhusudan Kaul Shastri, KSTS No. XXXIV, Srinagar 1921.

Utpaladeva, The Śivastotrāvalī with the Sanskrit Commentary of Kṣemarāja, edited with Hindy Commentary by Rājānaka Lakṣmaṇa, Chowkhamba Sanskrit Series No. 15, Varanasi 1964.

Utpalavaiṣṇava (Bhagavadutpalācāryā) The Spandapradīpikā: a Commentary on the Spandakārikā, edited by M.S.G. Dyczkowski, Varanasi 1990.

Vāmanadatta, Samvitprakāśa, edited and translated by R. Torella (forthcoming)

Vasubandhu, Abhidharmakośabhāṣyam, deciphered and edited by Prof. Prahlad Pradhan, revised with introduction and indices by Dr. Aruna Haldar, K.P. Jayaswal Research Institute, Patna 1975.

Vasubandhu, Vijñaptimātratāsiddhi, Deux traités de Vasubandhu. Vimśatikā accompagnée d'une explication en prose et Trimśikā avec le commentaire de Sthiramati, publié par Sylvain Lévi, Bibliothèque de l'Ecole des Hautes Etudes, fasc. 245, Paris 1925.

Vedajñāna, Śivāgamaparibhāṣāmañjarī (see Dagens 1979)

Vijñānabhairava with the commentary partly by Kṣemarāja and partly by Śivopādhyāya, edited by Mukunda Ram Shastri, KSTS VIII with commentary Kaumudī by Ānanda Bhatta, KSTS IX, Bombay 1918.

Vimalaprabhā Ṭīkā of Kalki Śrī Puṇḍarīka on Śrī Laghukālacakratantrarāja by Mañjuśrīyaśa, vol. I, edited by Pt. Jagannath Upadhyaya, Central Institute of Higher Tibetan Studies, Sarnath 1986.

Yoginihrdayam with Commentaries Dīpikā of Amrtānanda and Setubandha of Bhāskararāya, edited by Gopīnātha Kavirāja, Sarasvatībhavana-Granthamālā vol. 7, Varanasi 1979.

TRANSLATIONS AND STUDIES

Aklujkar, A.N. (1970) The Philosophy of Bhartrhari's Trikāṇḍī. Ph.D. Thesis (un-published), Harvard University. Cambridge (Mss.)

Aklujkar, A.N. (1972) The Authorship of the Vākyapadīya-vrtti, Wiener Zeitschrift für die Kunde Südasiens, 16.

Aklujkar, A.N. (1990) Mahābhārata and Bhartrhari's Trikāṇḍī. VIII World Sanskrit Conference. Wien (unpublished).

Aklujkar, A.N. (1993) Once again on the Authorship of the Trikāṇḍī-Vrtti, Proceedings of the First International Conference on Bhartrhari: University of Poona, January 6-8, 1992, Asiatische Studien - Études Asiatiques, XLVII, 1.

Alper, H.P. (1979) Śiva and the Ubiquity of Consciousness: the Spaciousness of an Artful Yogi. Journal of Indian Philosophy, 7, No.1.

Alper, H.P. (1987) Svabhāvam avabhāsasya vimarśam: Judgement as a Transcendental Category in Utpaladeva's Śaiva Theology. Adyar Library Bulletin, 51.

Bailly, C.R. (1990) Shaiva Devotional Songs of Kashmir: A Translation and Study of Utpaladeva's Shivastotravali, Delhi.

Bapat, P.V. (1975) Vohāra : vyāhāra : vyavāhāra. In: Sanskrit and Indological Studies. Dr. V. Raghavan Felicitation Volume, Delhi.

Bandyopadhyay, N. (1979) The Buddhist Theory of Relation between pramā and pramāṇa. Journal of Indian Philosophy, 7, No.1.

Biardeau, M. (1964) Théorie de la connaissance et philosophie de la parole dans le Brahmanisme classique. Paris-La Haye.

van Bijlert, V.A. (1989) Epistemology and Spiritual Authority. Wien.

