1. A Calculus of Creative Expression The Central Chapter of Dandin's Kavya Darsa John Fredrick Eppling (Thesis)

INFORMATION TO USERS

The negative microfilm copy of this dissertation was prepared and inspected by the

school granting the degree. We are using this film without further inspection or

change. If there are any questions about the content, please write directly to the

school. The quality of this reproduction is heavily dependent upon the quality of

the original material.

The following explanation of techniques is provided to help clarify notations which

may appear on this reproduction.

1. Manuscripts may not always be complete. When it is not possible to obtain

missing pages, a note appears to indicate this.

2. When copyrighted materials are removed from the manuscript, a note ap-

pears to indicate this.

3. Oversize materials (maps, drawings, and charts) are photographed by sec-

tioning the original, beginning at the upper left hand corner and continu-

ing from left to right in equal sections with small overlaps.

4. Most photographs reproduce acceptably on positive microfilm or micro-

fiche but lack clarity on xerographic copies made from the microfilm. For

any illustrations that cannot be reproduced satisfactorily by xerography,

photographic prints can be purchased at additional cost and tipped into

your xerographic copy. Requests can be made to the Dissertations Cus-

tomer Services Department.

U·M·I

Dissertation

Information Service

University Microfilms International

A Bell & Howell Information Company

300 N. Zeeb Road, Ann Arbor, Michigan 48106

Order Number 8914509

A calculus of creative expression: The central chapter of

Dandin’s “Kāvyādarśa”

Eppling, John Frederick, Ph.D.

The University of Wisconsin - Madison, 1989

Copyright ©1989 by Eppling, John Frederick. All rights reserved.

U·M·I

300 N. Zeeb Rd.

Ann Arbor, MI 48106

A dissertation entitled

A Calculus of Creative Expression:

The Central Chapter of Dandin's Kavyadarsa

submitted to the Graduate School of the

University of Wisconsin-Madison in partial fulfillment of

the requirements for the degree of Doctor of Philosophy

by

John Frederick Eppling

Degree to be awarded: December 19_____May 19_89August 19_____

Approved by Dissertation Readers:

Major Professor

Date of Examination

December 14, 1988

Dean, Graduate School

A CALCULUS OF CREATIVE EXPRESSION:

THE CENTRAL CHAPTER OF DANDIN'S KĀVYĀDARŚA

by

JOHN FREDERICK EPPLING

A thesis submitted in partial fulfillment of the

requirements for the degree of

Doctor of Philosophy

(South Asian Language and Literature )

at the

UNIVERSITY OF WISCONSIN - MADISON

1989

©

Copyright

by

John

Frederick

Eppling

1989

All

Rights

Reserved

Table of Contents

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . viii

Epigraphs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii

I. TEXTUAL CONTEXT . . . . . . . . . . . . . . . . . . . . . . . . . 1

A. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

The Text . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

Abbreviations and Editions Cited . . . . . . . . . . . . . . . . . . 43

Notes: Introduction . . . . . . . . . . . . . . . . . . . . . . . . 48

B. Chapter One . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

Homage to Sarasvatī . . . . . . . . . . . . . . . . . . . . . . . . 54

Daṇḍin’s Design and the Necessity of Kāvya . . . . . . . . . . . . 56

The Tradition and Possible Predecessors . . . . . . . . . . . . . . 62

Notes [1.1] - [1.9] . . . . . . . . . . . . . . . . . . . . . . . . 75

The Nature of Kāvya and its Formulaic Division /

On Metre and the Versatic Form . . . . . . . . . . . . . . . . . . 81

The Sargabandha or Mahākāvya . . . . . . . . . . . . . . . . . . . 100

Critical Acceptance as the Essential Criterion --

i

Not Formulaic Adherence . . . . . . . . . . . 109

The Ākhyāyikā or Kathā . . . . . . . . . . . 110

Again on the Freedom of the Kavi . . . . . . . . 113

Mixed Compositions -- The Campū . . . . . . . . 114

Notes [1.10] - [1.31] . . . . . . . . . . . . 130

Language and Kāvya . . . . . . . . . . . . . 142

Notes [1.32] - [1.38] . . . . . . . . . . . . 151

The Ten Guṇas (or "Qualities") and the Mārgas

(or "Styles") . . . . . . . . . . . . . . . 154

Notes [1.40] - [102] . . . . . . . . . . . . . 184

The Kavi and the Generation of Kāvya . . . . . . 186

Notes [1.103] - [1.105] . . . . . . . . . . . . 198

C. Chapter Three . . . . . . . . . . . . . . . 200

Yamaka or Variations of "Phonemic Repetition" . 200

Samdaṣṭa Yamaka . . . . . . . . . . . . . . . 212

Samudga Yamaka . . . . . . . . . . . . . . . . 213

Pāda Abhyāsa or the "Repetition of Pādas" . . . 214

Śloka Abhyāsa or "Stanzaic Repetition" . . . . . 215

Mahā Yamaka . . . . . . . . . . . . . . . . . 216

ii

Saṃsṛṣṭi or the "Combination of Yamakas" . . . 217

Pratiloma Yamaka or "Repetition in Reverse" . . 218

Duṣkara Śabda Alamkāras -- Those "Difficult to Construe" . . . . . . . . . . . . . 221

Niyama or "Phonemic Restriction" . . . . . . . 229

Prahelikā or the "Riddle" . . . . . . . . . . . 232

The Ten Doṣas or "Faults" and their Positive Transformations . . . . . . . . . . . . . . . . 247

Notes [3.1] - [3.185] . . . . . . . . . . . . . . 273

II. THE VERSES OF THE SECOND CHAPTER . . . . . . . 277

A. An Enumeration -- with English and Sanskrit Titles . . . . . . . . . . . . . . . . . . . . 278

B. The Central Text . . . . . . . . . . . . . . . . 333

Definition of Alamkāra . . . . . . . . . . . . . 333

The Intention of the Writer . . . . . . . . . . . 337

Indicating the Distinction between the Alamkāras Previously Discussed and Those About to be Discussed . . . . . . . . . . . . . . . . . . . . 339

iii

The Thirty-Five Artha Alamkāras . . . . . . . . 341

Notes [2.1] - [2.7] . . . . . . . . . . . . 344

Svabhāvokti Alamkāra . . . . . . . . . . . . 345

Notes [2.8] . . . . . . . . . . . . . . . . 386

Notes [2.9] - [2.13] . . . . . . . . . . . . 407

Upamā Alamkāra . . . . . . . . . . . . . . 408

Notes [2.14] . . . . . . . . . . . . . . . . 456

Notes [2.15] - [2.65] . . . . . . . . . . . . 584

Rūpaka Alamkāra . . . . . . . . . . . . . . 589

Notes [2.66] . . . . . . . . . . . . . . . . 596

Notes [2.67] - [2.96] . . . . . . . . . . . . 669

Dīpaka Alamkāra . . . . . . . . . . . . . . 671

Notes [2.97] . . . . . . . . . . . . . . . . 681

Notes [2.98] - [2.115] . . . . . . . . . . . . 725

Āvṛtti Alamkāra . . . . . . . . . . . . . . . 727

Notes [2.116] - [2.119] . . . . . . . . . . . . 738

Ākṣepa Alamkāra . . . . . . . . . . . . . . . 739

Notes [2.120] . . . . . . . . . . . . . . . . 750

Notes [2.121] - [2.168] . . . . . . . . . . . . 822

iv

Arthāntaranyāsa Alamkāra . . . . . . . . . 828

Notes [2.169] - [2.179] . . . . . . . . . . 860

Vyatireka Alamkāra . . . . . . . . . . . . 863

Notes [2.180] - [2.198] . . . . . . . . . . 908

Vibhāvanā Alamkāra . . . . . . . . . . . . 911

Notes [2.199] - [2.204] . . . . . . . . . . 925

Śamāsokti Alamkāra . . . . . . . . . . . . 926

Notes [2.205] - [2.213] . . . . . . . . . . 947

Atiśayokti [Atiśaya] Alamkāra . . . . . . . . 948

Notes [2.214] - [2.220] . . . . . . . . . . 971

Utprekṣā Alamkāra . . . . . . . . . . . . 972

Notes [2.221] - [2.234] . . . . . . . . . . 1001

Hetu Alamkara . . . . . . . . . . . . . . 1002

Notes [2.235] - [2.259] . . . . . . . . . . 1057

Sūkṣma Alamkāra . . . . . . . . . . . . . . 1060

Leśa Alamkāra . . . . . . . . . . . . . . . 1067

Yathāsaṃkhya [Samkhyāna / Krama] Alamkāra . . 1081

Notes [2.260] - [2.274] . . . . . . . . . . 1087

Preyas / Rasavat / Ūrjasvin Alamkāras . . . . 1089

v

Notes [2.275] - [2.294] . . . . . . . . . . . 1154

Paryāyokta Alamkāra . . . . . . . . . . . . . 1162

Samāhita Alamkāra . . . . . . . . . . . . . 1169

Udātta Alamkāra . . . . . . . . . . . . . 1174

Apahnuti Alamkāra . . . . . . . . . . . . . 1186

Notes [2.295] - [2.309] . . . . . . . . . . . 1199

Śleṣa [Śliṣṭa] Alamkāra . . . . . . . . . . . 1200

Viśeṣokti [Viśeṣa] Alamkāra . . . . . . . . . . 1247

Tulyayogitā Alamkāra . . . . . . . . . . . . 1264

Virodha Alamkāra . . . . . . . . . . . . . 1275

Notes [2.310] - [2.339] . . . . . . . . . . . 1293

Aprastutapraśaṃsā [Aprastutastotra]

Alamkāra . . . . . . . . . . . . . 1295

Vyājastuti Alamkāra . . . . . . . . . . . . . 1304

Nidarśana Alamkāra . . . . . . . . . . . . . 1318

Sahokti Alamkāra. . . . . . . . . . . . . . 1326

Parivrtti Alamkāra . . . . . . . . . . . . . 1326\1335

Āśiṣ Alamkāra . . . . . . . . . . . . . . . 1340

Saṃsṛṣṭi [Saṃkīrṇa] Alamkāra . . . . . . . . . . 1348

vi

Notes [2.340] - [2.363] . . . . . . . . . . . 1368

Bhāvika Alamkāra . . . . . . . . . . . . . . 1370

Notes [2.364] - [2.368] . . . . . . . . . . . 1390

III. TEXTUAL TRANSMISSION . . . . . . . . . . . . . . 1392

Kannada . . . . . . . . . . . . . . . . . . . 1395

Notes: Kannada . . . . . . . . . . . . . . . . . 1403

Sinhalese . . . . . . . . . . . . . . . . . . . 1406

Notes: Sinhalese . . . . . . . . . . . . . . . . . 1416

Pāli . . . . . . . . . . . . . . . . . . . . . 1419

Notes: Pāli . . . . . . . . . . . . . . . . . . . 1433

Tibetan . . . . . . . . . . . . . . . . . . . . 1435

Notes: Tibetan . . . . . . . . . . . . . . . . . . 1517

BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . 1546

vii

Acknowledgments

However inadequate these brief words may be I would

like to acknowledge with deep gratitude the generous

assistance of all those who contributed so materially and

psychically to the completion of this seemingly never-ending

project. My thanks to John Paul and the Fulbright-Hays

Foundation for the means to realize extended research

overseas. During my initial stay in England the renowned

Tibetan scholar Dr. David Snellgrove was extremely helpful

and courteous, facilitating access to the collections of the

University of London's School of Oriental and African

Studies. I would like to acknowledge my debt not only to

this institution, but to the kind and helpful staffs of the

Oriental Section of the British Museum and the India Office

Library. In India the staff of the Fulbright House in Delhi

under the guiding hand of Mrs. Sharada Nayak was always a

welcome oasis of efficiency. Mr. Chawla was always ready to

handle yet another complex arrangement, and my friends in

viii

the finance section Mr. Nehru and Ram Kumar were always

ready with a good word and consistently amazed me with their

ability to financially track me around the countryside.

Work on the Sanskrit material was conducted under the

auspices of Andhra Pradesh University in Waltair and I would

like to express my gratitude to this fine school.

My deepest thanks to the Venerable Zanthong Rinpoche,

Director of the Tibetan Institute in Sarnath, and the

inimitable Gene Smith for their hospitality, initial advice

and guidance on Tibetan material, and similarly to Dr.

Lokesh Chandra, Director of the Indian Academy of

International Culture, for sharing his thoughts and advice.

This research would not have been possible without the

excellent facilities offered by the Tibetan Library of Works

and Archives in Dharamsala, so generously made available by

its director Mr. Gyatso Tsering. Deepest thanks to my

colleagues and friends there -- to Sonam ever courteous and

ever ready to deal with another bizarre request, and Norbu

for all the favors granted; to the scholars Tashi Tsering

ix

and the Venerable Jampa Sampten always ready to help; and to

Thubten Tsering, invaluable translator and friend, companion

on all those memorable visits to Barshi Lha. The staff of

the Sikkim Research Institute of Tibetology in Gangtok

graciously allowed me access to their valued collection.

Warm thanks to Dr. David Jackson for the fine Mussooree

evenings of conversation, opening my eyes to the role of

Sa-skya Paṇḍita in the early transmission of kāvya into

Tibet, and of course to Richard and Chris for all their

sustaining humor.

Closer to home, valued teachers and advisors at the

University of Wisconsin, Madison, have been crucial

throughout. I would like to express my gratitude to Dr.

Stephan Beyer and Geshe Sopa for their generous Tibetan

teachings across the years, and to Dr. John Newman for his

commiseration and willingness to clarify final details; to

Dr. V. N. Rao, who indeed first pointed in the direction of

the Kāvyādarśa, and to Dr. Arthur Kunst, for his continual

advice and encouragement; to Professors Usha Nilsson, Vernon

x

Chadwick, and Andrew Sihler for their graciousness as

committee members; to Charlie and the staff of MACC for

their invaluable help in overcoming the foibles of computer

text generation; and a special thanks to Dr. Frances Wilson,

not only for her Sanskrit but also for stepping in and

seeing things through at the end.

Throughout this project I was extremely fortunate to

work closely with two of the finest scholars in their

respective fields. In Dharamsala during two extended

periods I received the oral teachings of Barshi Phun-tshog

Dbang-rgyal, the foremost contemporary Tibetan teacher of

poetic and linguistic material. It was an honor to be able

to spend time with him and I feel the weight of a debt now

to his memory that I can only hope to inadequately repay.

This thesis could not have been accomplished without the

invaluable and continual help of J. Prabhakara Shastry, an

Indian pandit thoroughly versed in the Sanskritic tradition.

Both in South India and the States, we worked closely

together developing the translation, discussing the text and

xi

the broad sweep of tradition. He is always generous with

his time and inestimable knowledge, always ready to clarify

yet another confusion. And to his family in Waltair I

express my thanks for their warmth and open hospitality.

In closing I offer my sincere gratitude to my parents

and family here and to my family in India for their

continual support and encouragement in an endeavor that no

doubt at times left one wondering. And surely any thanks

offered to my wife Shakti can only fall far short -- her

labor and strength sustained us over these past years and

her psychic support was unfailing.

xii

The entire threefold world

would become blind darkness

if the light whose name is language

did not shine throughout creation.

Dandin

[ Kāvyādarśa (c. 700)]

Because all objects which we can name or otherwise single out -- the simplest objects of the senses and the most recondite entities that speculation can conjecture, the most abstract constructions of the intellect and the most concrete aims of passion alike -- are projections of man's interests; because the Universe as it is known to us is a fabric whose forms, as we can alone know them, have arisen in and through reflection; and because that reflection, whether made by the intellect in science or by 'the whole soul of man' in poetry, has developed through language -- and, apart from language, can neither be continued nor maintained -- the study of the modes of language becomes, as it attempts to be thorough, the most fundamental and extensive of all inquiries.

I. A. Richards

[Coleridge on Imagination (1934)]

xiii

Textual Context

1

Introduction

The Kāvyādarśa of Dandin is unique, not only in its

seminal position among the long and vital tradition of

extant texts whose central concern is the explication of

kāvya -- the formal and exquisitely refined organization of

language whose focus and end is the generation of beauty --

but also in its method of realization, and the range and

depth of its impact. Dandin belongs among the ever-elect

company of writers accomplished both in the generation and

considered analysis of creative literature. Always grounded

in practice -- with an emphasis on illustration and open-

ended models -- he stands apart from those later writers

within the tradition who sought and affirmed respectively

varying absolute principles.

Indeed I would hold that the Kāvyādarśa is the single

most influential text of the classical Sanskrit tradition --

and perhaps the entire Indian literary tradition -- viewed

in a trans-cultural context -- as measured in range of

2

absorption, and literary and specific commentatorial

response. A striking statement no doubt, but just as this

work will attempt to move inside a specific text in a close

and detailed way, so also will it attempt to step back and

step out of the Sanskritic tradition, and consider the scope

of textual transmission. For the Kāvyādarśa was to become

the "poetic" of choice -- whether in immediate translation

or in direct adaptation -- not only throughout Southern

India and Śrī Laṅkā, but most especially and strikingly in

Tibet. We shall be moving on new ground here, and in

surveying this textual response across time in Tibet it is

to be hoped that a new and deeper awareness of Tibetan

literature and Indic textual influence will be gained.

It is certain that a considered and formal explication

of kāvya predated Daṇḍin -- who may be dated to the latter

7th and early 8th centuries and whose focus of activity was

most probably the southeastern city of Kāñcī -- but apart

from sections of the Nāṭyaśāstra traditionally attributed

to Bharata, it is to the Kāvyādarśa and the Kāvyālaṅkāra of

Bhāmaha that we turn as the earliest extant texts. And

however striking the points of comparison and contention

between these two works, and however much energy and ink

has been squandered in the attempt to establish the

temporal priority of the one over the other, I feel that to

posit resolution either way is questionable at best.