Bronkhorst, J. (1987) Mahābhāsyadīpikā of Bhartrhari, Fasc.IV: Āhnika I, critically edited and translated by ..., Post-Graduate and Research Department Series No. 28, Bhandarkar Oriental Research Institute. Poona.

Bronkhorst, J. (1988) Études sur Bhartrhari, 1: L'auteur et la date de la Vritti, Bulletin d'Études Indiennes, 6.

Bronkhorst, J. (1991) Two Literary Conventions of Classical India. Asiatische Studien - Études Asiatiques, XLV, 2.

Brunner-Lachaux, H. (1986) The Pāśupatas as seen by the Śaivas. In: Sanskrit and World Culture, Schriften zur Geschichte und Kultur des Alten Orients, vol. 18, Berlin.

Brunner-Lachaux, H. (1963-1977) Somāśambhupaddhati, parties I-III. Traduction, introduction et notes par ..., Publications de l'Institut Français d'Indologie No. 25 I-III. Pondichéry.

Brunner-Lachaux, H. (1985) Mrgendrāgama. Section des rites et section du comportement, avec la vrtti de Bhaṭṭanārāyaṇakanṭha. Traduction, introduction et notes par ..., Publications de l'Institut Français d'Indologie No. 69. Pondichéry.

Bühler, G. (1877) A Detailed Report on a Tour in Search of Sanskrit MSS made in Kasmir, Rajputana and Central India, Journal of the Bombay Branch of the Royal Asiatic Society, Extra No. 34a.

Cardona, G. (1974) Pānini's kārakas: Agency, Animation and Identity. Journal of Indian Philosophy, 2.

Chatterjee, J.C. (1914²) Kashmir Shaivism, Research and Publication Department. Srinagar (II Edition 1962).

Chaturvedī, R. (1986) The Sivadṛṣṭi of Śrī Somānanda Nātha with the Sanskrit Commentary and Hindi Translation by ... Varānaseya Samskrta Samsthāna Granthamālā No. 35. Varanasi.

Dagens, B. (1979) Śaivāgamaparibhāṣāmañjarī (Le Florilège de la doctrine śivaite), édition critique, traduction et notes, Publications de l'Institut Français d'Indologie No. 60. Pondichéry 1979.

Dreyfus, G. (1991) Dharmakīrti's Definition of pramāṇa and its Interpreters. In: Steinkellner 1991 (Ed.).

Dvivedi, V.V. (1983) Luptāgamasamgraha, Part II, collected and edited by ..., Yogatantra-Ratnamālā 10. Varanasi.

Dyczkowski, M.S.G. (1987) The Doctrine of Vibration: An Analysis of the Doctrines and Practices of Kashmir Shaivism, SUNY Press. Albany.

Dyczkowski, M. S. G. (1988²) The Canon of the Śaivāgama and the Kubjikā Tantras of the Western Kaula Tradition. Albany (II Edition Delhi 1989).

Dyczkowski, M.S.G. (1990) Self-Awareness, Own Being and Egoity. Varanasi.

Franco, E. (1991) The Disjunction in Pramāṇavārttika, Pramāṇasiddhi Chapter 5. In: Steinkellner 1991 (Ed.).

Frauwallner, E. (1934) Dharmakīrtis Sambandhaparīkṣā. Text und Übersetzung. Wiener Zeitschrift für die Kunde des Morgenlandes, 41.

Frauwallner, E. (1959) Dignāga, sein Werk und seine Entwicklung. Wiener Zeitschrift für die Kunde Südasiens, 3.

Frauwallner, E. (1961) Landmarks in the History of Indian Logic. Wiener Zeitschrift für die Kunde Südasiens, 5.

Frauwallner, E. (1962) Aus der Philosophie der śivaitischen Systeme, Deutsche Akademie der Wissenschaften zu Berlin Vorträge und Schriften 78. Berlin.

Funayama, T. (1992) A Study of kalpānāpodha: A Translation of the Tattvasamgrahapañjikā by Kamalaśīlā on the Definition of Direct Peception. Zinbun: Annals of the Institute for Research Humanities. Kyoto University, 27.

Gaurinath Shastri (1959) The Philosophy of Word and Meaning: Some Indian Approaches with Special Reference to the Philosophy of Bhartrhari, Calcutta Sanskrit College Research Serics No. V. Calcutta.

Gaurinath Shastri (1980) A Study in the Dialecticts of Sphota. Revised new Edition. Delhi.

Gnoli, R. (1957) Śivadṛṣṭi by Somānanda. East and West, 8.

Gnoli, R. (1959) Vāc. Il secondo capitolo della Śivadṛṣṭi di Somānanda. Rivista degli Studi Orientali, 34.

Gnoli, R. (1960) see Dharmakīrti, Pramāṇavārttikam. The First Chapter with the Autocommentary

Gnoli, R. (1968) The Aesthetic Experience according to Abhinavagupta, Chowkhamba Sanskrit Studies No.LXII. II Ed. Varanasi.

Gnoli, R. (1972) Abhinavagupta, Luce delle Sacre Scritture (Tantrāloka), introduzione, traduzione c note di ... Torino.

Gnoli, R. (1985) Il Commento di Abhinavagupta alla Parātrimśikā (Parātrimśikātat-tavavivaranam), traduzione e testo, Serie Orientale Roma LVIII, IsMEO. Roma.

Gonda, J. (1963) The Vision of the Vedic Poets. The Hague.

Gonda, J. (1977) Medieval Religious Literature in Sanskrit, in Gonda, J. (Ed.) History of Indian Literature, vol. II fasc. I. Wiesbaden.

Goudriaan, T (1992) (Ed.) Ritual and Speculation in Early Tantrism. Studies in Honor of André Padoux, SUNY Press. Albany.

Goudriaan, T. and Gupta, S. (1981) Hindu Tantric and Śākta Literature. In: Gonda, J. (Ed.) History of Indian Literature, vol. II fasc. II. Wiesbaden.

Hara, M. (1966) Review of Silburn, L. La Bhakti: Le Stavatāmani de Bhaṭṭanārāyana, text traduit et commenté. Paris 1964. Indo-Iranian Journal, IX, No. 3.

Hattori, M. (1968) Dignāga. On Perception, being the Pratyakṣapariccheda of Dignāga's Pramāṇasamuccaya, Harvard Oriental Series No. 47. Cambridge (Mss.).

Hattori, M. (1977) The Sautrāntika Background of the apoha Theory. In: Kawamura and Scott (Eds.) 1977.

Hattori, M. (1980) Apoha and pratibhā. In: Nagatomi and al. (Eds.) 1980

Hayes, R.P. (1983) Jñāendrabuddhi. Journal of the American Oriental Society, 103, No.4.

Herzberger, R. (1986) Bhartrhari and the Buddhists. An Essay in the Development of Fifth and Sixth Century Indian Thought, Studies of Classical India 8. Dordrecht.

Hulin, M. (1978) Le principe de l'ego dans la pensee indienne classique: la notion d'ahamkāra, Publications de l'Institut de Civilisation Indienne 44. Paris.

Ingalls, D.H.H. (1990) (Ed.). The Dhvanyāloka of Ānandavardhana with the Locana of Abhinavagupta, translated by D.H.H. Ingalls, J.M. Masson and M.V. Patwardhan, edited with an Introduction by ...., Harvard Oriental Series No.49. Harvard (Mss.) - London.

Iwata, T. (1991) Sahopalambhaniyama. Struktur und Entwicklung des Schlusses von der Tatsache, dass Erkenntnis und Gegenstand aussliesslich zusammen wahrgenommen werden, auf deren Nichtverschiedenheit, Teil I-II, Alt- und Neu-Indische Studien 29, Universität Hamburg. Stuttgart.

Iwata, T. (1991) On the Classification of the Three Kinds of Reason in Pramāṇaviniścaya III - Reduction of Reasons to svabhāvahetu and kāryahetu. In: Steinkellner (Ed.) 1991.

Iyer, K.A.S. (1964) Bhartrhari on vyākarana as a means of attaining mokṣa. Adyar Library Bulletin, 28.

Iyer, K.A.S. (1965) The Vākyapadīya of Bhartrhari with the Vrtti. Chapter I. English Translation.. Deccan College. Poona.