On the tradition that was to follow Daṇḍin and Bhāmaha

the opinion of Louis Renou is just, "Quel riche domaine est

celui de le Poétique sanskrite, tout engagée dans le vif

des spéculations, et sans cesse animée par le contact des

grandes oeuvres littéraires." Yet although he affirms that

"on a identifié aujourd'hui plus de huit cents traités de

poétique en Sanskrit,"1 we should realize that the vital

extant tradition extends to the mid-17th century,

effectively coming to a close with the Rasagaṅgādhara of

Jagannātha, and is primarily expressed by up to twenty-five

central texts

The essential kāvya śāstras (apart from those works

whose focus is drama and theatre) include:

(1) the Kāvyalañkāra of Bhāmaha [7th-8th centuries]

(2) the Kāvyādarśa of Daṇḍin [7th-8th centuries

(3) the Kāvyalañkārasūtrāṇi of Vāmana [8th-9th centuries]

(4) the Kāvyalañkārasārasaṅgraha of Udbhaṭa [8th-9th centuries]

(5) the Kāvyalañkāra of Rudraṭa [9th century]

(6) the Dhvanyāloka of Ānandavardhana [9th century]

(7) the Kāvyamīmāṃsā of Rājaśekhara [10th century]

(8) the Vakroktijīvita of Kuntaka [10th-11th centuries]

(9) the Abhinavabhāratī and Kāvyālokalocana of Abhinavagupta [10th-11th centuries]

(10) the Aucityavicāracarcā and Kavikanṭhābharana of Kṣemendra [11th century]

(11) the Sarasvatīkanṭhābharana and Śṛṅgāraprakāśa of Bhoja [11th century]

(12) the Vyaktiviveka of Mahimabhaṭṭa [11th century]

(13) the Kāvyaprakāśa of Mammaṭa [11th-12th centuries]

(14) the Alamkārasarvasva of Ruyyaka [12th century]

(15) the Kāvyānuśāsana of Hemacandra [12th century]

(16) the Vāgbhaṭālamkāra of Vāgbhaṭa (I.) [12th century]

(17) the Chandrāloka of Jayadeva [13th century (?)]

(18) the Kāvyānuśāsana of Vāgbhaṭa (II.) [14th century (?)]

(19) the Ekāvalī of Vidyādhara [14th century]

(20) the Pratāparudrayasobhūṣaṇa of Vidyānātha [14th century]

(21) the Sahityadarpana of Viśvanātha [14th century]

(22) the Vrttivārttika, Citramīmāṃsā, and Kuvalayānanda of Appayya Dīkṣita [16th century]

(23) the Rasagangādhara of Jagannātha [17th century]

It shall be a fundamental contention of this thesis

that the presentation of Daṇḍin and the Kāvyādarśa as found

throughout the contemporary literature (by which I mean

that written from the latter 19th century forward) --

whether "Western" or Indian -- is frequently marred by

misconception and distortion. I regret the frequent polemic

tone but there is nothing else for it. As we proceed

through Daṇḍin's text, one of our primary concerns then

shall be to cut away accumulated detritus. Despite much

endeavor the explication of kāvya śāstra in contemporary

exposition has been generally but poorly served.

For English readers this is most surely seen in the

lack of adequate translations. Although a few of the

primary texts have been published in English over the last

century, with the possible exception of J. L. Masson and M.

V. Patwardhan's translation of the Rasādhyāya of the

Nāṭyaśāstra,2 these more frequently obscure than clarify.

And I should immediately offer that I feel that my own

position toward translation is hardly highbrow. That is, as

much as I may be sympathetic to and recognize the

criticisms expressed in, for example, Henry Heifetz's

dissertation "Issues of Literary Translation from Sanskrit

and Tamil" (sadly emasculated in publication),3 and

although I would consider a strained or jarring style of

translation in the extreme grounds for rejection (and this

perhaps can only be pointed to in specific example), there

is a wide and grey area into which translation may fall

where dismissal is more a matter of personal distaste.

Rather I would hold to -- insofar as such things may

be grasped -- a more pedestrian line, and ask of a

translation that (1) it be literate (in the literal sense

of the word) in the target language; (2) that its stylistic

presentation is not one of accumulative distortion; and (3)

that it remain within the semantic bounds -- where evident

-- of the original. (It is thus that following my own

restricted sense of the term I would reject, say, the poems

of Ezra Pound's Cathay as "translations" -- but this is not

to simultaneously degrade them, nor does it entail a

failure to recognize that such "transmutations," or call

them what you will, may exist on or reflect (as in this

case) an extremely elevated plane of poetical awareness.)

For to focus on the kāvya śāstra texts the element of

informational or procedural "transfer," however frequently

illuminated by kāvya itself, is central, and the centrality

of this message, however obscure at times at the edges, does

indeed allow itself to be delimited.

A number of the translations relevant to this work fail

at a most fundamental level in their realization of an

English that is as often as not sorely pressed. When this

is combined with occasional distortions of original

"meaning," we have the primary reasons for the continued

obscurity of the study of creative language in classical

Sanskrit.

The English translations of the Kāvyādarśa to date are

cases in point.4 That of V. Narayana Iyer may be rejected

on the first point alone, with such verses as, for example,

"What is called Udara by which all sequence (of words) find

their excellence when the sequence (of word) is uttered its

excellent quality is clear" (KD [1.76]); or "This

decoration of the ear stands in the way of expansion (of

the eye). 'Thus (thinking) probably, by your eye the utpala

flower in your ear is besieged" (KD [2.224]).5

Where that of S. K. Belvalkar, even allowing a

vocabulary that reflects "Indologese" at its best, displays

such a reinforced degree of stylistic distortion that I

feel Dandin's message is severely marred. As in, for

example, "As if chiselled out of the lunar orb, as if

extracted from lotus-interior, is, O slender-bodied one,

thy face. . . ." (KD [2.41]); or "The eyes of the deer have

no dancing eyebrows and are not through liquor tinged red;

this thy pair of eyes however, is adorned with those

qualities" (KD [2.191]).6

A number of Dandin's verses from the Second Chapter

also appear throughout Edwin Gerow's Glossary of Indian

Figures of Speech. And here translation occasionally fails

-- and sadly when this occurs in sufficient number the

remainder, whether justifiably or not, tends to become

suspect -- due to excessive semantic distortion. We shall

touch on many of these verses not merely in correction, but

in view of the various aspects and questions of translation

that may be raised.

The fundamental goal and basis of this work is thus an

adequate translation of Dandin's central, highly technical

yet revealing Second Chapter. The actual practice followed

in the contained translations is based upon a team approach.

Throughout I have worked closely with J. Prabhakara Shastry,

an Indian pandit extremely well-versed in (among other

things) kāvyā and kāvya śāstra, and fluent in English. Each

verse initially would be pulled apart with an emphasis on

resolving questionable word meanings and cultural

references. We would then shift to the sense of the verse

as a whole -- a sense by no means immediately apparent in

every case and which would frequently have to be refined and

drawn out through an extended questioning dialectic. I

would then proceed to an actual translation. The eventual

product would be returned to after a period of time with

Shastry scanning for more obvious errors.

In conjunction with translation an attempt has been

made at detailed explication, for a translation of this

material alone -- however accurate -- at this temporal and

cultural remove would but partially convey the issues

involved. The approach here is radically different from the

usual methodology. For one should be aware that the

standard critical approach to classical Indian literature is

one of broad historical sweep, where textual analysis is

generally reduced to a summation of content.

We may point to, for example, A History of Sanskrit

Literature by A. A. Macdonell (1899); the epical Geschichte

der indischen Litteratur in three volumes by Moriz

Winternitz (1904-20); the sections on literature by Louis

Renou in L'Inde Classique (1953); ; A History of Sanskrit

Literature by A. B. Keith (1928); History of Classical

Sanskrit Literature by M. Krishnamachariar (1937); A History

of Sanskrit Literature: Classical Period by S. N. Dasgupta

and S. K. De (1946); and more recently A. K. Warder's

Indian Kāvya Literature (1972-) (now in five volumes), and

Siegfried Lienhard's A History of Classical Poetry:

Sanskrit-Pali-Prakrit (1984).7

The historical approach is also followed in the

foremost works more specifically devoted to kāvya śāstra --

P. V. Kane's History of Sanskrit Poetics (1923), S. K. De's

History of Sanskrit Poetics (in two volumes) (1923 and

1925), and Edwin Gerow's Indian Poetics (1977). In

exception one perhaps might mention V. Raghavan's extensive

study of Bhoja's Śṛṅgāraprakāśa (1963) (though here too the

emphasis is more comparative across time), and the

exceptional work of Marie-Claude Porcher, as in, for

example, Figures de Style en Sanskrit (1978).8

Our focus shall rather be on a single text -- although

comparative points within the tradition will be drawn, we

will move inside and primarily remain within the Kāvyādarśa

and attempt to examine in detail what Daṇḍin is about.

This considered appreciation of the text itself will allow

us (it is to be hoped) to cut through much of the

misconceived generalities which the historical approach

alone has so frequently offered. I would see two pervasive

fallacies in the literature that this thesis will attempt to

counter.

Among the primary texts of the classical Indian

literary critical tradition, one is struck by the

variability and indeed individuality of approach which

their authors generally display. This is a situation quite

otherwise than the interminable commentaries on

commentaries stemming from a given and absolute textual

authority that one finds in the broader literature as a

whole -- "Contrairement à ce qui se passe dans la plupart

des disciplines indiennes, la Poétique n'a pas eu un texte

de base, un code 'révéle' que tous les ouvrages ultérieurs

se seraient efforcés de suivre ou de commenter. . . . De

là aussi, par voie de conséquence, une liberté de mouvement

qu' on ne retrouve pas au même degré dans les autres

branches du savoir."9

It is this unusual situation that I would posit

underlies one of the most extensive fallacies found in the

contemporary literature: The essentially "revisionist" view

that (1) assumes the quest for an absolute, inherent

principle of kāvya is the valid critical methodology; (2)

that (given this) this was in fact an invariable concern of

all the traditional writers, and thus that they may be

grouped according to various "theoretical" positions or

"schools"; and (3) that this quest was only fully realized

by the later Dhvani theorists (as epitomized by

Ānandavardhana in the Dhvanyāloka [9th century], and

Abhinavagupta in the Kāvyālokalocana [10th-11 centuries]),

whose position is thus conceived as the evaluative standard

by which all other (and especially earlier) writers are to

be judged.

For the lack of an absolute textual authority has

generated a tension or unease among many, a circumstance

seen by many as essentially aberrant, and the need to bring

the kāvya śāstra into equilibrium with the wider spectrum

of Indian literature has been one of the prime motivating

factors leading to the elevation of absolute principles

that we do indeed find among many of the writers

themselves. Yet the force of this continuing circumstance,

for it was by no means resolved within the central

tradition itself, has also led to a seeming need on the

part of many contemporary writers to reinterpret -- in

their consequent evaluative elevation and sweeping

application of such principles and procedures -- the entire

tradition and to project backwards in this light a false

equilibrium.

It is thus that we find what I would consider the

incorrect presentation of the various kāya śāstras (those

explicatory texts whose concern is kāvya or the beautiful

in literature) as a linear progression from initial somewhat

confused stumblings to the complete realization of final

truth (a view that thus serves -- in the minds of those who

accept it -- to solidify and thus satisfactorily ground

what was a quite open situation); and the concomitant

classification of the various writers into "schools"

according to their perceived adherence to particular

principles or features.

Thus we find, for example, Johannes Nobel projecting

upon the exposition of kāvya a conception far more

indicative of late 19th century Europe (of Darwin and the

"Crystal Palace"):

From the beginning to modern times there has been

a steady growth and development. The views of the

older works were rejected or modified, one theory

has substituted for another, and poetry was

regarded from quite different points of view; in

short, there was scarcely one theme that did not

assume a new aspect in the course of the

historical development. Compared with other

branches of human knowledge this progress and

growth was throughout natural and in accordance

with the progressive methods employed in treating

abstract matters.10

Or again Louis Renou, now on the perceived universal

"quête de Unité": "La Poétique, envisagée dans son

développement historique, s'attachera à déterminer un point

central, un principe d'explication permettant de rendre

compte de tous les faits. C'est cette même quête de

l'Unité que nous observons dans l'ensemble des systèmes

philosophiques. Il s'ensuit que chaque école sera tentée

de dévalorise les explications antérieures pour instaurer

un principe nouveau."11

And as Edwin Gerow points outs, "De, Dasgupta, and

Keith, the standard Western or Westernized interpreters of

Indian poetics, prefer to see all poetics addressed to an

abstract genre 'poetry', to the nature of poetry per se,

and to account for the variety of poetics in terms of

varyingly adequate responses to that problem"

(Glossary/71); "and from De, [we seem to get] a sense that

the texts are interesting only insofar as they fulfill some

predefined potential of ideal aesthetic 'progress'."12

Yet Gerow's own views on the issue are hardly clear.

Although seeming to criticize De in the above, he also

writes, "The theory of literature that developed on the

Indian soil . . . is itself exclusively concerned with

purposes and forms of literature, and not at all with its

occasion: it is, in other words, literary philosophy or

æsthetics, rather than criticism."13

Certainly S. K. De, one of the most respected and

prolific writers on kāvya, must yet be approached with

caution. We should recognize his tendency to seek "in every

author notions of poetic essence" (Glossary/43), and we

should be sensitive to statements where presupposition is

presented as evaluative absolute, as in, for example: "The

Indian theorists have almost neglected perhaps the most

important part of their task, viz. a definition of the

nature of the subject of a poem as a product of the mind of

the poet; this problem is the main issue of Western

aesthetics."14 Yet in either tendency he is hardly alone.

Gerow himself, certainly one of the foremost American

scholars in the field, tends I feel to develop logical

constructs that may not necessarily have a basis in the text

under cor:ideration -- that frequently we again find a,

perhaps more subtle, "spin" cast upon the material. And

too his presentation is often marred by a style frequently

opaque, a curious academic "philoso-speak." We find in a

discussion of the "figures" (alaṁkāras), for example, "The

primary characteristic of the figurative universe is not

its fixity, but its selectivity. The figures realize the

potentialities implicit in the norms of grammar and logic

in no set or predetermined archetectonic. . . . Categories

considered as genera in a large number of cases are taken

as basic, especially when these genera appear to remove the

subjacent figures from immediate cross-relevance. . . .

(Glossary/53). (And I fear the words of John Crow Ransom

(in regard to the writings of R. P. Blackmur) perhaps apply,

"I have nearly always seemed to sense an esoteric effect in

his language when he generalizes, which makes him often hard

reading, and I have wondered if it did not cover a very real

philosophical confusion; for, at any rate, philosophical

discourse is explicit, and never esoteric".)15

Indeed, it seems that the majority of writers on the

alaṁkāra theorists frequently tend to hypostasize what are

in fact their own projections -- we should be especially

wary when an author moves away from the relevant text

itself, whether in interpretation or in judgment.

The immediate danger, however, lies in the concrete

distortions that are generated with regard to the

Kāvyādarśa, stemming from the misconceived projection of

"theory" as central to all writers, and its consequent

reification in the form of the various theoretical

"schools."

The "classic" position of this approach is perhaps not

surprisingly presented by S. K. De, "It is probable that

the Rīti school, if we use this term to separate those

writers who put an emphasis on rīti as the most important

element of poetry, had an independent origin and history,

and existed for a long time side by side with the sister

schools, which threw into prominence the elements of rasa,

alaṁkāra or dhvani, respectively"; in which schema Daṇḍin

"stands midway in his view between the Alaṁkāra system of

Bhāmaha and the Riti-system of Vāmana." And further, "At

the same time there can be no doubt that in theory he allies

himself distinctly with the views of Vāmana" (this last an

interesting maneuver considering that Vāmana's views did not

exist at Dandin's time).16

And of course once the legitimacy of the schools was

accepted a conceived scholarly endeavor was to align

oneself according to one's preference. Thus P. V. Kane

mirrors De (albeit with a conclusion that at least hints at

the truth), "Dandin's Kāvyādarśa is to some extent an

exponent of the Rīti School of Poetics and partly of the

Alañkāra school. He gives, however, such an exhaustive

treatment of Guṇa and Alaṅkāras that it is not possible to

identify him with any particular school."17

And reflecting a prevalent practice in the secondary

literature -- the paraphrasing of one's predecessors without

acknowledged attribution --18 D. K. Gupta affirms, "[Dandin]

should be regarded . . . as an alamkāra theorist with the

same force with which he is associated with the rīti school.

In fact, he affiliates himself to both the schools and it

should be clearly understood that he cannot be linked

exclusively with either of the two."19

I would thoroughly reject such views and their

underlying presupposition, and it will thus be one of the

central concerns of this work -- the validity of which I

feel will be self-evident as we examine the text itself --

to demonstrate that Daṇḍin was not concerned with

"theoretical" questions, with "explaining," or with

assuming a given position according to the projected tenets

of a hypostatized school. And far from thus displaying an

envisioned lack of critical awareness, I would posit that

Daṇḍin was very much aware of what he was doing in

consciously developing a presentation that "shows," that

"points to."

For the implications of a circumstance that are

commonly passed by cannot be overly stressed. Daṇḍin

himself is the only writer of a major kāvya śāstra who was

also a major writer of kāvya. A consideration of the best

kāvya work in "prose" (gadya) by H. T. Colebrooke, a

pioneering scholar in the Western tradition of Sanskrit

studies, reflects a traditional given, "The most celebrated

are the Vāsavadattā of Subandhu, the Daśacumāra of Danḍī,

and the Kādamabarī of Bāṇa."20

Danḍin was intimately concerned and convernsant with

the generation of kāvya itself -- in marked contrast to

contemporary scholars -- as well as with the realm of kāvya

śāstra. I am willing to posit (and risk invoking the

"intentional fallacy") that he felt that "The experience of

poetry like any other experience, is only partially

translatable into words"; that "Even the most accomplished

of critics can, in the end, only point to the poetry which

seems to him to be the real thing."21 That Danḍin as a

master of language recognized the limitations of language,

and perhaps recognized the pursuit of ultimate meanings --

themselves expressed in words -- as essentially academic, an

endless web of individual presumption woven by scholars by

and primarily for themselves.

The second pervasive misconception found throughout

the contemporary literature that an accurate reading of the

Kāvyādarśa will dispel is what I term the "prescriptive

fallacy." As with the preceding (and with any number of

minor errors) its fundamental cause is the failure to ground

oneself in the text itself. We seem to have writers

invariably accepting at face value prior summaries of

previous writers, themselves often summaries of prior

summaries, with the original text left unread, lost one is

left to assume in some pre-Cambrian fog.

Again we turn to S. K. De for a model exposition of

this view:

The attempts of these exponents of the Alamkāra

School are limited to a systematic classification

of poetic expression into fixed rhetorical

categories; and from this formal treatment their

works have the general appearance of technical

manuals comprising a collection of definitions,

illustrations and empirical canons elaborated for

the benefit of the aspiring poet. Poetry is

regarded, more or less, as a mechanical series of

verbal devices, in which a desirable sense must

prevail, and which must be diversified by means of

certain tricks of phrasing, which consist of the

so-called poetic figures and to which the name

Alamkāra is restricted.22

The misconception that kāvya śāstra invariably

comprises a collection of "rules" is well-entrenched, and

indeed frequently dropped in passing by some of the most

perceptive of contemporary writers. Henry Heifetz notes,

for example, "The prescriptive rather than evaluative tenor

of Sanskrit formal aesthetics. . . ." (and again, that we

are invariably dealing with "aesthetics").23 Or again, as

Leonard Nathan writes on Daṇḍin's elaboration of the

Mahākāvya (or Sargabandha) in the first chapter of the

Kāvyādarśa [1.14-20], "Indian critics have tried to set

forth the nature and purpose of classical poems and to a

great degree have succeeded, though to our modern way of

thinking their dicta may seem overly dogmatic."24 Where in

fact if he had been reading the actual text -- or a

reasonable translation -- rather than a summary by someone

else, he would have read Daṇḍin's concluding verse to this

sequence [1.20], one verse among many as we shall see, that

explicitly belies this projection of proscribed "dicta": "A

kāvya although short of some of these features is not

necessarily defective / If the excellence of those employed

pleases the wise," that is, the "connoisseur" of literary

excellence.

Surely much of this must be seen as a break down in

scholarly method and rigor, yet even where the need to

approach the text itself is recognized in principle there

remains it seems a glaring failure to actually put this

need into practice with regard to the kāvya śāstras

themselves -- a tradition which thus remains opaque to

modern literary scholarship and criticism as a whole. I

shall close with one of the more striking and recent

examples of this failure, offering an indication of how

serious this problem is.

Gwendolyn Layne in "Orientalists and Literary Critics"

(1982) catalogues the presumptions and failures of Western

and Indian scholars, especially concerning the critical

assessment of Bāṇa's Kādambarī, and cites at length the

practice of repetitive paraphrase (if not plagiarism) from

one author to the next.25 She affirms, "Unfortunately there

are no serious literary critics, nor have there been any in

the recent history of literary criticism, who practice their

craft on Sanskrit Literature."26

With no mention of the śāstra writers themselves we

are left in doubt as to what she might mean by "serious

critics." A doubt soon dispelled however, when she offers

her own approach (as detailed previously in her disserta-

tion (1979)) based on the "Chicago School." For indeed, as

she writes in conclusion, if the "Orientalists" would only

turn to these "real critics," "they in turn would educate

Orientalists in the various methodologies of the discipline

of literary criticism (first rule, read the text itself).

Since a tradition of literary criticism may not have

developed in India, and since the discipline is not known

and practiced by Indologist, such a change in the situation

could only be for the better."27

Yet what is ultimately clear is that her rejection of

the central writers of kāvya śāstra and indeed of the

entire tradition does not stem primarily from an apparent

idiosyncratic conception of what literary criticism might

be, but -- in quoting in approval Nirad C. Chaudhuri (whose

publications on kāvya śāstra are nil) -- from a decline

into the prescriptive fallacy in its most severe form:

"Sanskrit rhetoric and poetics -- Alamkara or Rasa

Sastra as these were called -- were as pretentious

as they were arid. . . . The only service that

these rhetoricians and analysts rendered to

Sanskrit literature was by preserving as

illustrations to their categories some gems of

lyric poetry, which otherwise might have been

lost. Their writings gave Sanskrit literature a

bad name as a collection of mere artificial

prettiness and far-fetched conceits."28

The acceptance of such distortion, and the obvious lack

of familiarity with the Indian critical tradition, indeed

reveals that Layne has not followed her own "first rule."

As we work through our text, we shall rather find that

Dandin's presentation is anything but prescriptive. That

one of the Kāvyādarśa's most distinctive features is a

creative openness, with continual indications that -- in

the case of the alamkāras -- we have guidelines, models

that may provide the basis for yet further development and

variation. That artificial distinctions are to be

rejected; that the ultimate source of poetic validation

lies not in blind adherence to prescribed rules, but in the

acceptance of the "wise," the kavis and refined connoisseurs

themselves. It is well to keep Daṇḍin's conclusion to the

Kāvyādarśa [3.368] firmly in mind: "The Path of alamkāras

is thus displayed / Condensing within limits its endless

expansion / Practice alone can reveal the fine points /

transcending the range of words."

31

We shall develop our understanding of the Kāvyādarśa

then in three sections. The first locates the Central

(Second) Chapter within its immediate Textual Context -- our

introductory remarks are followed by an extensive

consideration, with translation of prominent verses, of

Chapters One and Three. Both touch on a number of issues

which shall be developed in explication, and present a

number of features an awareness of which shall allow us to

approach the central section with a degree of background.

The second section is devoted to the translation and

explication of the Second Chapter. This is the focused

heart of Dandin’s text, what I choose to term a "calculus"

of creative expression, and which reflects his contribution

-- in elaboration of varieties and illustration -- to the

fullest. And in the third section we shall trace the

immense impact of the Kāvyādarśa as textual model not only

throughout South and Central India and Śrī Laṅkā, but - for

the first time in textual detail -- into Tibet.

The Text

The text itself of the Kāvyādarśa is quite well-established, with numerous extant manuscripts available.29

The published editions and translations of the Kāvyādarśa include the following:

[1862] The Kāvyādarśa of Sir Dandin. Edited by Pandit Premachandra Tarkabagisa, with his own commentary entitled Mālinyapronchani. Fasc. 1 and 2. Calcutta: Asiatic Society of Bengal, 1862.

[1863] The Kāvyādarśa of Śrī Dandin. Edited with a commentary entitled Mālinyaproñcanī by Premachandra Tarkavāgīśa. Bibliotheca Indica, vol. 40, New Series nos. 30, 33, 38, 39, 41. Calcutta: Asiatic Society of Bengal, 1863. Reprint. Kāvyādarśah Śrī Dandyācāryaviracitah

Śrī Premachandra Tarkavāgīśa Bhaṭṭācārya viracita Mālinyapronchanī nāmaka ṭīka sahitah Śrī Bhavadeva

Caṭṭopādhyā yena samṣkrtaḥ Calcutta: New School Press,

1881. Reprint. Osnabrück: Biblio Verlag, 1981.

[1874] Kāvyādarśa. Śrī-Daṇḍy-ācārya-viracitah. Śrī

Jīvānanda-Vidyāsāgara-Bhaṭṭācārya-kṛta-vivṛtisametaḥ

Calcutta: Sarasvati Press, 1874.

[1882] Kāvyādarśa. Edited by Jīvānanda Vidyāsāgara

Bhaṭṭācārya, with his own commentary known as the Jīvānanda

Vidyāsāgara ṭīkā. Calcutta, 1882. 2nd edition. Calcutta,

1890. 4th edition. Calcutta, 1925.

[1890] Daṇḍin's Poetik (Kāvyādarśa). Sanskrit text

with German translation by Otto Böhtlingk. Leipzig: Verlag

von H. Haessel, 1890.

[1909] Kāvyādarśa (Chapter 2, verses 14–96 only). In

Otto Böhtlingk's Sanskrit-Chrestomathie. Edited by Richard

Garbe. 1909.

[1910] The Kāvyādarśa of Daṇḍin, with the Commentary of

Taruṇavācaspatī, and also with an anonymous incomplete

Commentary known as Hṛdayaṅgamā. Edited by M. Rangacharya.

Madras: Brahmavādin Press, 1910.

[1919] Daṇḍin's Kāvyādarśa (Chapters 1 and 4).

Edited with translation and Notes by S. Subrahmanya Sastry.

Allahabad: National Press, 1919.

[1919] Daṇḍin's Kāvyādarśa: Pariccheda I. Edited with a

new Sanskrit Commentary and Notes by S. K. Belvalkar and

Rangacharya B. Raddi. Bombay: The Department of Public

Instruction, 1919.

[1920] Daṇḍin's Kāvyādarśa: Pariccheda II. Edited with

a new Sanskrit Commentary and English Notes by S. K.

Belvalkar and Rangacharya B. Raddi. Bombay: The Department

of Public Instruction, 1920.

[1920] Kāvyādarśa. English translation of Chapters 1

and 2 by S. K. Belvalkar. Poona, 1920.

[1921] Dandin's Kāvyādarśa (Chapter 1). Translated

literally into English with full explanatory and critical

notes by P. N. Patankar. Indore: City Press, 1921.

[1924] Kāvyādarśa of Dandin. Edited with Sanskrit text

and English translation by S. K. Belvalkar. Poona: The

Oriental Book-Supplying Agency, 1924.

[1925] Kāvyādarśa, with the contemporary commentary

entitled Kusumāpratimā. Edited by Nṛsimhadeva Śāstrī.

Lahore: Mehrchand Lakshmandas, 1925. 2nd edition. Lahore,

[1929] Kāvyādarśa (First Pariccheda), with Commentary

by Pandit R. V. Krishnamachariar. Kumbako Nam: Komalamba

Press, 1929.

[1930] Kāvyādarśa, with the commentary of Vādijaṅghāla.

Edited with English translation and Notes by V.

Krishnamachariar and V. Hanumanthachar. Madras: Educational

Publishing Co., 1930.

and Tarunavācaspatī, and a contemporary commentary entitled

Mārjanā. Edited by V. Krishnamachari. Tiruvadi: Srinivasa

Press, 1936.

[1938] Kāvyādarśa of Dandin. Edited by Vidyābhūsana

Pandit Rangacharya Raddi Shastri, with his own commentary

entitled Prabhā. Poona: Bhandarkar Oriental Research

Institute, 1938. 2nd edition. Poona, 1970.

[1941] Kāvyādarśa, with the anonymous Hrdayañgamā

commentary and the commentaries of Vādijaṅghāla and

Tarunavācaspatī. Edited by D. T. Tatacharya. Bombay,

[1942] Kāvyādarśa of Dandin. Edited by S. Viswanathan.

English notes and translation of the first parichchheda and

of the second parichchheda up to the end of the rūpakachakra

by C. Sankara Rama Sastri. Madras: Sri Balamanorama Press,

1942. 2nd edition. Madras, 1959. 3rd edition. Madras,

[1952] Kāvyādarśa, with the commentary of Jīvānanda

Vidyāsāgara. Edited with English translation by V.

Narayana Iyer. Madras: Ramaswamy Sastrulu, 1952. Reprint:

Madras, 1964.

[1957] Kāvyalaksana of Dandin (also known as Kāvyā

darśa), with the Commentary entitled Ratnaśrī by

Ratnaśrījñāna. Edited by Anantalal Thakur and Upendra Jhā.

Darbhanga: Mithila Institute of Post Graduate Studies and

Research in Sanskrit Learning, 1957.

[1958] Kāvyādarśa. Sanskrit and Hindi texts, with the

commentary entitled Prakāśa in Sanskrit and Hindi by

Ramchandra Mishra. Varanasi: Chowkhamba Vidyābhavan, 1958.

2nd edition. Varanasi, 1972.

[1961] Kāvyādarśa, with the commentary entitled

Tarkavāgīśa [?] by Premachandra. 2nd edition. Calcutta: K.

Ray, 1961.

[1973] Kāvyādarśa. Sanskrit text with Hindi paraphrase

and commentary entitled Sudarśana by Dharmendra Kumāra

Gupta. Delhi: Mehrcand Lachmandas, 1973.

We shall be following the Sanskrit of and translating

Rangacharya Raddi's second edition (1970) of the Kāvyādarśa,

and occasionally referring to his Sanskrit commentary

entitled "Prabhā."30 The text appears in three paricchedas

or "chapters," which is the norm, with 105 verses to the

first, 368 to the second, and 187 to the third. Some

editions are found however, with four chapters, a

reflection of the division of the usual third chapter with

the final section on the doṣas or potential "faults" in

kāvya now distinct.

The first chapter appears stable, yet the number of

verses in the second and third chapters occasionally varys.

Variation in the second chapter stems from either the

retention or deletion of any of three verses. Two of these

appear in our text at [2.155-56] as varieties of ākṣepa

alaṃkāra. Although Raddi considers these interpolations --

which I too feel is very much the case -- and marks them as

such, he retains them. This appears to be a standard

practice (whether the editor is aware of their questionable

nature or not), and given that the primary reference works

(those of P. V. Kane, S. K. De, D. K. Gupta, for example)

all mirror Raddi's numbering, to avoid confusion in cross

referentiation I have followed suit. In a truly "critical"

edition these two verse would most probably be dropped.

The third variable verse appears in our edition as [2.362],

and again it is quite possible that we have an

interpolation. The question of interpolation will be

discussed under the respective verses.

Thus for example, Raddi's edition and the Calcutta

edition of Premachandra (1863) include all three of these

verses, although marking them as uncertain, and thus give a

total of 368 verses in the second chapter. In the edition

of the text reconstructed from, and including, Ratnaśrī's

commentary (1957), and in all of the Tibetan editions

(which are closely related to the former) none of these

verse appear and thus the second chapter in each displays a

total of 365 verses.

The third chapter appears stable, although when it is

divided variation may occur. Thus in the edition of M.

Rangacharya (1910), which is in four chapters, two

additional verses appear at the end of what is now the third

chapter, and two are added to the fourth chapter, one at

the beginning and one in the middle.

The second commentary on and (reconstructed) text of

the Kāvyādarśa to which we shall occasionally refer is that

of the Ceylonese Buddhist monk Ratnaśrijñāna [c. 900].31 S.

K. De notes, "The author was a Ceylonese monk who wrote

under the patronage of a Rāṣṭrakuṭa king, named Tuṅga,

under the overlordship of Rājyapāla of Gauḍa and Magadha

(c. 908 a.d.) [the commentary itself states that it was

written in the 23rd regnal year of a Rājyapāla]. Authors

quoted, besides Aśvaghoṣa and Kālidāsa, are Mātrceṭa,

Āryaśūra, Kohala, Rāmaśarman, Medhāvirudra, Kambala,

Harivrddha, Bhāmaha, Bhartṛmeṇṭha, Gunādhya, Mallanāga, and

Dharmakīrti."32

The editors of the published edition, Anantalal Thakur

and Upendra Jha, have reconstructed the text from Ratnaśrī's

commentary, itself based upon a single palm leaf manuscript

(with verses [1.1-3] and [3.50-56] missing). They believe

the title of Daṇḍin's work to be "Kāvyalakṣaṇa" based on a

misreading of verse [1.2], a view which is unwarranted and

otherwise unsubstantiated.

Ratnaśrī's work is of extreme interest, not only

because it is most probably the earliest extant commentary

[10th century], but also for its intimate role in the

Tibetan transmission and interpretation of the Kāvyādarśa.

We shall discuss this in our final section, but we should

note now that it is highly probable that Ratnaśrī studied

and taught at one (or more) of the northern Buddhist

monasteries so central for the transmission of Buddhism

beyond India, and that the text he was following is

extremely similar to the Tibetan versions. This text then

may very possibly closely reflect the version of the

Kāvyādarśa which was first brought into Tibet, or it may be

that it was brought in somewhat later and utilized in the

revisions of the initial Tibetan translation. It is

certain, however, that Ratnaśrī's commentary was utilized by

the Tibetans at a very early date. The editors note and

affirm

that Ratnaśrī generally agrees with the Tibetan

version of the Kāvyalaksana [the editors

attributed title of Dandin's work]. As our author

hails from Ceylon where . . . Dandin's text was

highly popular, it is normally expected that the

author should follow the southern text of the

Kāvyalaksana. But the commentary shows that he

was influenced by the culture of Magadha [in the

North], and the text of Dandin's work as found

there at the time [10th century] was acceptable to

him. . . . That the work of Dandin was popular in

Magadha and adjoining regions is proved by

quotations from it even in the philosophical works

of Vācaspati Miśra. The Tibetan text is also

based on manuscripts from the monasteries of

Magadha [an assertion for which unfortunately no

concrete evidence is provided].33

43

Abbreviations and Editions Cited

The following texts are of central importance to our

study and are the editions cited within the narrative by the

corresponding abbreviations where marked. The initial

edition listed is the one cited unless otherwise noted (a

following edition is one regularly consulted as well).

Primary Explicative Texts Cited

(RŚ/ ) Ratnaśrī [10th century]

Kāvyalaksana of Dandin (also known as Kāvyādarśa), with

the Commentary entitled Ratnaśrī by Ratnaśrījñāna.

Edited by Anantalal Thakur and Upendra Jhā. Darbhanga:

Mithila Institute of Post Graduate Studies Research in

Sanskrit Learning, 1957.

(RR/ ) Rangacharya Raddi [20th century]

Kāvyādarśa of Dandin. Edited by Vidyābhūsana Pandit

Rangacharya Raddi Shastri, with his own commentary

entitled Prabhā. Poona: Bhandarkar Oriental Research

Institute, 1938. 2nd edition. Poona, 1970.

(Notes 1 / )

Dandin's Kāvyādarśa: Pariccheda I. Edited with a new Sanskrit Commentary and Notes by S. K. Belvalkar and Rangacharya B. Raddi. Bombay: The Department of Public Instruction, 1919.