Iyer, K.A.S. (1969) Bhartrhari. A Study of the Vākyapadīya in the light of the Ancient Commentaries. Deccan College, Poona.

Iyer, K.A.S. (1971) The Vākyapadīya of Bhartrhari. Chapter III, part I. English Translation. Deccan College, Poona.

Iyer, K.A.S. (1974) The Vākyapadīya of Bhartrhari. Chapter III, part II. English Translation. Delhi.

Jha, H.M. (1981) Trends of Linguistic Analysis in Indian Philosophy. Varanasi.

Jha, V.N. (1990) The Philosophy of Relations, Bibliotheca Indo-Buddhica Series No.66. Delhi.

Kajiyama, Y. (1963) Trikapañcakacintā : Development of the Buddhist Theory on the Determination of Causality. Miscellanea Indologica Kiotensia, 4-5.

Kajiyama, Y. (1966) An Introduction to Buddhist Philosophy. An Annotated Translation of the Tarkabhāṣā of Mokṣākaragupta. Memoirs of the Faculty of Letters. Kyoto University. 10.

Kano, K. (1991) On the Background of PV II 12ab - The Origin of Dharmakīrti's Idea of arthakriyā. In: Steinkellner 1991 (Ed.).

Katsura, S. (1984) Dharmakīrti's Theory of Truth. Journal of Indian Philosophy, 12.

Katsura, S. (1986) Jñānaśrīmitra on apoha. In: Matilal and Evans 1986 (Eds.).

Katsura, S. (1991) Dignāga and Dharmakīrti on apoha. In: Steinkellner 1991 (Ed.).

Kaviraj, G. (1966) Aspects of Indian Thought, University of Burdwan.

Kaw, R.K. (1967) The Doctrine of Recognition (Pratyabhijñā Philosophy), Vishveshvaranand Institute Series No.40. Hoshiarpur.

Kaw, R.K. (1975-76) Pratyabhijñākārikā of Utpaladeva, parts I-II, Sharada Peetha Indological Research Series 11-12. Srinagar.

Kawamura, L.S. and Scott, K. (1977) Buddhist Thought and Asian Civilisation. Essays in Honour of H.V. Guenther on his Sixtieth Birthday. Emeryville.

Kielhorn, L.F. (1874) The Paribhāsenduśekhara of Nāgojībhatta, edited and explained by ...., part II: translation and notes. Bombay.

229

Kitagawa, H. (1955) A Refutation of Solipsism (Annotated Translation of Samtānāntarasiddhi ). Journal of the Greater India Society, 14.

Kitagawa, H. (1960) A Note on the Methodology in the Study of Indian Logic. Journal of Indian and Buddhist Studies, VIII, No.1.

Larson, G.J. (1976) The aesthetic (rasāsvāda) and the religious (brahmāsvāda) in Abhinavagupta's Kashmir Shaivism. Philosophy East and West, 26.

de La Vallée Poussin, L. (1901-1902) Le Bouddhisme d'après les sources Brahmaniques. Le Muséon, 2-3.

de La Vallée Poussin, L. (1923-1931) L'Abhidharmakośa de Vasubandhu, traduit et annoté par..., vols. I-VI, Société Belge d'Études Orientales. Louvain-Paris.

de La Vallée Poussin, L. (1928-48), Vijñaptimātratāsiddhi. La Siddhi de Hiuan-Tsang, traduite et annotée par..., tomes I-III, Buddhica: Documents et Travaux pour l'étude du Bouddhisme publiés sous la direction de Jean Przyluski. Paris.

Lévi, S. (1932) Matériaux pour l'étude du système Vijñaptimātrata. Bibliothèque de l'École des Hautes Études, fasc. 260, Paris.

Lindtner, Ch (1984) Marginalia to Dharmakīrti's Pramāṇaviniścaya I-II, WZKSO,28.

Lindtner, Ch. (1993) Linking up Bhartrhari and the Bauddhas. Proceedings of the First International Conference on Bhartrhari: University of Poona, January 6-8, 1992, Asiatische Studien - Études Asiatiques, XLVII, 1.

Masson, J.L. and Patwardhan, M.V. (1969) Śāntarasa and Abhinavagupta's Philosophy of Aesthetics. Bhandarkar Oriental Research Institute. Poona.