(Notes 2 / )

Dandin's Kāvyādarśa: Pariccheda II. Edited with a new Sanskrit Commentary and Notes by S. K. Belvalkar and Rangacharya B. Raddi. Bombay: The Department of Public Instruction, 1920.

(Böhtlingk7 )

Dandin's Poetik (Kāvyādarśa). Sanskrit text with German translation by Otto Böhtlingk. Leipzig: Verlag von H. Haessel, 1890.

(Glossary )

Edwin Gerow. A Glossary of Indian Figures of Speech. The Hague: Mouton, 1971.

Primary Texts Cited

NŚ Bharata [2nd-3rd centuries(?) ]

The Nātyaśāstra ascribed to Bharata-Muni, vol. 1 (Chapters 1-27). Edited by Manomohan Ghosh. Calcutta: Manisha Granthalaya, 1967.

Nāṭyaśāstra of Bharatamuni, with the Commentary

Abhinavabhārati by Abhinavaguptācārya. Edited by M.

Ramakrishna Kavi, 2nd rev. edition by K. S. Ramaswami

Shastri, vol. 1. (Chapters 1-7), vol. 2 (Chapters

8-18). Baroda: Oriental Institute, 1956.

Bhaṭṭi [6th-7th centuries]

The Bhaṭṭi-kāvya of Bhaṭṭi, with the Commentary

(Jayamaṅgalā) of Jayamaṅgalā. Edited by Vināyak

Nārāyan Shāstri Joshi and Srīnivāsa Venkatrāma Śarmā.

Bombay: Nirnaya Sāgar Press, 1887. 5th edition.

Bombay, 1914.

KA Bhāmaha [7th-8th centuries]

Kāvyālaṅkāra of Bhāmahā. Edited with English

translation by P. V. Naganatha Sastry, 2nd edition.

Delhi: Motilal Banarsidass, 1970 (1927).

Kāvyālaṅkāra, with the Udyāna Vṛtti. Edited by D. T.

Tatacharya. Tiruvadi, 1934.

KAS Vāmana [8th-9th centuries]

Kāvyālaṅkārasūtrāṇi, with the Kāvyālaṅkāradhenu

Sanskrit commentary by Gopendrā Tripurahara Bhūpāla.

Edited with Hindi translation by Bechana Jhā. Varanasi:

Chowkhamba Sanskrit Series Office, 1971.

KASS Udbhata [8th-9th centuries]

Kāvyālaṅkārasārasaṅgraha of Udbhata, with the

Laghuṿṛtti commentary of Indurāja. Edited with

introduction and notes by Naryana Daso Bhatti. 2nd

edition.

Poona: Bhandarkar Oriental Research

Institute, 1982.

KA Rudrata [9th century]

Kāvyālaṅkāra (A Treatise on Rhetoric) of Rudrata, with

the Commentary of Namisādhu. Edited with the Prakāśa

Hindi Commentary by Rāmadeva Śukla. Varanasi: The

Chowkhamba Vidyabhavan, 1966.

Ānandavardhana [9th century]

Dhvanyāloka of Ānandavardhana. Critically edited

Sanskrit text, with revised English translation by K.

Krishnamoorthy. Dharwar: Karnatak University, 1974

(1955).

The Dhvanyāloka of Śrī Ānandavardhanāchārya, with the

Lochana and Bālapriyā Commentaries by Śrī Abhinavagupta

and Panditrāja Sahrdayatilaka Śrī Rāmaśāraka. Edited

by Pandit Paṭṭābhirāma Śāstri. Benares: Chowkhambā

Sanskrit Series Office, 1940.

Agni Purāṇa (Alamkāra Section) [c. 900 (?)]

Agni Purāṇa. A Collection of Hindu Mythology and

Traditions. Edited by Rajendralala Mitra. Vol. 3

(Chapters 269-382). Calcutta: Bibliotheca Indica, 1879.

Reprint. Osnabrück: Biblio Verlag, 1985.

47

SKA

Bhoja [11th century]

Sarasvatīkanthābharanālañkārah Edited by Viśvanātha

Bhaṭṭācāryah Vol. 1. Varanasi: Banaras Hindu

University, 1979.

ŚP Bhoja

Śṛṅgāraprakāśah Edited by G. R. Josyer. Vol. 2

(Chapters 9-14). Mysore: Coronation Press, 1963.

KP Mammaṭa [11th-12 centuries]

The Poetic Light: Kāvyaprakāśa of Mammaṭa. Edited with

an English translation by R. C. Dwivedi. Vol. 1.

Delhi: Motilal Banarsidass, 1967.

AS Ruyyaka [12th century]

The Alamkārasarvasvam of Rājānaka Ruyyaka. With the

commentary of Jayaratha. Edited by Mahāmahopādhyāya

Paṇḍita Durgāprasād and Paṇḍuraṅga Parab.

Kāśināthasarma. Bombay: Nirnaya Sagara Press, 1893.

Reprint. Delhi: Bharatiya Vidya Prakashan, 1982.

Notes: Introduction

1. Louis Renou, "La Réflexion sur la Poésie dans L'Inde," in Sanskrit et Culture: L'Apport de l'Inde a la Civilisation Humaine (Paris: Payot, 1950), p. 143.

2. J. L. Masson and M. V. Patwardhan, Aesthetic Rapture: The Rasādhyāya of the Nāṭyaśāstra, 2 vols. (Poona: Deccan College, 1970).

3. Henry S. Heifetz, "Issues of Literary Translation from Sanskrit and Tamil," Ph.D. dissertation, University of California, Berkeley, 1983. Heifetz rejects most of the translations to date due to what he perceives as their lack of sensitivity to the rhythm and sound features of the original Sanskrit, but also primarily in their display of a style which he terms "Indologese":

The absence of poetic acumen among partisans of 'the tradition' leads to their assumption that an elevated tone is obtained through stiff academic diction and grammar, copious Latinisms and archaisms, an affection for Victorian and Renaissance inversion, and perhaps such (incorrect) cosmetic issues as an avoidance of contractions. What results is a sort of sentimentalism of elevation, a ragbag of cliches vaguely associated with higher social strata in America or England now or back through the past few hundred years. A genuinely elevated tone in writing is not obtained through superficial decoration but by the over-all management of diction, rhythm, and placement (p. 191).

4. Apart from the translations of V. Narayana Iyer and S.

K. Belvalkar, which themselves are quite rare, three additional yet unavailable prior translations of the Kāvyādarśa, in whole or in part, may be cited:

(1) Dandin’s Kāvyādarśa (Chapter 1 and 2), edited with translation and notes by S. Subrahmanya Sastry (Allahabad: National Press, 1919).

(2) Dandin’s Kāvyādarśa (Chapter 1), translated literally into English with full explanatory and critical notes by P. N. Patankar (Indore: City Press, 1921).

(3) Kāvyādarśa, edited with English translation and notes by V. Krishnamachariar and V. Hanumanthachar, with the commentary of Vādijaṅghāla (Madras: Educational Publishing Co., 1930.

5. Dandin, Kāvyādarśa, edited with English translation by V. Narayan Iyer, with the commentary of Jīvānanda Vidyāsāgra (Madras: Ramaswamy Sastrulu, 1952), Reprint (Madras, 1964), p. 41 and p. 146.

6. Dandin, Kāvyādarśa of Dandin, edited with Sanskrit text and English translation by S. K. Belvalkar (Poona: The Oriental Book-Supplying Agency, 1924), p. 16 and p. 31.

7. A. A. Macdonell, A History of Sanskrit Literature, Reprint (Delhi: Motilal Banarsidass, 1976 (1899)); Moriz Winternitz, Geschichte der indischen Litteratur, 3 vois. (Leipzig, 1904-20) (for English translation see the bibliography); Louis Renou in Louis Renou and Jean Filliozat, L’Inde Classique, vol. 2 (Paris: Imprimerie Nationale, 1953); A. B. Keith, A History of Sanskrit Literature (London: Oxford University Press, 1928); M. Krishnamachariar, History of Classical Sanskrit Literature, 2nd edition, Reprint (Delhi: Motilal Banarsidass, 1974 (1937)); S. N. Dasgupta and S. K. De, A History of Sanskrit Literature: Classical Period, 2nd edition (Calcutta:

University of Calcutta, 1975 (1946)); A. K. Warder, Indian Kāvya Literature, vols. 1-5 (Delhi:Motilal Banarsidass, 1972- ); and Siegfried Lienhard, A History of Classical Poetry: Sanskrit-Pali-Prakrit (Wiesbaden: Otto Harrassowitz, 1984).

8. P. V. Kane, History of Sanskrit Poetics, 3rd edition, Reprint (Delhi: Motilal Banarsidass, 1971 (1923)); S. K. De, History of Sanskrit Poetics, 2nd edition, two vols. in one, Reprint (Calcutta: Firma KLM Private, 1976 (1923 and 1925); Edwin Gerow, Indian Poetics (Wiesbaden: Otto Harrassowitz, 1977); V. Raghavan, Bhoja's Śṛṅgāraprakāśa, 3rd rev. edition (Madras: V. Raghavan, 1978 (1963)); Marie-Claude Porcher, Figures de Style en Sanskrit: Théories des Alamkāraśāstra Analyse de Poèmes de Veṅkatādhvrin (Paris: Institut de Civilisation Indienne, 1978).

9. Louis Renou, "La Réflexion sur la Poésie dans L'Inde Ancienne," in Sanskrit et Culture (Paris: Payot, 1950), p. 137.

10. Johannes Nobel, The Foundations of Indian Poetry and their Historical Development (Calcutta: R. N. Seal, 1925), p. 9.

11. Louis Renou, "La Reflexion sur la Poésie dan L'Inde," pp. 138-39.

12. Edwin Gerow, Indian Poetics (Wiesbaden: Otto Harrassowitz, 1977), p. 218, n. 2.

13. Edwin, Gerow, Indian Poetics, p. 218.

14. S. K. De in Kuntaka, The Vakrokti-Jīvita, edited by S. K. De, 3rd rev. edition (Calcutta: K. L. Mukhopadhyay, 1961), p. xix, n. 19.

15. John Crow Ransom, "Ubiquitous Moralists," The Kenyon Review, 3 (1941), pp. 96-97.

16. S. K. De, History of Sanskrit Poetics, vol.2, Reprint (1976), p. 75 and p. 76.

17. P. V. Kane, History of Sanskrit Poetics, pp. 89-90.

18. For a critical review of this practice see Gwendolyn L. Layne, "Kādambarī: A Critical Inquiry into a Seventh-Century Sanskrit Narrative," 2 vols., Ph.D. dissertation, University of Chicago, Chicago, 1979; and also "Orientalists and Literary Critics: East is East, and West is West, and it is in the Professional Interest of Some to Keep it that Way," The Western Humanities Review, vol. 36, n. 2 (Summer, 1982), pp. 165-75.

19. D. K. Gupta, A Critical Study of Dandin and hisWorks (Delhi: Meharchand Lachmandas, 1970), p. 184.

20. H. T. Colebrooke, "On Sanskrit and Prākrit Poetry" (1808), in Miscellaneous Essays (1827), Reprint, Essays on History, Literature, and Religions of India, vol. 2 (New Delhi: Cosmo Publications, 1977), p. 134.

For the kāvya of Dandin see Avantisundarīkāthā [and Avantisundarīkathāsāra], edited by M. Ramakrishna Kavi and S. K. Rāmanātha Sastri (Madras, 1924); Avanti Sundarī of Acharya Dandin, edited by K. S. Mahadeva Sastri (Trivandrum: Suranand Kunjan Pillai, 1954); The Daśa Kumāra Charita; or The Adventures of the Ten Princes: A Series of Tales in the Original Sanskrit, edited with introduction by Horace H. Wilson (London: Society for the Publication of Oriental Texts, 1846); Reprint, "Introduction to the Daśa Kumāra Charita," in Essays Analytical, Critical and Philological on Subjects Connected with Sanskrit Literature, vol 1, collected and edited by Reinhold Rost (New Delhi: Asian Educational Services, 1984, 342-79); Dasakumaracharita of

Dandin, revised in one volume by Ganesh Janardan Agashe from

the first edition of Buhler and Peterson in two parts, 2nd

ed., Bombay Sanskrit and Prakrit Series, 10 and 42 (Bombay:

Government Central Press, 1919); The Daśakumāracarita of

Dandin, translated with introduction by M. R. Kale, 3rd ed.,

(Bombay, 1925); Reprint, (Delhi: Motilal Banarsidass, 1966);

Daśakumāracaritam: Pūrvapiṭhikā, Sanskrit text with English

translation, Introduction and annotation by C. Sankara Rama

Sastri, edited by S. Viswanatham (Madras, 1944); Reprint

(Delhi: Motilal Banarsidass, 1978); Dandin's

Daśakumāracaritam: Die Abenteur der zehn Prinzen, Zum ersten

Male aus dem Sanskrit ins Deutsche übersetzt von Johann

Jakob Meyer (Leipzig: Verlag, 1902); Dandin's

Dasha-Kumara-Charita: The Ten Princes, translated by Arthur

W. Ryder (Chicago: The University of Chicago Press, 1927);

Die Erlebnisse der zehn Prinzen; eine Erzählung Dandins,

translated by Walter Ruben (Berlin: Akademie-Verlag, 1952).

21 T. S. Eliot, "Introduction" to The Use of Poetry andthe

Use of Criticism (London: Faber and Faber, 1980), pp. 17-18.

22. S. K. De, "The Problem of Poetic Expression" (1947),

in Some Problems of Sanskrit Poetics, Reprint (Calcutta:

Firma KLM, 1981 (1959)), pp. 12-13.

23. Henry Heifetz, "Issues of Literary Translation from

Sanskrit and Tamil," Ph.D. dissertation, University of

California, Berkeley, 1983, p. 184.

24. Leonard Nathan, The Transport of Love: The Meghadūta of

Kālidāsa (Berkeley: University of California Press, 1976),

p. 9.

25. Gwendolyn Layne, "Orientalists and Literary Critics,"

Western Humanities Review, vol. 36, n. 2 (Summer, 1982),

pp. 165-75.26. Gwendolyn Layne, "Orientalists and Literary

Critics," p. 168.

27. Gwendolyn Layne, "Orientalists and Literary Critics, "

pp. 174-75.

28. Gwendolyn Layne, "Orientalists and Literary Critics,"

p. 175, n. 21; quoting Nirad C. Chaudhuri, Hinduism:

AReligicn to Live By (New York: Oxford University Press,

1979), p. 218.

29. For a listing of manuscripts of the Kāvyādarśa

available in India see the New Catalogus Catalogorum,

edited by V. Raghavan and K. Kunjunni Raja, vol. 4 (Madras:

University of Madras, 1968), p. 108.

30. Dandin, Kāvyādarśa of Dandin, edited by Pandit

Rangacharya Raddi Shastri, with his own commentary entitled

Prabhā , 2nd edition (Poona: Bhandarkar Oriental Research

Institute, 1970 (1938).

31. Kāvyalakṣana of Dandin (also known as Kāvyādarśa),

with the Commentary entitled Ratnaśrī by Ratnaśrījñāna,

edited by Anantalal Thakur and Upendra Jha (Darbhanga:

Mithila Institute of Post Graduate Studies and Research in

Sanskrit Learning, 1957).

32. S. K. De, History of Sanskrit Poetics, vol. 1, pp.

71-72.

33. Kāvyalakṣana of Dandin, edited by Anantalal Thakur and

Upendra Jha, (1957), p. 17.

Chapter One

Homage to Sarasvatī -- Goddess of Kavis (Poets)

"May all-white Sarasvatī / -- a hamsī among clusters of

lotuses -- the faces of Caturmukha / Forever play in the

Mānasa lake of my mind" [ caturmukhamukhāmbhojavanahamsa-

vadhūrmama | mānase ramataṃ nityaṃ sarvaśuklā sarasvatī || ]

[1.1]

sarasvatī : Goddess of poets and writers, of speech

and music, wife of "Caturmukha," (the "Four-faced One,"

that is, Brahmā). Sarasvatī appears in Vedic times as the

female personification of a powerful northwestern river

(since dried). In Ṛg Veda [2.41.16] she is praised as "Best

mother, best of rivers, best of goddesses."1 From this

early association with a clean and life-giving river, she

drew the attributes of purity and procreation. Her eventual

role as muse, as source and bestower of artistic creativity,

would appear to be a logical extension.

54

The wise king saw before his eyes the goddess

Sarasvatī herself, and saluted her with bowed head

and folded hands. This conqueror of enemies

praised her with reverent words and fell like a

log to the ground, saying, 'I have come to you for

help! I worship the great chaste goddess who is

before me, the divinity of speech, who is without

beginning or end. I praise the womb of the world,

the excellent Yoginī, the supreme spouse of the

Golden Embryo, the three-eyed moon- topped

goddess! I honor her who knows supreme bliss, a

portion of the highest consciousness, the

embodiment of Brahman. Protect me, supreme

goddess, who has come to you for refuge!'" 2

caturmukha : The "Four-faced One," Brahmā the Creator,

who with Viṣṇu the Preserver and Śiva the Destroyer forms

the "trimūrti," the essential triad of forces of later

Hinduism. Without the sectarian support offered to Viṣṇu

and Śiva, the later mythic characteristics of Brahmā reflect

a biased manipulation. According to the Śaivites, "Brahmā

originally had five heads, each one appearing as he turned

to gaze at his newly created daughter-wife, Sarasvatī. The

fifth head was destroyed by Śiva (in some accounts cut off

by a swipe of Śiva's left thumb nail), who was once annoyed

with Brahmā for being presumptuous enough to deny that

deity's superiority; or according to another legend, because

he had violated Pārvatī, the wife of Śiva."3

Dandin's Design and the Necessity of Kāvya

"Synthesizing earlier śastras and examining their

practices / We shall present the distinctive character of

kāvya to the best of our ability" [ pūrvāśātrāṇi saṃhṛtya

prayogānupalakṣya ca | yathāsāmarthyamasmābhiḥ kriyate

kāvyalakṣaṇam | |] [1.2].

lakṣaṇa / In this case, "characteristic or distinctive

attributes." In varying versions of Vātsyāyana's

Nyāyasūtrabhāṣya [prior to 400 a.d.4], we find two

definitions of lakṣaṇam : (1) "lakṣaṇa is the property that

distinguishes the essence of something specifically

designated" / uddiṣṭasya tattvavyavacche-dako dharmaḥ

lakṣaṇam ; and (2) "lakṣaṇa is the property that

distinguishes something specifically designated from other

things" / uddiṣṭasyātattvavya-vacchedako dharmaḥ lakṣaṇam.