Masson, J.L. and Patwardhan, M.V. (1970) Aesthetic Rapture, 2 vols., Deccan College. Poona.

Matilal, B.K. (1968) The Navya-Nyāya Doctrine of Negation, Harvard Oriental Series 46. Cambridge (Mss.).

Matilal, B.K. (1968a) Indian Theories of Knowledge and Truth: Review Article of: Mohanty, J. Gaṅgeśa's Theory of Truth, Santiniketan 1966. Philosophy East and West, XVIII, 4.

Matilal, B.K. (1985) Logic, Language and Reality: An Introduction to Indian Philosophical Studies, Delhi.

Matilal, B.K. (1986) Perception. An Essay on Classical Indian Theories of Knowledge. Oxford.

Matilal, B.K. (1991) The World and the Word. Delhi.

Matilal, B.K. and Evans, R.D. (1986) (Eds.) Buddhist Logic and Epistemology. Studies in the Buddhist Analysis of Inference and Language, Studies of Classical India 7. Dordrecht.

Mikogami, E. (1979) Some Remarks on the Concepts of arthakriyā. Journal of Indian Philosophy, 7, No.1.

Mookerjee, S. and Nagasaki, H. (1964) The Pramāṇavārttikam of Dharmakīrti. An English Translation of the First Chapter with the Autocommentary and with Elaborate Comments [Kārikās I-L.I], The Nava Nālandā Mahāvihāra Research Publication Vol.IV. Patna.

Muller-Ortega, P.E. (1988) The Triadic Hearth of Śiva: Kaula Tantricisn of Abhinavagupta in the Non-Dual Shaivism of Kashmir, SUNY Press. Albany.

Nagatomi, M. (1967-68) Arthakriyā. Adyar Library Bulletin, 31-32.

Nagatomi, M. and al. (1980) (Eds.) Sanskrit and Indian Studies. Essays in Honour of D.H.H. Ingalls, Studies of Classical India 2. Dordrecht.

230

Padoux, A. (1975) La Parātrīśikālaghuvṛtti de Abhinavagupta, Publications de l'Institut de Civilisation Indienne. Paris.

Padoux, A. (1990) Vāc. The Concept of the Word in Selected Hindu Tantras, SUNY Press. Albany.

Padoux, A. (1986) (Ed.) Mantras et diagrammes rituels dans l'Hindouisme. Centre National de la Recherche Scientifique. Paris.

Padoux, A. (1990) (Ed.) L'image divine. Culte et meditation dans l'Hindouisme, Centre National de la Recherche Scientifique. Paris.

Pandey, K.C. (1954) Bhāskarī, vol.III. Īśvarapratyabhijñāvimarśinī, English Translation, The Princess of Wales Sarasvati Bhavana Texts No.54. Lucknow.

Pandey, K.C. (1963) Abhinavagupta. An Historical and Philosophical Study, II Edition. Varanasi.

Raghavan, V. (1981) Abhinavagupta and his Works, Chaukhambha Oriental Research Studies No. 20. Varanasi.

Randle, H.N. (19262) Fragments from Diṅnāga. London (II Ed. Delhi 1981).

Rastogi, N. (1977-78) Recognition in Pratyabhijñā School: A Study in Epistemology. Annals of Bhandarkar Oriental Research Institute (Diamond Jubilee Volume), 58.

Rastogi, N. (1979) The Krama Tantricism of Kashmir, vol.I. Delhi.

Rastogi, N. (1986) Theory of Error according to Abhinavagupta. Journal of Indian Philosophy,14.

Rastogi, N. (1987) Introduction to the Tantrāloka. Delhi.

Ray, S.C. (1955) Studies on the History of Religion in Ancient Kāśmīra. The Journal of the Bihar Research Society, XLI, Part II.

Rucgg, D.S. (1959) Contributions à l'histoire de la philosophie linguistique indienne, Publications de l'Institut de Civilisation Indienne. Paris.

Sanderson, A. (1985) Purity and Power among the Brahmans of Kashmir. In: M. Carrithers, S. Collins, S. Lukes The Category of the Person, Cambridge University Press. Cambridge (Mss.).