"One is inclined to translate lakṣaṇa as 'characteristic

trait' rather than as 'definition'. . . . Rather than

indicating an exhaustive description of the object to be

defined, it focuses on that property that belongs to that

object and to no other."5

A lakṣaṇa as "definition" must be free of three

errors: (1) ativyāpti / a referential range that is too

great, thus including characteristics of things other than

that object which one wishes to define; (2) avyāpti / a

referential range that is too small, thus excluding

elements that should be included in the scope of the

definition; and (3) asambhava / a definition that is

impossible. "A correct definition [lakṣaṇa] is negatively

defined as one which is free from any of these three faults,

and more positively by Vātsyāyana [Nyāyasūtrabhāṣya [1.1.2]]

as an attribute which differentiates what is defined from

all things other than itself."6

"By the grace of languages alone -- those grammatically

analysed by the authorities and the rest -- the way of the

world proceeds" [iha śistānuśistānām śistānāmapī sarvathā |

vācāmeva prasādena lokayātrā pravartate ||] [1.3].7

"The entire threefold world would become blind

darkness if the light whose name is language did not shine

throughout creation" [ idamandhamtamah kṛtsnam jāyeta

bhuvanatrayam | yadi śabdāhvayam jyotirāsamsārānna dīpyate

||] [1.4].

"The image of fame of earlier kings reflected in the

mirror of literature / See! It does not perish even in

their absence" [ ādirāja-yaśobimbamādarśam prāpya vāñmayam |

teṣāmasamnidhānepi na svayam paśya naśyati ||] [1.5].

vañmayam / as "literature" in the narrower (and more

usual) sense, that is, "creative written works (in whatever

presented or received form)."

Rājaśekhara writing at a much later date [9th-10th

centuries], in the beginning of the second chapter of the

Kāvyamīmāṃsā understands vañmaya in a somewhat wider sense:

"Vāñmayam [comprehends] simultaneously both śāstra and kāvya. Since the śāstras precede in time, one should apply oneself to the śāstras before the kāvyas" / iha hi vāñmayamubhayathā śastram kāvyam ca | śastrapūrvakatvāt kāvyānām pūrvam śastreṣvabhiniviśeta.

Siegfried Lienhard also appears to accept this wider view. "Poetry is of course only one part of all the writing comprehended in the Sanskrit word vāñmaya, which is used in some texts that deal with literature [?] and other writings to include everything that is expressed in words."

And in the Agnipurāṇa [327.1], we find vāñmaya divided into four "linguistic" components: dhvani/"sound"; varṇa/"letter," "phoneme"; pada/"word"; and vākya/"sentence." Yet in the immediately following verse [327.2], we have vāñmaya reflecting three primary genres, śāstra, itihāsa, and kāvya.

It is perhaps with this in mind that one of our commentators, Ratnaśrī, chose the following analysis: "' vāg' in this case refers to itihāsas, kathās, and so on"/vāgiha itihāsa kathādilakṣaṇā (RŚ/4).

60

I feel, however, that Dandin in this case is

indicating a very special capacity of a very special kind

of writing. Our other commentator, Rangacharya Raddi, is

closer to the mark: "Vāñmaya is kāvya created through the

imagination (pratibhā) of excellent kavis" / vāñmayam

satkavipratibhäprasūtam kāvyam (RR/4).

That kāvya should extol the "fame of kings," keeping

it ever alive, reflects the activity of but one side of a

common, fruitful, symbiotic relationship. The kings of

nobility on their part frequently provided the supportive

environment that allowed the kavis to focus on their

creative task.

To the king of India we unquestionably owe most of

the poets of repute; patronage by the king was at

once the reward of skill in panegyric and the

means of obtaining the leisure for serious

composition and a measure of publicity for the

works produced. It was the duty of the king to

bridge the gulf between wealth and poetic talent,

of the poet to save his patron from the night of

oblivion which else must assuredly settle on him

when his mortal life closed. 10

"It is said by the wise that language properly

employed is a wish-yielding cow / Poorly employed it merely

conveys the ox-headedness of the user" [ gaurgauh kāmadughā

samyak prayuktā smaryate budhaiḥ i duṣprayuktā punargotvaṃ

prayoktuḥ saiva śaṃsati ||][1.6].

"Therefore a flaw in kāvya however slight should not be

neglected -- A body however beautiful would become ugly

through a single blemish" [ tadalpamapi nopeksyaṃ kāvye

duṣṭaṃ kathamcana | syādvapuḥ sundaramapi svitrenaikena

durbhagam || ] [1.7].

"How could one ignorant of the [kāvya] śāstras

distinguish between the guṇas [the "qualities"] and doṣas

[the "faults"] ? Is there discrimination for one blind

between the perceptions of various colors?" [ guṇadoṣāṇa-

śāstrajñāḥ katham vibhajate janaḥ | kimandhasyādhikārosti

rūpabhedopalabdhiṣu || ] [1.8].

62

The Tradition and Possible Predecessors

"Therefore the learned -- with an eye towards the

education of kavis -- have formulated the method of

composing kāvyas in the various mārgas" [atah prajānām

vyutpattimabhisamdhāya sūrayah | vācaṃ vicitramārgāṇaṃ

nibabandhuḥ kriyāvidhim ||] [1.9].

śāstras: In this verse and the preceding Daṇḍin is

referring to alaṃkāra or kāvya śāstras, that is, texts which

present a formal explication and analysis of kāvya.

This explicit confirmation, and that of the preceding

[1.2], of prior kāvya śāstras is clearly of some importance.

There are certainly sufficient indications throughout the

Kāvyādarśa (mention of "previous authorities," "others," and

so on), and in Bhāmaha's Kāvyālaṅkāra, to justify the belief

in a number of earlier studies devoted to kāvya. That

Daṇḍin himself drew from these earlier authors and their

works is also explicitly stated. At the beginning of the

Second Chapter he writes, "The basis of these postulations /

was demonstrated by earlier teachers" [2.2]; and following

his list of the artha alaṁkāras, "Thus the alamkāras of

kāvya / described by earlier teachers" [2.7].

As translated by Hari Chand, the anonymous Hrdayamgamā

commentary glosses [1.2] of the Kāvyādarśa as: "J'ai réuni

et examiné àfond les définitions données par les anciens

maitres, Kāśyapa, Vararuci, etc.; j'ai bien observé les

applications chez Kālidāsa, etc.; c'est ainsi que j'ai

composé, dans la mesure de mes moyens et de mes facultés, ma

propre définition de la poésie" / pūrveṣāṁ kaśyapavararuci-

prabhṛtīnām ācāryāṇām lakṣaṇa-śāstrāṇi saṁhṛtya paryālocya

kālidāsaprabhṛtīnām prayogān upalakṣya ca yathāsāmarthyam

buddhyanurūpam asmābhiḥ kāvyalakṣaṇaṁ kriyate (we note that

Hari Chand chose to translate "lakṣaṇa" as "definition,"

rather than presume that it indicates the title of the

text).11 And Vādijaṅghāla in his Śrutānupālinī commentary

again mentions Kāśyapa, as well as Brahmadatta and

Nandisvāmin as predecessors.

Kāśyapa is unknown except for the rare item. Pānini

cites a Kāśyapa under Astādhyāyī [8.4.67]. In later works,

the Sinhalese Siyabaslakara (Svabhāṣālaṅkāra) of the mid-8th

(or possibly the 12th) century, a derivative of the

Kāvyādarśa (see under "Sinhalese" within the Transmission

section ), "begins with homage to Brahma, Śakra, Bṛhaspati,

the saint Kāśyapa, the excellent Vāmana [or Bhāmaha] Daṇḍin

and other masters."12 The Pañcasāyaka of Nānyadeva in

[4.19] mentions Kāśyapa as an authority on erotics, where

the Agnipurāṇa considers him an authority on metrics.

Abhinavagupta notes that he was a sage that preceded

Bharata. And Kallinātha in his commentary on the

Saṅgītaratnākara of Śārṅgadeva [first half of the 13th

century] under [2.2.31], quotes three verses attributed to

Kāśyapa.13

Vararuci, the other "early master" that the Hrdayamgamā

mentions, is dated to the time of a Nanda king [4th century

B.C.] (predecessors of Candragupta Maurya) by later versions

of Guṇadhya's Bṛhatkathā in their (legendary) presentation

of the origins of Sanskrit grammar.14 Aśvaghoṣa in the

Sūtrālamkāra assigns him to the same period and further

cites six verses of Vararuci's addressed to this Nanda

king.15 And Patañjali in the Mahābhāṣya under Pāṇini

[4.3.101] mentions a Vārarucakāvya.

As we mentioned above, Daṇḍin at the beginning of

Chapter Two [2.2] again expresses his debt to predecessors

in the field: "The basis of these postulations / was

demonstrated by earlier teachers" [ kimtu bījaṃ vikalpanāṃ

pūrvācāryaiḥ pra-darśitam ]. Our commentators under this

verse merely mention, "Medhāvi, Śyāmava, and so on" (RŚ/67);

"Bharata and so on" (RR/112).

Bhāmaha specifically mentions a "Medhāvin" as a

predecessor. After listing the seven defects possible in

upamā alamkāra [2.39], he notes in [2.40ab], "These seven

faults have been mentioned by Medhāvin" [ ta eta upamā-doṣāḥ

sapta medhāvinoditāḥ| ]. And in [2.88cd] he writes, "In some

places utprekṣā is called 'samkhyāna' by Medhāvin

[ saṃkhyānamiti medhāvinot-prekṣābhiḥitā kvacit || ]. Yet as

66

P. V. Kane points out, Dandin notes in Kāvyādarśa [2.273]

that samkhyāna is rather another name for for yathāsamkhya

alamkāra (as is krama), which leads him to affirm,

"Therefore the passage in Bhāmaha's work seems to be

corrupt. If we read medhāvī notprekṣā etc. then there is a

correspondence with Dandin's words, the meaning being

'Medhāvin calls yathāsamkhya by the name samkhyāna and in

some places (in some works on alaṅkāra) utprekṣā has not

been spoken of as an Ālaṅkāra'. "16

And as with Dandin, Bhāmaha collectively and

impersonally refers a number of times to earlier (or

perhaps contemporaneous) writers: "others"/apare ([1.31],

[2.6], [4.6]); anyaih ([2.4]); anye ([3.4], [4.12];

"some"/kaiścit ([2.37]); kecit ([2.93]).

A personal note is added by Rājaśekhara [9th-10th

centuries] in his Kāvyamīmāṃsā. He indicates that

Medhāvirudra (the presumed long form of the name) was blind

from birth: "Yet for those who possess pratibhā ["creative

illumination"] even those without sight, [the caravan of

words and ideas] is as though vividly present. It is thus

that one hears of kavis blind from birth, Medhāvirudra,

Kumāradāsa, and so on."17

On Kumāradāsa, who was from Śrī Lañkā, Louis Renou

notes, "A poet well known, author of the Jānakīharana (5th -

6th centuries ?). Tradition considers him a contemporary of

Kālidāsa. In the last verse of his poem he alludes to a

sickness that he had contracted as a child."18

Namisādhu [mid-11th century] in his tippana

("commentary") on Rudraṭa's Kāvyālaṅkāra mentions a

Medhāvirudra under [1.2], along with Daṇḍin and Bhāmaha /

nanu daṇḍimedhāvirudra bhāmahādikṛtāni santyevalamkāra-

śāstrāṇi.19 And under [2.2] he notes that "Medhāvirudra and

so on" consider "words" (śabda) to have only four categories

/ eta eva catvāraḥ śabdavidhā iti . . . tairmedhāvirudra-

prabhṛtibhiḥ karmapravacanīyā noktā bhaveyuḥ | .20

Yet following [11.24] we again have the name as cited

in Bhāmaha. It is thus probable that the full form of the

name is "Medhāvirudra."21 Here Namisādhu mentions "Medhāvi

and others" in connection with seven defects possible in

upamā alamkāra (as does Bhāmaha above). P. V. Kane feels

that "the manner in which he deals with this topic suggests

that the examples he gives are taken from Medhāvin's work."

Yet, having indicated that five of the seven illustrative

verses that Namisādhu cites are found in Bhāmaha (as [2.40,

47, 55, 63]), his conclusion is questionable, "if the verses

were Bhāmaha's he would have probably so stated. Therefore,

Bhāmaha should be taken as quoting five verses from

Medhāvin."22 Namisādhu does not just mention Medhāvin

alone, but "Medhāvin and others" / atra ca svarūpopādāne

satyapi catvāra iti grahanādyanmedhāviprabhṛtibhiruktam

yathā . . . iti saptopamādoṣāḥ. |.23 We may conjecture that

these examples may have been drawn from a work prior to

Bhāmaha, but as to whose we have no assurance.

In Bhāmaha's Kāvyālaṅkāra there are further references

to possible earlier writers. In [2.19] yamakas ("sound

repetitions") and prahelikās ("riddles") are mentioned as

occurring in the Acyutottara, attributed to a "Rāmaśarma."24

69

And again, in [2.58] an example of the defect in upamā known

as adhikatva (unbalanced parallelism between upameya and

upamāna) is attributed to Rāmaśarma. In the same section

[2.47], the example which illustrates the defect asambhava

(an improbable comparison) is attributed to a Śakhāvardhana.

The Rājamitra appears apparently as a text in [2.45], and

again in [3.10] with an example of samāhita alaṃkāra drawn

therefrom.25

Of further works where there is some indication of

earlier writers of kāvya śāstra we might add that Pāṇini in

Aṣṭādhyāyī [4.3.11] refers to a Nāṭasūtra of Kṛśāśvin, and

in [4.3.110] to a Nāṭasūtra of Śilālin.26 We have

previously considered kāvya's mythological origins as

presented by Rājaśekhara.27 Yet he also subsequently

provides a mythological list of the various originators --

"etres célestes versés dans la science poétique"28 -- of

various elements in kāvya, and as such we may consider them

as Daṇḍin's "mythological" predecessors. Substantively,

"this entire enumeration reflects the content of 'ancient' poetics prior to dhvani. . . ."29

Thus Sahasrākṣa has transmitted the secret doctrine of the kavi (kavirahasya); Uktigarbha the poetic locutions (auktika < ukti, "a poetical term covering the ensemble of poetical figures"30); Suvarnanābha the r̥tis ("styles");

Pracetas31 that which concerns ānuprāsika (< anuprāsa); Citrāñgada the yamakas and citras ["vaiegated" poetical expressions presenting a verbal puzzle or pattern, whether of meaning or sound]; Śeṣa the śabda śleṣas [where a single verbal string may be variously analysed syllabically, yielding different words]; Pulastya the vāstavās [alam̐kāras based on things as they are; one of the four categories of artha alam̐kāras presented by Rudraṭa in the Kāvyālañkāra [7.9ff.]]; Aupakāyana those alam̐kāras based upon similarity (aupamya); Pārāśara those based upon atiśaya [poetical exaggeration or "intensity"]; Utathya the artha śleṣas [where a single word yields more than one meaning]; Kubera the ubhaya alam̐kāras [those displaying the manipulation of

both sound and sense]; Kāmadeva the poetical diversions

(vainodika) ("Its is here a useful convenience marking the

distinction between the body of knowledge taught (the

collection of vinoda, which may correspond to the krīda of

Kāmasūtra [1.4.42]) and its instructor, 'the God of Love'

(Kāmadeva and also the King Kādamba patron of poets and the

author of a verse anthology";32 Bharata who described the

rūpakas [here "plays"]; Nanikeśvara who promulgated the

rasas [the eight or nine purified emotive nodes in kāvya];

Dhiṣaṇa who presented the doṣas ("defects") possible in

kāvya; Upamanyu who presented the guṇas ("qualities")' and

Kucamāra who taught the esoterica (aupanisadika) ("without

doubt this is based upon the model of the Kāmasūtra . . .

which closes with a section on the occult"33 -- Thus each of

these has composed their respective individual texts.34

It should go without saying that a kavi of Daṇḍin's

skill would be well-versed not only in prior śāstras on

kāvya, but in prior kāvyas as well. Daṇḍin himself

provides an extremely valuable survey of authors with whom

he was familiar. At the opening of his extended kāvya in

gādya ("prosaic") form, the Avantisundarī, we find twenty-

seven (somewhat fragmentary) verses in praise of kāvis that

have come before.35

After offering homage to Sarasvatī, Vālmiki, Vyāsa, and

kavis in general, Daṇḍin devotes himself to specific

writers. Thus we find: (1) Subandhu (verse 6), associated

as minister with King Bindusāra (3rd century b.c.), the son

of the great Candragupta Maurya (an earlier Subandhu than

the later author of the mahākāvya Vāsavadattā). Patañjali

mentions the Vāsavadattā of this Subandhu as an example of

an ākhyāyikā (a kāvya in gādya or prose form). (2) Guṇāḍhya

(verse 7), author of the lost Bṛhatkathā, "a work which

ranked beside the Mahābhārata and the Rāmāyaṇa as one of the

great storehouses of Indian literary art,"36 and which

Daṇḍin explicitly mentions again in [1.38]. (3) Mūladeva

(verse 8), the "personification of all trickery," the

masterful rogue to whom the Corasūtras (treatises on theft)

are attributed. (4) Śūdraka [4th century a.d. (?)] (verse

9), reputed to have been a king of Ujjain and the author of

the plays Mṛcchakaṭikā and Padmaprabhṛtaka. ((5) The name

is missing.) (6) Bhāsa (verse 11) [c. 300 a.d.], reputed

author of a variety of plays, including those in one act

such as the Madhyamavyāyogā, the Dūtaghatotkaca, the

Karnabhāra, and the Ūrubhaṅga; the Pañcarātra in three acts;

and the longer Bālacarita, Avimāraka, Pratijñāyaugandharā-

yana, Svapnavāsavadattā, and the fragmentary Carudatta.37

(7) King Sarvasena [1st half of the 4th century] (verse 12),

author of the lost Prākṛta mahākāvya, the Harivijaya. (8)

Pravarasena [5th century] (verse 13), author of the

Setubandha ("The Building of the Bridge"). also known as the

Rāvanavaha ("The Killing of Rāvana"), in Māhārāṣṭrī Prākṛta

(and mentioned by Daṇḍin in Kāvyādarśa [1.34]; ((9) The name

is missing.) (10) Kālidāsa [4th-5th centuries] (verse 15).

(11) Mentions only one "afflicted in the eye" (verse 16).

This may possibly refer to the Sinhalese kavi Kumāradāsa,

author of the Jānakīharaṇa, who was born blind (see

above).38

(12) Nārāyaṇa (verse 17), possibly Bhaṭṭa

Nārāyana, the author of the play Venīsamhāra -- "peut-être

contemporain des drames de Bhavabhūti (on a même présumé le

7e siècle). . . ."39

(13) An incomplete verse praising a

"cakravartin of kavis," which "may in all possibly point to

the celebrated Bhāravi," author of the Kirātārjunīya.40

(14) And finally Bāṇa [7th century], with Daṇḍin and

Subandhu, justly considered the "third great master of

Sanskrit prose" -- author of the mahakāvya Harṣacarita, the

incomplete Kādambarī, and presumably the Caṇḍīśataka, a

stotras in 102 stanzas praising Caṇḍī (Umā/Pārvati, Śiva's

consort); and Mayūra [7th century], with Bāṇa also reputed

to have written under the patronage of King Harṣavardhana

[606-47 a.d.], and author of the Sūryaśataka, a stotra

praising the sun, and the shorter Mayūrāṣṭaka in eight

stanzas.41

Notes [1.1] - [1.9]

1. Margaret and James Stutley, Harper's Dictionary of Hinduism (New York: Harper and Row, 1977), p. 270.

2. Kūrma Purāṇa [1.23.13-27], in Classical Hindu Mythology, edited and translated by Cornelia Dimmitt and J. A. B. van Buitenen (Philadelphia: Temple University Press, 1978), p. 241.

3. Benjamin Walker, Hindu World, vol. 1, (London: George Allen and Unwin, 1968), p. 165.

4. A. B. Keith, Indian Logic and Atomism (Oxford: The Clarendon Press, 1921), p. 28.

5. Madeleine Biardeau, "La Définition dans la Pensée Indienne," Journal Asiatique, 245 (1957), p. 372.

6. A. B. Keith, Indian Logic and Atomism, (1921), p. 154. See also Gotama, The Nyāya Sūtras of Gotama, translated by Satiśa Chandra Vidhyābhūṣana (Allahabad, 1913) Reprint (New York: AMS Press, 1974); Gautama, The Nyaya-Darshana: The Sūtras of Gautama and Bhāṣya of Vātsyāyana, with the Khadyota and Bhāsyachandra commentaries, edited by Gangānātha Jha and Dhundhirāja Shastri Nyāyopādhyāya (Benares: Chowkhamba Sanskrit Series Office, 1925); Gautama, Nyāya Philosophy: Literal Translation of Gautama's Nyāya-Sūtra and Vātsyāyana's Bhāṣya by Debiprasad Chattopadhyaya and Mrinalkanti Gangopadhyaya, Part 1: First Adhyāya (Calcutta: Indian Studies Past and Present, 1967).

7. Belvalkar and Raddi provide an excellent example of the all too common distorted and quite naive approach to "primitive" languages (and we should note that there is no implication of a hierarchy in Daṇḍin's verse) : "The interpretation which we prefer, and according to which the

śiṣṭānusiṣṭa forms of language include Sanskrit, the Prākṛts

and in fact all forms that have reached the grammar-stage;

the remaining forms (śiṣṭānām) comprising all the dialects

of the vulgar or the ignorant or the uncivilized people

that are not general or consistent or advanced enough to

demand a grammatical treatment. In fact even the most

primitive and uncivilized man needs some kind of language

howsoever crude and unpolished" (Notes 1/4).

8. Rājaśekhara, Kāvyamīmāṃsā of Rājaśekhara, edited by C.

D. Dalal and R. A. Sastry, revised and enlarged by K. S.

Ramaswami Sastri Siromani, 3rd edition (Baroda: Oriental

Institute, 1934), p. 2.

9. Siegfried Lienhard, A History of Classical Poetry:

Sanskrit-Pali-Prakrit, (1984), p. 1.

10. A. B. Keith, A History of Sanskrit Literature, (1920,

pp. 52–53.

11. Hari Chand, Kālidāsa et L'Art Poétique de L'Inde

(Paris: Libraire Ancienne Honoré Champion, 1917),, p. 62.

12. Hari Chand, Kālidāsa, p. 62.

13. Cited in S. K. De, History of Sanskrit Poetics, vol. 1,

pp. 67–68.

14. See: Somadeva Bhaṭṭa, Kathāsaritsāgarah (Delhi:

Motilal Banarsidass, 1970), Chapter 1: The Kathā sarīt

sāgara or Ocean of the Streams of Story, translated by C. H.

Tawney, 2nd edition (Delhi: Munshiram Manoharlal, 1968).

Kṣemendra, The Brihatkathāmañjari of Kshemenḍra, edited

by Mahāmahopādhyāya Pandit Śivadatta and Kāśināth Pāṇḍurang

Parab, 2nd edition (Bombay: Nirnaya Sagar Press, 1931),

Chapter 1.

15. Aśvaghoṣa, Sūtrālaṃkāra: Traduit en Francais sur la

version chinoise de Kumārajīva par Edouard Huber. Paris:

Ernest Leroux, 1908, p. 88.

16. P. V. Kane, History of Sanskrit Poetics, p. 63.

17. Rājaşekhara, Kāvyamīmāmsā of Rājaşekhara, edited by C.

D. Dalal and Pandit R. A. Sastry, third edition (Baroda:

Oriental Institute, 1934), Chapter 4, p. 12: pratibhāvataṫ

punarapażyato 'pi pratyakṣa iva | yato medhāvirudrakumāra-

dāsādayo jātyandhāḥ kavayaḥ şrūyante |.

18. (Rājaşekhara, La Kāvyamīmāmsā de Rajashekhara,

translated by Nadine Stchoupak and Louis Renou (Paris:

Imprimerie Nationale, 1946), p. 58, n. 23).

See: Kumāradāsa, The Jānakāharaṇa of Kumāradāsa,

edited by S. Paranavitana and C. E. Godakumbura (Colombo:

Sri Lanka Sahitya Mandalaya, 1967).

19. Rudrata, Kāvyālaṇkāra (A Treatise on Rhetoric) of

Rudrata, with the commentary of Namisādhu, edited with the

Prakāşa Hindi commentary by Rāmadeva Śukla (Varanasi: The

Chowkhamba Vidyabhavan, 1966), under [1.2] p. 3.

20. Rudrata, Kāvyālaṇkāra, (1966), under [2.2], p. 20.

21. See: P. V. Kane, History of Sanskrit Poetics, p. 63.

22. P. V. Kane, History of Sanskrit Poetics, p. 64.

23. Rudrata, Kāvyālaṇkāra [11.24], (1966), p. 361.

24. S. K. De accepts the Acyutottara as a work of Rāmaşarma

(History of Sanskrit Poetics, vol. 2, p. 84,

n. 18).

25. P. V. Nāganātha Śāstry writes, "Śākhāvardana and his

two works." This attribution is not evident from the

Kāvyālaṇkāra itself, and I have been unable to find any

substantiation (Bhāmaha, Kāvyālaṅkāra, translated by P. V.

Nāganātha Śāstry, second edition (Delhi: Motilal

Banarsidass, 1970), p. xvii).

26. Pāṇini, The Aṣṭādhyāyī of Pāṇini, edited and translated

by Śrīśa Chandra Vasu, vol. 1 (1891); Reprint (Delhi:

Motilal Banarsidass, 1962), p. 789.

27. Rājaśekhara, Kāvyamīmāṃsā of Rājaśekhara, (1934), p. 2.

28. Rājaśekhara, La Kāvyamīmamsā de Rājaśekhara, (1946),

p. 23.

29. Rājaśekhara, La Kāvyamīmamsā de Rājaśekhara, (1946),

p. 26, n. 52.

30. Rājaśekhara, La Kāvyamīmamsā de Rājaśekhara, (1946),

p. 23, n. 22.

31. Accepting the reading of Renou rather than pracetāyana

in Rājaśekhara, La Kāvyamīmamsā de Rājaśekhara, (1946), p.

23, n. 25.

32. Rājaśekhara, La Kāvyamīmamsā de Rājaśekhara, (1946),

p. 25, n. 41.

On the vinodas in kāvya ^@śāstra see: Bhoja's

Sarasvatīkaṇṭhābharaṇa following [5.93].

33. Renou later translates the related word "upaniṣad" as

"cause mystérieuse de la poésie." Rājaśekhara, La

Kāvyamīmamsā de Rājaśekhara, (1946), p. 26, n. 51 and pp.

61–62.

34. Paraphrasing the French translation of Nadine

Stchoupal and Louis Renou in La Kāvyamīmāṃsā Rājaśekhara,

(1946), pp. 23–26.

Rājaśekhara, Kāvyamīmāṃsā of Rājaśekhara, (1934), p. 1:

tatra kavirahasyam sahasrākṣah samāmnāsīt

auktikamuktigarbhaḥ rītinirnayaṃ suvarnanābhaḥ ānuprāsikaṃ pracetāyanaḥ yamakāni citraṃ citrāṅgadaḥ śabdaślesaṃ śeṣaḥ vāstavam pulastyah aupamyamaupakāyanaḥ atiśayaṃ pārāśaraḥ arthaślesamutathyāḥ ubhayālaṅkārikaṃ kuberāḥ vainodikaṃ kāmadevaḥ rūpakanirūpanīyaṃ bharataḥ rasādhikārikaṃ nandikeśvaraḥ guṇaupādānikamupamanyuḥ aupaniṣadikaṃ kucamāraḥ iti tataṣte pṛthak pṛthak svaśāstrāṇi viracayāñcakruḥ ‖

35. Daṇḍin, Avantisundarī, edited by K. S. Mahādeva Śāstri (Trivandrum, 1954), pp. i–3.

For a discussion on these previous kavis see: M. Ramakrishna Kavi, "Proceedings and Transactions of the Second Oriental Conference," Calcutta, 1922, pp. 193–201.

36. A. B. Keith, History of Sanskrit Literature (London: Oxford University Press, 1920), p. 266.

37. That Bhāsa was the author of all of these plays is open to doubt. See: Louis Renou and Jean Filliozat, L'Inde Classique, vol. 2, (1953), pp. 265–70; and A. B. Keith, The Sanskrit Drama (London: Oxford University Press, 1924), pp. 91–26.

38. M. Ramakrishna Kavi, Avantisundarī-Kathā of Daṇḍin, p. 199. See: A. B. Keith, History of Sanskrit Literature, (1920), pp. 119–24.

39. Louis Renou, L'Inde Classique, vol. 2, (Paris: Imprimerie Nationale, 1953), pp. 286–87. See: A. B. Keith, The Sanskrit Drama, pp. 212–19.

40. M. R. Kavi, Avantisundarī-Kathā of Daṇḍin, p. 199. Kavi incorrectly considers Bhāravi to be Daṇḍin's grandfather; based upon the Avantisundarikathā this would be Damodhara. Bhāravi was rather a friend of Damodhara.

41. See: Mayūra, The Sanskrit Poems of Mayūra, edited

with the text and translation of Bāṇa's Candīśataka by G. P.

Quakenbos (New York: Columbia University Press, 1917).

C. P. Quakenbos, "The Mayūrāṣṭaka, An Unedited

Sanskrit Poem by Mayūra," Journal of the American Oriental

Society, 1911, pp. 343-54.

Mayūra, The Sūryaśataka of Mayūra, with the commentary

of Tribhuvanapāla, edited by Pt. Durgāprasād and K. P.

Parab (Bombay, 1889), Reprint (Bombay: Nirṇaya Sāgar Press,

81

The Nature of Kāvya and its Formulaic Division / On Metre

and the Versaic Form

"And by these [earlier writers] the body and ornaments

of kāvyas are shown / Basically the body is a succession of

words distinguished with desired meaning" [ taiḥ śarīraṃ ca

kāvyānāṃ alaṃkāraśca darśitaḥ | śarīraṃ tāvadiṣṭārthavya-

vacchinnā padāvalī ||] [1.10].

"It is classified in a threefold way: Whether

stanzaic, prosaic, or their mixture / The stanzaic consists

of four padas and is either in vṛtta or jāti" [ padyam

gadyam ca miśram ca tat tridhaiva vyavasthitam | padyam

catuṣpadī tatra vṛttaṃ jātiriti dvidhā ||] [1.11].

The latter half of Daṇḍin's verse touches on metre or

"chandas," a wide-ranging organizational element that

extends far beyond kāvyas as such -- "Sanskrit literature is

chiefly in verse. The poems and plays, the histories and

legends, treatises on law, divinity, astronomy,

mathematicks, and indeed nearly all literature being in

metre."1 And keeping in mind the important proviso that

"metrical form in poetry is not merely a matter of fixed

accentual, quantitative or syllabic patterns but involves

the whole issue of how rhythm is articulated in the units of

poetry for the communication of meaning and feeling,"2 we

may procede to a brief examination.3

A padya ("stanza") consists of four pādas ("quarters").

In a vṛtta padya the metre is determined by the number and

position of the syllables (akṣaras) in each pada, and their

"weight" -- whether "light"/laghu, where the vowel is short

and not followed by more than one consonant; or "heavy"/

guru, where the vowel is long, or if short it is followed by

two or more consonants (whether in another word or not), or

if the syllable contains either an anusvāra or visarga.

Three fundamental classes of vṛtta padyas are

distinguished: samavṛtta, where the pādas are all similar;

ardhasamavṛtta, where alternate pādas are similar; and

viṣamavṛtta, where the pādas are all dissimilar. The number

of syllables in a given pāda may (theoretically) vary from

one to twenty-six, and each may be (generally) either heavy

or light.

It should go without saying that the actual number of

syllables found in practice is far less than potentially

possible. H. D. Velankar in examining the "prosodial

practice" of twenty-eight prominent kavis, found that the

anuṣṭubh (eight syllables to the pāda), the upajāti (eleven

syllables to the pāda), and the vamśastha (twelve syllables

to the pāda) were the most commonly utilized in continued

narration within the sections of a stanzic mahākāvya.4

A yati or specified "break" should occur between words

or members of a long compound at the end of each pāda, with

the break at the half-padya somewhat stronger. In longer

metres, breaks may occur at fixed positions within the pādas

themselves. For example, in the ubiquitous anuṣṭubh (or

śloka) metre -- "as well as being frequently used in

Classical poetry, it is the staple metre of Sanskrit epic

and of the many didactic works composed in verse"5 -- with

its eight syllables to the pāda, the fifth syllable should

be light, the sixth heavy, the seventh alternately heavy and

light, and the eighth either heavy or light.

In a jāti padya or stanza the metre is determined by

"quantity," that is, by the number of "sound instants" or

mātrās in a given pāda. The duration of a short vowel is

equivalent to one mātrā, that of a long vowel to two mātrās

(a mātrā is not equivalent to the classical Greek mora as is

occasionally affirmed: a mora only refers to a short

syllable, where morae refers to a long syllable). The most

prevalent variety of the jāti category is the āryā metre

with nine subvarieties (Renou posits sixteen varieties in

Sanskrit, twenty-seven in Prākṛt6). Thus for example, the

variety of the jāti āryā category itself termed āryā should

display twelve mātrās in the first and third pādas, eighteen

in the second, and fifteen in the fourth.7

Jacobi's assertion that "Metrical compositions were

originally designed to be sung,"8 is perhaps true if we

only consider the most basic and accessible metres,

especially those of the jāti āryā type which "primitivement

sans doute était chanté."9 Yet the motivating force behind

the growth in complexity and length found in the classical

Sanskrit metres is most probably to to be found in the

coordinate and increasingly refined development of kāvya.

That "This richness and elaboration of metre, in striking

contrast to the comparative freedom of Vedic and epic

literature, must certainly have arisen from poetical use; it

cannot have been invented for grammatical memorial verses

[or for song], for which a simple metre might better

suffice."10

"Its [the stanzaic form's] complete exposition is

revealed in the Chandoviciti -- This branch of knowledge is

a ship for those wishing to enter the ocean of kāvya"

[ chandovicitvāṃ sakalastatprapańco nidarśitaḥ | sā vidyā

naurvivikṣūṇāṃ gambhīraṃ kāvyasāgaram ||] [1.12].

Although some writers believe that the "Chandoviciti"

refers to yet another work of Dandin's11 -- now on metre --

it is far more probable, as P. V. Kane believes, "that the

work Chandoviciti means simply chandas-śastra . . . and is

generally taken as referring to the Vedānga on metrics

ascribed to Piṅgala."12 And as the commentator Taruṇavā

caspati glosses this verse, "chandaḥprapaṅcaśchandovicityāṃ

piṅgalanāgena darśitaḥ paryālocanīyaḥ |".13

However we should note that a text with this name (as

its first and second verses indicate) was found somewhat

recently in Central Asia,14 and which apparently drew from

the earlier works of Yāska (one of the early masters of

metrics that Piṅgala mentions),15 and from Piṅgala himself:

[1d] yās(ka) piṅga[l] (asūtr)[e]bhyah / [2.cd] ______ kṛtiḥ

| candovici[t]iḥ ______ ||.