Sanderson, A. (1986) Maṇḍala and Āgamic identity in the Trika of Kashmir. In: Padoux 1986 (Ed.).

Sanderson, A. (1988) Śaivism and the Tantric Tradition. In: Sutherland, S. et al. (Eds.) The World's Religions. London.

Sanderson, A. (1990) The Visualisation of the Deities of the Trika. In: Padoux 1990 (Ed.).

Sanderson, A. (1992) The Doctrines of the Mālinīvijayottaratantra. In: Goudriaan 1992 (Ed.).

Sastri, S.S.S. (1982) The Śaivaparibhāṣā of Śivāgrayogin, University of Madras, Madras.

Shah, N.J. (1967) Akalaṅka's Criticism of Dharmakīrti's Philosophy: a Study, L.D. Series No.11, L. D. Institute of Indology. Ahmedabad.

Sharma, P.S. (1972) The Kālasamuddeśa of Bhartrhari's Vākyapadīya (together with Helarāja's commentary translated from the Sanskrit for the first time). Delhi.

Siderits, M. (1986) Was Śāntarakṣita a "Positivist"? In: Matilal and Evans 1986 (Eds.).

Sinha, J. (19382) Indian Realism. London (II Ed. Delhi 1972).

Stcherbatsky, T. (1930–322) Buddhist Logic, vols. I–II, Bibliotheca Buddhica XXVI. Leningrad (repr. New York 1962).

Stcherbatsky, T. (1936) Madhyānta-vibhaṅga. Discourse on Discrimination between dhu and Extremes, ascribed to Bodhisattva Maitreya, commented by Vasubandhu and Sthiramati. Translated from the Sanskrit by ... Bibliotheca Buddhica XXX. Leningrad (repr. Soviet Indology Series No. 5. Calcutta 1971).

Stcherbatsky, T. (1969) Establishment of the Existence of Other Minds. Free Translation of Dharmakīrti's Santanāntara-siddhi with Viñītadeva's Santanāntara-siddhi-tīkā. In: Papers of Stcherbatsky, translated for the first time into English by H.C. Gupta, edited with an Introduction by D. Chattopadhyaya, Soviet Indology Series No.2. Calcutta.

Steinkellner, E. (1966) Bemerkungen zu Īśvarasenas Lehre vom Grund. Wiener Zeitschrift für die Kunde Südasiens, 10.

Steinkellner. E. (1967) Dharmakīrti's Hetubindu, Teil I, Tibetischer Text und rekonstruierter Sanskrit-Text; Teil II, Übersetzung und Anmerkungen. Österreichische Akademie der Wissenschaften. Wien.

Steinkellner, E. (1971) Wirklichkeit und Begriff bei Dharmakīrti. Wiener Zeitschrift für die Kunde Südasiens, 15.

Steinkellner, E. (1972) New Sanskrit Fragments of Pramāṇaviniścaya, First Chapter. Wiener Zeitschrift für die Kunde Südasiens, 16.

Steinkellner, E. (1973) Dharmakīrti's Pramāṇaviniścaya. Zweites Kapitel: Svārthānu-mānam. Teil I, Tibetischer Text und Sanskrittexte*, Österreichische Akademie der Wissenschaften. Wien.

Steinkellner, E. (1974) On the interpretation of the svabhāvahetu. Wiener Zeitschrift für die Kunde Südasiens, 18.

Steinkellner, E. (1979) Dharmakīrti's Pramāṇaviniścaya. Zweites Kapitel: Svārthānu-mānam. Teil II, Übersetzung und Anmerkungen. Österreichische Akademie der Wissenschaften. Wien.

Steinkellner, E. (1984) Svabhāvapratibandha again. In: Studies on Mysticism in Honour of the 150th Anniversary of Kobo-Daishi's Nirvāṇam. Acta Indologica-ca. VI.

Steinkellner, E. (1992) The Logic of the svabhāvahetu in Dharmakīrti's Vādanyāya. In: Steinkellner 1992 (Ed.).

Steinkellner, E. (1992) (Ed.) Studies in the Buddhist Epistemological Tradition, Proceedings of the Second International Dharmakīrti Conference, Vienna, June 11-16, 1989, Österreichische Akademie der Wissenschaften, Philophisch-Historische Klasse, Denkschriften, 222. Band, Beiträge zur Kultur-und Geistesgeschichte Asiens Nr.8. Wien.

Taber, J.A. (1986) Utpaladeva's Īśvarasiddhi. Adyar Library Bulletin, 52.

Tola, F. and Dragonetti, C. (1990) Some Remarks on Bhartṛhari's Concept of pratibhā. Journal of Indian Philosophy, 18.

Torella, R. (1979) Śivasūtra con il commento Di Ksemarāja. Roma.

Torella. R. (1979) Due capitoli del Sarvadarśanasamgraha: Śaivadarśana e Pratya-bhijñādarśana. Rivista degli Studi Orientali, LIII, fasc. III-IV.

Torella, R. (1987) Examples of the Influence of Sanskrit Grammar on Indian Philosophy. East and West, 37.

Torella, R. (1988) A Fragment of Utpaladeva's Īśvarapratyabhijñā-vivṛti. East and West, 38.

Torella, R. (1991) Importanza di Utpaladeva. In: Atti del IV e V Convegno Nazionale

232

di Studi Sanscriti (Torino, 24 gennaio 1986 - Milano, 8 novembre 1988), editi da

O. Botto, a cura di S. Piano e V. Agostini. Torino.

Torella, R. (1992) The Pratyabhijñā and the Logical-epistemological School of Bud-

dhism. In: Goudriaan 1992 (Ed.).

Torella, R. (1994) On Vāmanadatta. In P.S. Filliozat, C.P. Bhatt, S.P. Narang

(Eds.) N. Ramachandra Bhatt Felicitation Volume, Delhi.

Torella, R. (1998) The kañcukas in the Śaiva and Vaiṣṇava Tantric Tradition:

A Few Considerations between Theology and Grammar, in: G. Oberhammer

(Ed.), Studies in Hinduism, II, Miscellanea to the Phenomenon of Tantras,

Österreichische Akademie der Wissenschaften, Philosophisch-Historische

Klasse, Sitzungsberichte, 662. Band, Wien.

Torella, R. (forthcoming) The Word in Abhinavagupta's Ḅrhadvimarśinī. In:

Torella, R. (Ed.), Le Parole e i Marmi. Studi in onore di Raniero Gnoli nel

suo 70° compleanno, Serie Orientale Roma, IsIAO, Roma (forthcoming).

Tucci, G. (1930) The Nyāyamukha of Dignāga [...] after Chinese and Tibetan

Materials, Heidelberg.

Tucci, G. (1956) Minor Buddhist Texts. Part 1, Serie Orientale Roma IX, Roma.

Vetter, T. (1964) Erkenntnissprobleme bei Dharmakīrti, Österreichische Akademie der

Wissenschaften, Wien.

Vivanti, M.P. (1985) Il Kiranagama. Testo e traduzione del Vidyāpāda, Supplemen-

to n. 3 agli Annali dell'Istituto Orientale di Napoli vol. 35 (1975), fasc. 2.

Vyas, S.P. (1989) Siddhitrayī of Rājanaka Utpala Deva, edited and translated [into

Hindi] by ..., Kashi Sanskrit Series 259. Varanasi.

Wezler, A. (1982) Paralipomena zum Sarvasarvāmakatvavāda I: Mahābhāṣya zu Pān.

4.3.155 und seine einheimischen Erklärer. Wiener Zeitschrift für die Kunde

Südasiens, 26.

Wezler, A. (1992) Paralipomena zum Sarvasarvāmakatvavāda II: On the Sar-

vasarvāmakatvavāda and its Relation to the Vr̥kṣāyurveda. Studien zur Indologie

und Iranistik , 16.

Yamaguchi, S. - Mayer, H. (1929) Examen de l'object de la connaissance (Ālam-

banaparīkṣā): text tibétain et chinois et traduction des stances et du commentaire:

eclaircissement et notes d'après le commentaire tibétain de Vinitadeva. Journal

Asiatique, 214.

233