Enough of the text remains for the editor Dieter

Schlingloff to conclude that Halāyudha [10th century], who

illustrated with examples and commented upon Piṅgala's

sūtras, knew of and utilized this text. And that, contrary

to Albrecht Weber's earlier opinion that Halāyudha's

examples were either his own or taken from well-known

poetical works,16 it now appears that a large part of his

material was drawn from this text.17 And in noting a number

of contrasts with Bharata's Nātyaśāstra he writes, "Diese

Unterschiede machen es unwahrscheinlich die Quelle unseres

Textes oder unser Text die des Nātyaśāstra ist."18

Yet there is really no basis for John Brough, in

reviewing Schlingloff's edition, to affirm, "Until now

nothing has been known of it [the Chandoviciti] except the

name; but the name has long been familiar as that of a wo-k

on metrics mentioned in the Kāvyaḍarśa of Daṇḍin. There is

no reason to doubt that this is the text to which Daṇḍin

referred."19 The ambiguity remains.

"A detailed description of such stanzaic forms as:

Muktaka, Kulaka, Kośa, and Samghāta will not be presented

-- These are all categories subsumed by the Sargabandha"

[ muktakaṃ kulakaṃ kośaḥ saṃghāta iti tadrṣaḥ |

sargabandhāṃśarūpatvādānuktaḥ padyavistarāḥ ||] [1.13].

muktaka [ < *muc /"release," "set free" ] : a single

padya "detached" yet self-sufficient. In the Agni Purāna

[337.36cd] we read: "The muktaka is but a single stanza yet

capable of generating poetic beauty among the wise"

[ muktakaṃ sloka ekaikaścamatkārakṣamaḥ satām ||.]

The muktaka compresses the "maximum of poetic message

into a limited space," utilizing to the full the syntactical

and semantic compression that Sanskrit allows. They may

capture the essence of a scene, a moment, or develop

multiple layers of meaning that yet interact, semantically

expanding outward. "The single stanza of muktaka poetry is

without context and, as it is a complete poem, an

artistically rounded whole, each part of it shows far

greater elaboration than is found in epic stanzas."20

The element of "detachment" reflects the view of the

Sargabandha or Mahākāvya, the "great" extended Kāvya, as

all-embracing. Lienhard would see this as essentially a

late development. "This clearly does not reflect the old

designation of short poems; it is rather a product of the

later, mistaken conception of a short poem as being really

only a stanza "freed" from its context and of single-stanza

poetry on the whole as being a secondary form derived from

long poems. . . . "21

I am not really sure, however, that one can so easily

dismiss (or even apply the question of truth or falsity) to

an attitude held throughout the kāvya tradition as a

"mistaken conception." It is not so much that the muktaka

and the other briefer forms are considered "secondary," as

it is that the mahākāvya is "considéré tacitement par la

théorie indienne comme le kāvya par excellence."22

Lienhard continues:

In reality, exactly the reverse is true. Long

before the rise of kāvya, a category of

single-stanza poetry, muktaka [What of the "old

designation"?] reached maturity and held a key

position right from the very beginning in Old and

Middle Indian litera-ture. . . . Not only did

short poetry influence other genres to a degree

that has hardly been realized so far, it also

became a living part of the way of life and

outlook of those classes who handed down the

traditions of poetry, a highly cultivated, largely

urban society. Moreover, the finest works of the

classical poets have been written in the poetic

miniature painting that is the muktaka genre.

90

An older, authentic and factually correct name for independent stanzas is gāthā [as in Theragā-thā/Therīgāthā].23

I have included Lienhard's statement (as part of the most recent history of kāvya) because it touches on an

important issue, and yet contains a number of fallacies that one should be aware of. That kāvya may have developed

from the single stanza (that this may be disputed we have noted), and that it appears in these various stanzaic forms

hardly invalidates the recognition in the developed tradition that the mahākāvya in its sweep and scope, in its

totality that is more than the sum of its stanzaic parts, is the ultimate test and expression of the kavi. This leads us

to presumption, and its frequent ensuing companion,

projection. One cannot presume to speak of "mistaken conceptions" in contexts that have no relevant bearing: the

fact that the single stanza kāvya came to be known as "muktaka" reflects an actuality of development within the

tradition. By the very creation of the longer forms the

entire relational position of all other forms changed; they

were seen, and quite naturally, in a different light. One

cannot presume that a word carries its full etymological

weight in an evaluative sense, the focus here is primarily

structural as we shall see.

Whenever one reads such signposts over the unknown as

"in reality" or "long before the rise of kāvya," one should

pause. One cannot assume such assurance. No one knows when

kāvya first arose, much less what went on before. One does

not know when the "very beginning" was. One has to tread

very lightly with such words as "authentic" and "factually

correct." Authentic as opposed to what? Is it factually

incorrect that a later term comes into play in order to

reflect a different "reality"? And if a writer presumes to

speak of "finest wurks," let he or she grant at least a

hint of what they mean.

kulaka / a brief kāvya of up to fifteen padyas

(although some would consider five the maximum number24),

where the entire group forms a single syntactical unit

construing with a single verb placed either at the

beginning or at the end. Dandin’s coming description of

the mahākāvya [1.15-19] mirrors this type syntactically.

samghāta / a series of padyas greater in number than

the kulaka. All are in the same metre and all pertain to a

common theme, yet each is now syntactically distinct,

capable of standing alone.

kośa (or koṣa) / a "treasure," that is, an extended

anthology of individual stanzas or muktakas. These may be

by a single kavi or by a number of kavis; the arrangement

may be arbitrary, or (more usually) according to a

particular principle, such as theme, metre, or even

alphabetically (according to the first word of each padya).

The included padyas are "quotations from literary

Sanskrit works by known or unknown authors, being either

descriptive verses or single poetical verses standing by

themselves in which the poet concisely depicted a single

phase of emotion, or a single interesting situation within

the limits of a finely finished form."25

The earliest known kośa, and one most probably known

to Dandin and thus of some concern, is Hāla's Sattasaī

(also known as the Gāhākośo, Gāthakośa, or Gāthāsapta-

śatī).26 Hāla is often equated with King Śātavāhana, who

ruled from the city of Pratiṣṭhāna on the banks of the

Godāvarī River (in Māhārāṣṭra). The text appears in seven

recensions, with the total number of stanzas or gāthās

varying from roughly 700 to 1000, with perhaps some 430

believed to be "core verses." All of the gāthās are

written in Māhārāṣṭrī Prākṛt (as is Pravarasena's

Setubandha, which Dandin mentions in [1.34]), they are

undivided with respect to categories, and are most probably

drawn from a variety of sources with many added by Hāla

himself. Dating is vague, ranging between 200 - 600 AD

(perhaps tending toward the former).27

As Hāla himself wrote (and we note the reference to

alamkāras as a distinguishing feature, rather than, as the

dhvani revisionists would have us believe, "dhvani" or even

"rasa"). "Out of the ten millions of gāthās adorned with

alamkāras, seven hundred have been compiled by One

Compassionate towards Kavis (kavivatsala), Hāla"

[(transposed into Sanskrit:) sapta śatāni kavivatsalena

koṭayāḥ madhye | hālena viracitāni sālaṅkārāṇām gāthānām ||]

[1.3].28

Sternbach affirms that, "The gathas, intended to be

sung, contain poetry of the highest type. . . . Each gāthā

forms a unity in itself and only in some cases two or three

gāthās are combined to constitute a song. Not infrequently

a gāthā forms an epigram or an aphorism expressing a certain

truth in a few words and only rarely a gāthā contains

well-rounded narrative verses borrowed from another poem or

drama."29 And Keith (who might have been casting a

felicitous eye across the rolling Kent countryside) offers

a soothing picture, "The prevailing tone is gentle and

pleasing, simple loves set among simple scenes, fostered by

the seasons. . . ."30

Well, not quite. Here is Radhagovinda Basak in the

introduction to his edition, glossing a few verses on an

extremely common theme throughout the text:

The presence of unchaste women in societies

cannot be unthinkable, as we find in this treatise

mention of unchaste women in various contexts

(2.4, 65-66, 3.28, 94-95). As such unchaste women

knew the art of enticing chaste ladies to violate

their morality, the latter lived in consternation

against the activities of unchaste female

neighbours (1.36). An unchaste women does not

often fail to cite friends who could certify her

as possessing good character (2.97). It is

curious that unchaste women could easily

understand the entry into her house- precincts of

her husband or her paramour by the particular

barkings of her own dog (7.62). Wanton women

often went to meet their lovers through snow-clad

sesamum fields (7.93). We read of an unchaste

lady of high family to contract secret love with a

barber (5.17). We find a description of a wanton

woman besmearing her body with the cremation ashes

of her paramour (5.8). Under various pretexts

unchaste women conceal from their husbands their

connection with their paramours (4.1). There is

mention of a call of a physician paramour given by

an unchaste wife even in presence of her husband,

under the pretext of a treatment for a scorpion

bite (3.37). Such a bad woman often introduces to

her husband her paramour as a person seeking for a

refuge in her house (3.97). This anthology contains description of how harlots succeed in tarnishing the character of young men and fleecing their finance," and so on.31

We may assume that Dandin was aware of if not actually familiar with the Gāthāsaptaśatī. In [1.34] he mentions the Setubandha of Pravarasena, also written in Māhārāṣṭrī Prākṛt. He was certainly aware of Bāṇa (one of the kavis praised in the opening verses to the Avantiṣundarī), who refers to Hāla in [1.13cd] of the Harṣacarita: "Sātavāhana [Hāla] created a kośa replete with excellent sayings (subhāṣitas) of the purest nature like jewels" [ viśuddha-jātibhiḥ kośaṃ ratnairiva subhāṣitaiḥ ||].

We note that Bāṇa refers to the text simply as the "kośa." V. V. Mirashi believes that Hāla's anthology was known primarily as the "Gāthākośa" or simply as the "Kośa" down to the 9th century, and cites a number of instances, some of which would fall before Dandin.32

Following close to Dandin's time is the Vajjālagga (Kośa) by the Śvetāmbara Jaina Jayavallabha.33 Sternbach

believes its date is not much later than Hāla's work.34

Where M. V. Patwardhan in the introduction to his

translation of the text places it between the "broad

limits" of 750-1337 A.D.35

The Vajjālagga is also in Māhārāṣṭrī Prākṛt and

contains some 1350 gāthās, with perhaps 400 considered to

be core verses. Where Hāla's Gāthāsaptaśatī was not

divided into various thematic sections, the Vajjālagga is

broken into 95 "vajjas." This term (as vrajyā) marking a

thematic division became the norm for Sanskrit kośa. As we

find in the later Sahityadarpana [6.565] of Viśvanātha

[14th century]: "A kośa is a compilation of independent

stanzas (ślokas), classified according to vrajyās -- This

indeed is exceedingly pleasing" [ kośaḥ ślokasamūhastu

syādanyonyāna-pekṣakaḥ | vrajyākramaṇa racitaḥ sa

evātimanoramāḥ ||].

The vajjās themselves are grouped into three broad

categories according to three (of the four, excluding

mokṣa) ends or goals of human life: dharma, kāma (the

majority), and artha (see note 40 under Notes [1.10] -

[1.31]). The text does not provide any indication of the

various authors.

It is not until the 11th-12th centuries that we have

the first (extant) kośa (or subhāṣita samgraha) in

Sanskrit, the Subhāṣitaratnakoṣa of Vidyākara.36 It

contains 1739 verses, divided into 50 vrajyās, with perhaps

one-third of the verses attributed to specific authors.

There are sections on, for example, the Buddha and

Bodhisattvas, Śiva and Viṣṇu, the seasons, love and women,

the stages of life and times of day, good men and bad, the

flattery of kings and the praises of poets.37

We may briefly note three additional and specific

stanzaic forms: the yugmaka (yugma, yugala, yugalaka) in

two pādyas; the sandānitaka (viśeṣaka) in three pādyas; and

the kāpalaka in four pādyas.38

And the khandakāvya ("kāvya of a single fragment")

which, with its capability of interweaving a focused theme

(with secondary variations) and an embracing story line,

may be placed between, say, the sanghāta and the more

developed mahākāvya.

By this term [the Sanskrit critics] indicate that

the type is concerned with any of the subjects

assigned to the great kāvya but that it treats of

only one or of a small selection of the subjects

so assigned. In actual fact the khanda-kāvyas

preserved to us from the classical period may be

more narrowly characterized. With few exceptions

they fall into two categories: messenger-poems

(samdeśa-kāvya) and verse- sequences such as the

Centuries (śataka).39

100

The Sargabandha or Mahākāvya

"A Sargabandha is a Mahākāvya / Its distinctive

characteristics are: At its beginning there is either

benediction, salutation, or a statement of the theme"

[ sargabandho mahākāvamucyate tasya lakṣaṇam | āśīr-

namaskriyā vastunirdeśo vāpi tanmukham ||] [1.14].

"It is born from stories of the Itihāsas or from other

works based upon historical characters / It revolves around

the quest for the Four Goals40 with a protagonist lofty and

noble" [ itihāsakathodbhūtamitaradvā sadāśrayam |

caturvargaphalāyattam caturodāttanāyakam !!]. [1.15]

itihāsa [ < iti (+) ha (+) āsa ] /"thus it was."

"A generic name which includes all chronicles, legendary

tales, and heroic sages. . . ." (Notes 1/13). It is

"history" only in the loose sense that it concerns events

believed to have occurred in the past. Presenting such

popular material, it is not surprising that the itihāsas

(and the proto-Purāṇas) "were current in the early Vedic period."41

Yet their status was somewhat equivocal. Not generally included in the Vedas, they share common story elements (the "gambler's lament" [Rg Veda 10.34], for example). In certain later works they are classified as a fifth Veda, as in the Chāndogya Upaniṣad [3.4.1ff.]; and in the Artha Śāstra [1.2ff] we read: "The Sama Veda, the Rg Veda and the Yajur Veda constitute the trilogy of the Vedas. These, the Atharva Veda, and the Itihāsa Veda (the Veda of history and legends) make up the Vedas.

Phonetics, ritual, grammar, etymology, metrics, and astronomy -- these are the limbs [ancillary branches of knowledge] of the Veda."42

nāyaka /the hero or male protagonist. "Le héros est le personnage principal du drame, celui dont les aventures à la poursuite de l'object qu'il désire forment le sujet de la pièce et qui recueille au dénouement le profit suprème de

l'action. . . . C'est lui qui conduit les événements dans la

mesure des forces humaines et de sa volonté. Dans la

plupart des genres dramatiques, le héros doit être un

modèle presque accompli de vertus."43 And we should be

aware that the models of the "hero" and "heroine"

explicitly presented in the context of the drama apply in a

less structured way to the prosaic and stanzaic forms of

kāvya as well.

The Nāṭyaśāstra [34.17-21] distinguishes four types of

nāyaka in the play:

(1) dhīrodhata /"noble, firm self-controlled, and

haughty." They are "dominated by pride and jealousy; they

employ magic, ruse, and so on."44 This category generally

pertains to gods.

(2) dhīralalita /"noble, self-controlled, firm, and

light-hearted." "He does not have any worries, for his

friends look after his interest; he loves the fine arts,

song, dance and so on; he is devoted to pleasure, and

especially love; and finally he is naturally happy and

compassionate."45 This category generally pertain to

kings.

(3) dhīrodātta /"noble, self-controlled, firm and

exalted, superior. "He has a great heart which is never

dominated by depression or anger, and so on; a character

extremely profound, patient . . . faithful to his

promises."46 This category generally pertains to generals

and ministers.

(4) dhīrapraśāstra /"noble, self-controlled, firm, and

calm." "The calm hero differs primarily from the light-

hearted hero by reason of his birth, for he is a Brahmin or

merchant. . . ."47

From Dandin's Daśakumāracarita let us see a few of the

nāyaka's attributes revealed in practice.

(1) Bravery and facility at arms:

(Somadatta speaks) "Avoiding the tangled struggle of

the hostile hosts, wantonly delighting in my strength of

arm, I shot a shower of shafts and struck down my foes.

Then, guiding my splendid chariot-horses toward the enemy

king, I swiftly overtook his chariot and cut off his

head."48

(2) Cunning:

(Puspodbhava speaks) 'I received a message

from Balachandrika that she was preparing to visit

Daruvarman, having been summoned to sport in the

love-chamber by this victim caught in the tangling

toils of my device. Therefore I deftly affixed to

the proper portions of my person the frippery

appropriate to a pretty girl. . . .

For just twenty seconds he stood there

chattering, and laughing as he talked; then, blind

with passion, showed a mind to fondle the sweet

maiden's bosom.

My turn had come. Red with wrath, I dashed him

headlong from the couch and drubbed him dead with

fists and knees and feet.'49

(3) Supernatural power:

(Rājavāhana) "Rājavāhana . . . previously instructed

in the mechanics of disappearance, floated like a specter

into the maidens' apartments."50

(4) Virtue:

(Apahāravarman speaks) "Since I desired to bring

these gentry to orthodox thinking by revealing the

perishable nature of riches, I resolved to tread the path

of scientific thievery."51

And, allowing for his biased anger, we note King

Chandavrman's less than felicitous enunciation of

Rājavāhana's qualities: "Aha! Here he is, the friend of

Puṣpodbhava, that foreign son of a merchant, that money-mad

prig, that husband of Balachandrika who caused my younger

brother's death -- damn her! Here he is, the handsome

coxcomb, the arrogant artist, who tickles the silly

townsfolk with his skill in a pack of juggling tricks, and

fools them by shamming the dignity of something superhuman!

A bogus robe of virtue outside, and rottenness inside! A

mountebank! A quack!"52

nāyikā /the heroine or female protagonist. "The

heroine . . . is the one among the female characters who

drives or guides the weave of primary events. The

character of the heroine does not contribute less than that

of the hero in giving to the drama its [']physionomie

spéciale[']."53

The Nāṭyaśāstra [34.25cd-26ab] lists four types:

(1) divyā /a goddess; (2) nrpapatnī /a queen; (3)

kula-strī / a women of high family; and (4) ganikā /a

courtezan.

These are in turn are characterized according to

variations in four attributes (Nāṭyaśāstra [34.26cd-28]):

whether (1) dhīrā /"self-controlled," "restrained"; (2)

lalitā /"playful," "light-hearted"; (3) udāttā /"exalted,"

"superior"; or (4) nibhṛtā /"modest." Thus a goddess or

queen will display all four attributes; a women of high

family will display superiority, exaltedness, and modesty;

and a courtezan (or one proficient in all the "skills")

will display playfulness and superiority.

We should also be aware of a threefold typology which

regularly appears in kāvya, where the nāyikā may be

classified according to her physical development and sexual

experience: (1) mugdhā /"die Naive" ("Die mugdhā ist noch

rech ungeschickt in Liebesdingen, sehr sanft, wo sie zürnen

müsste, und überaus verschämt"); (2) madhyā /"die Mittlere"

107

("Die madhyā ist schon viel anstelliger. In der Liebe ist

sie schon erfahrener und auch körperlich ist sie schon mehr

entwickelt"); and (3) pragalbhā (prauḍhā) /"die Leiden-

schaftliche" ("Die pragalbhā dagegen ist liebesblind und

begehrt einen schrankenlosen Liebesgenuss. Sie ist das

Weib in der vollen Blüte der Jugend").54

From Dandin's Daśakumāracarita we have a description of

the nāyikā, indeed a revealing catalogue of many of kāvya's

recurring descriptive attributes, a number of which we shall

see in the examples of the Kāvyādarśa's second chapter:

She shone, a creation of Love. Yes, Love had

fashioned a paragon of women, as if he wished, in

wistfull memory of Charm, to image forth this

duplicate. He formed her feet from the sweetness

of two autumn lilies in his own pleasure pool; the

languid grace of her gait from the course of a

swan down a long lake in a planted garden;

her calves from a quiver's curve; her comely

thighs from the shapeliness of two plantain stems

by the door of a summer-house; her generous hips

from the sweep of conquering chariots; her navel

(which seemed an eddy in Ganges' stream) from the

semblance of an early- flowering ornamental lotus

bud; . . . her breasts from the beauty of two full

golden bowls; her arms from the delicacy of vines

in a bower; her neck from the symmetry of a conch

of victory; her lip, like a bimba fruit, from the

redness of mango flowers that maidens fondly wear above the ear; her sweet smile from the splendor of Love's flower-arrows; her every word from the witchery of the soft song of Love's first messenger, the cuckoo; the breath of her sigh from the gentleness of the southern breeze, leader of all Love's soldiers; her eyes from the pride of two fishes figured on a conquering banner; her brows from the curve of a bow; her face from the spotless enchantment of Love's first friend, the moon; her hair from the similitude of a pet peacock's fan.55

"With description of cities, oceans, mountains, the seasons / the rising of the moon and sun / play among gardens and pools / drinking and festivals of love"

[ nagarārṇavaśailartucandrārkodayavarnanaih | udyāna-salilakrīḍāmadhupānarāratotsvaih || ] [1.16].

"With descriptions of the separation and marriage of lovers / the births of sons / diplomacies, ambassadors, expeditions, battles / and the success of the protagonist"

[ vipralambhairvivāhaiśca kumārodayavarnanaih. | mantradūtaprayāṇājināyakābhyudayairapi || ] [1.17].

"It should be embellished and be not too condensed / endowed with a continuous stream of Rasas and Bhavas / with

109

sargas not excessively long / possessing melodious metres

and effective transitions" [ alamkrtamasamsiptam

rasabhavanirantaram |sargairanativistirnaih sravyavrttaih

susamdhibhih ||] [1.18].

"With the final metre of each sarga different from all

preceding / Such a kavya displaying a profusion of alamkaras

shall be pleasing to the world and endure for yet another

eon"56 [ sarvatra bhinnavrttantairupetam lokaraijanam |

kavyam kalpantarasthayi jayate sadalamkrti || ] [1.19].

Critical Acceptance as the Essential Criterion --

Not Formulaic Adherence

"A kavya although short of some of these features is

not necessarily defective if the excellence of those

employed pleases the wise" [ nyunamapyatra yaih

kaiscidangaih kavyam na dusyati | yadyupattesu sampatti-

raradhayati tadvidah ||] [1.20].

The Ākhyāyikā or Kathā

"The prosāic is a succession of words devoid of pādas

/ Its varieties are two: Ākhyāyikā and Kathā / Of these

some affirm that the Ākhyāyika [1.23] should be narrated

just by the protagonist / The other by the protagonist or

someone else / Here there is no defect in presenting one's

own qualities as one proclaims the truth" [1.24] [ apādah

padasantāno gadyamākhyāyikā kathā | iti tasya prabhedau duau

tayorākhyāyikā kila ||] [1.23] nāyakenaiva vācyānyā

nāyakenetarena vā | svagunāviṣkriyādoṣo nātra bhūtārtha-

saṃsināḥ ||] [1.24].

"However, the lack of such restriction is actually

seen / since in the Ākhyāyikā also there is narration by

others / Whether the narrator is the protagonist or somone

else / What basis for distinction is this?" [ api tvaniyamo

drṣṭastatrāpyanyairudīranāt | anyo vaktā svayaṃ veti

kīdṛgvā bhedakāraṇam ||] [1.25].

"If the distinguishing marks of the Ākhyāyikā are

either the vaktra or aparavaktra metre and its division into

ucchvāsas / Then in kathās as well -- due to their relevance

-- [1.26] / Why not utilize the vaktra or aparavaktra

metres just as it already employs the āryā? / And as

'lambha' and so on mark the kathā's divisions / Let

'ucchvāsa' be used as well -- What of it?" [1.27] [vaktraṃ

cāparavaktram ca socchvāsatvaṃ ca bhedakam | cihnamākhyāyi-

kāyāścet prasariṇgena kathāsvapi || āryādivat praveśaḥ kiṃ na

vaktrāparavaktrayoḥ | bhedasca dṛṣṭo lambhādirucchvāso vāstu kiṃ tataḥ ||].

"Therefore the Kathā and Ākhyāyikā are really one

genre marked by two names / And within this any remaining

types of prosaic narration will be subsumed" [ tat

kathākhyāyiketyekā jātiḥ samjñādvayāṅkitā | atraivāntar-

bhaviṣyanti śeṣāścākhyānajāatayaḥ ||] [1.28].

"Abduction of virgins, battles, the separations of

lovers, triumphs, and so on, are certainly common in

Sargabandhas as well -- These are not distinguishing

112

attributes" [kanyāharaṇasaṃgrāma vipralambhodayādayah |

sargabandhasamā eva naite vaiśeṣikā guṇāḥ ||] [1.29].

Daṇḍin specifically rejects a number of points evident

in Bhāmaha’s Kāvyalaṅkāra [1.25-29]. Bhāmaha maintains, for

example, that the ākhyāyikā should be narrated only by the

protagonist immediately involved with events, where the

kathā should be narrated by some character(s) other than the

protagonist; or that the ākhyāyika must present the

"abduction of virgins, battles, the separation of lovers,

and triumphs" (the Sanskrit in each case here is exactly the

same). Daṇḍin, as a master of the extended prose form

himself, rejects a distinction seemingly made for its own

sake, with little basis in actual practice. One should be

aware, however, that in approaching the secondary literature

one will frequently find this artificiality maintained,

with the kathā and ākhyāyika presented as clearly defined

genres. S. K. De, for example, affirms that "The ākhyāyika

was more or less a serious composition dealing generally

with facts of actual experience with an autobiographical or

semi-autobiographical interest; while the kathā was

essentially a fictitious narrative -- which may sometimes

(as Daṇḍin contends) possess an autobiographical form, but

whose interest chiefly resides in its invention."57

This error is carried to an extreme with A. K. Warder,

who not only solidifies these distinctions, and falsely

equates them with Western literary genres, but condemns

Daṇḍin for not adhering to these misconceived views: "Of

Daṇḍin's rather idiosyncratic ideas about literature,

particularly that the distinction between history and

fiction should not be recognized. . . ."; or speaking of

Daṇḍin's "deliberately confounding history and fiction, or

biography and novel. . . ."58

Again on the Freedom of the Kavi

"A feature realized here through the thought of the

kavi is not wrong elsewhere / Among the accomplished what

indeed cannot be an opening onto the achievement of their

goals?" [ kavibhāvakṛtam cihnamanyatrāpi na duṣyati |

mukhamistārthasamsiddhau kim hi na syāt kṛtātmanām || ]

[1.30].

Mixed Compositions -- The Campū

"Mixed compositions are the nāṭakas and so on / Their

detailed treatment is found elsewhere / Another such

variety -- abounding in both the prosaic and stanzaic is

called campū " [ miśrāṇi nāṭakādini teṣāmanyatra vistaraḥ |

gadyapadyamayī kāciccampūratyabhidhīyate || ] [1.31].

"nāṭakas and so on" /that is, the ten primary types of

rūpakas or "plays," "dramas":

(1) the nāṭaka as such: "La comédie héroique est le

type le plus complet de l'oeuvre dramatique."59 According

to the Nāṭyaśāstra [20.7] it may display "all the vṛttis

("styles") and a number of varied situations"

[sarvavṛttiviniṣpannam nānāvāsthāsamaśryam ||].

There are four vṛttis (NŚ [6.24cd-25ab]): bhāratī /the

"verbal"; sāttvatī /the "elegant"; kaiśikī /the "graceful";

and ārabhaṭī /the "vigorous." ("Une variante des ṛti [see

verses [1.40] ff.] est constituée par les vṛtti ou 'modes':

cette discrimination entre les styles élégant, ordinaire,

grossier, émané de la Dramaturgie et n'a été appliqué que

secondairement, et non sans gaucheries, à la Poétique."60

On the nāṭaka Lévi writes, "Le style doit en être noble

et harmonieux; les parties en prose veulent des expressions

sans recherche et des composés de peu d'étendue . . .; les

vers, une langue claire et douce."61 Its title should

reflect the subject matter, which should be divided into

five to ten acts (aṅka). Later self-styled nāṭakas may have

less (the Janakīparinaya of Madhusūdana [18th century] is in

four acts) or more ("Il existe même un drame en quatorze

acts, sorte de monstre, attribué à Hanumat: le

Hanuman-Nāṭaka"62).

Its material should be well-known, that is, not

invented. It may utilize any number of rasas, yet

primarily employs vīra (the "heroic") and śṛṅgāra (the

"erotic"). And the protagonist (nāyaka) should have an

elevated, superior nature (udātta). "Where the behavior of

kings reflected in their joys or sorrows is variously

realized through actions displaying the rasas and bhāvas --

This should be known as the 'nāṭaka'" [ nrpatīnāṃ yaccaritam

nānārasabhāvaceștitairbahudhā | sukhaduhkhotpattikṛtam

tajjñeyaṃ nāṭakaṃ nāma || (NŚ [20.12]).

(2) the prakaraṇa ("comédie bourgeoise") follows the

structure and development of the nāṭaka (NŚ [20.50]), but

now with a plot generated through the creative power of the

kavi (kavirātmakaśktyā) (NŚ [20.48]). The nāyaka may be a

brahmin, merchant, minister, officer, or caravan leader (NŚ

[20.51]), "toujours du genre noble et calme." The nāyikā

may be of similar status as that of the nāyaka, she may be

a courtezam; or two women may appear drawn from each of

these two categories.

As with the nāṭaka, the prakaraṇa should have from

five to ten acts, and it should possess the various rasas

and bhāvas (NŚ [20.57]). The name of the individual play

may be formed from the name of the nāyaka, the nāyikā, or a

conjunction of both.

(3) the samavakāra ("le drame surnaturel") presents

the adventures of gods and demons in three acts. It may

have up to twelve nāyakas (NŚ [20.64-65]) -- "tous du genre

noble et supérieur; chacun d'eux poursuit un object

particular qu'il finit par atteindre."63

(4) the īhāmṛga displays divine beings in conflict

over love (NŚ [20.78]). "It is to abound in venement

Heroes and to have its construction dependent on feminine

anger which is to give rise to commotion, excitement and

conflict" (NŚ [20.79]).64 The vṛttis and rasas that apply

are the same as in the vyāyoga (NŚ [20.81]) ( see below).

(5) the dima ("le drame fantastique") presents a

well-known plot and an exalted nāyaka (NŚ [20.84]). It

should have four acts and display the various rasas except

śṛṅgāra (the "erotic") and hāsya (the "comic") (NŚ [20.85]),

and display sixteen nāyakas, who may be devas, asuras,

rākṣasas, bhūtas, yakṣas, and nāgas (NŚ [20.87-88]). "La

magie, la sorcellerie, les combats, les fureurs, les

éclipses de lune et de soleil contribuent à augmenter

l'horreur de l'action."65

(6) the vyāyoga ("le spectacle militaire") should have

only one act, representing the passage of one day (NŚ

[20.90-91]. The single nāyaka should be a well-known

(though not divine) royal sage (rājarṣi). The action is of

battle and conflict, evoking "exciting"/"blazing" rasas

(that is, all but śṛṅgāra or hāsya).

(7) the utsṛṣṭikāṅka ("l'acte en dehors" or isolated

act) has a plot that is usually well-known with non-divine

male characters (NŚ [20.94]). It should express karuṇa

(the "compassionate") rasa, be in the bhārati ("verbal")

vṛtti, and it should concern women in mourning who describe

recently completed combat (NŚ [20.95-96]). "Le nom d'acte

en dehors est donné à ce genre pour le distinguer de l'acte

simple, qui est une des divisions de la comédie héroïque

[nāṭaka]. Certain théoriciens l'entendent: acte en dehors

des règles ordinaires."66

(8) the prahasana ("la comédie bouffe" or farce) is

in one act and should primarily evoke hāsya rasa. There

are two types: śuddah ("pure"), involving comical

arguments among ascetics, brahmins, heretics, and so on

(NŚ [20.103-4]); and miśra or sañkīrṇa ("mixed"), involving

harem guards, eunuchs, courtezans, "galants," and so on (NŚ

[20.105]). "Some popular topic [of scandal] or incident of

hypocrisy should be introduced. . . ."67

(9) the bhāṇa ("le monologue") is in one act with a

single character who should be either a dhūrta or viṭa who

relates either his own or another's actions (how much in

even the smallest instance translation may reveal of

cultural conditioning: A. B. Keith translates this

character as "parasite"; Sylvain Lévi as "un bel esprit").

(10) the vīthī ("la guirlande") is also in one act

with either one or two characters, who may be of high,

middle, or low status. It may evoke any of the rasas (NŚ

[20.112-13]); -- "on l'appelle la guirlande, parce qu'elle

est composée de parties successives."68

campū / The conjunction of the gadya and padya forms

(prosaic/stanzaic) in Indian literature certainly goes back

to an early date -- "We can safely accept the view that the

form is quite old . . . it may be admitted that the prose-

poetic form goes back beyond the beginning of the first

century b.c. . . ."69 It appears, for example, in the

early Buddhist avadānas and sūtras, and in the early story

collections such as the Pañcatantra.

In passing we may mention the dated "ākhyāna theory"

of H. Oldenberg, which postulated that certain of the

dialoque hymns of the Ṛg Veda represented such a mixed

format, originally including prose explanations that were

later lost (as in, for example, the following hymns:

[1.170, 171], [1.179], [8.91], [8.100], [10.51-53],

[10.95], [10.124]).70 A. B. Keith, who effectively refutes

this theory, summarizes:

We are . . . to conceive of a form of literature

which was essentially a mixture of prose and

verse, and which was narrative in character. But

with the natural liking of people for direct

speech, the narrative every now and then took the

dialogue form. . . . And in these passages verse was normally used. It was not necessarily confined to these passages, but it might occur whenever there was a heightening of the interest or of the feeling."71

And (among various other reasons presented by Keith) I would agree with Sylvain Lévi's polite assessment:

"L'hypothèse est ingénieuse, mais elle ne s'impose pas. L'exposition est en général si nette, le dialogue si bien suivi, qu'un commentaire narratif paraîtrait superflu."72

Lévi himself would see in the Rg Vedic dialogue hymns "la structure pré-dramatique ou semi-dramatique"73 (following the initial proposition of Max Müller (1869)74).

Yet Dandin in speaking of "mixed compositions" is referring to something other than the mere conjunction of anything that might appear in the metrical stanza or the prosaic line. The verses of the early stories do "mark a heightening of the interest, for the verses often contain in summary form the point of the narrative. But . . . the essential nature of the verses is gnomic. . . ."75

Displaying a straitforward style that suitably conveys

the fabled message, the stories do not fail to entertain

and instruct -- "La prose du Pañcatantra est en général

aisée, sans raffinements (bien qu'elle utilise à l'occasion

quelques effets de style); les versets dont elle s'orne

sont plus simples eux aussi que ceux de la poésie gnomique

ultérieure."76 But we do not have kāvya.

What is clear is that in the earliest extant nāṭyas

forward we find a conjunction of the two forms -- as

befitting kāvya -- at a more refined level. The prosaic

line of dialogue or narrative would seem to be a natural

extension of the stories. Its role now is to carry the

story forward in a manner that can entertain and capture an

audience.

Au tēmoignage du conte et de la fable s'ajoute

par un lien natural celui des portions dialoguées

du théâtre. Le dialogue dramatique en prose n'a

suivi que d'assez loin la progression vers

l'artifice qui marque les portions strophiques.

Certes on rencontre, suivant les circonstances de

l'action, des passages élaborés. . . . Main en

gros, il existe une tradition persistante de style

simple, direct, visant évidemment à reproduire le langage courant. . . ."77

The stanzas, however, are nodal points of importance,

compressing summation, revelation, the evocation of the

appropriate rasa, and so on , into their brief space. "The

place of poetry in the drama is extremely important. When

a situation calls for the expression of a truth, the

evocation of a sentiment, the recollection of a significant

event, it calls for poetry. The stanza may be at once

narrative and self-contained, but it is always the climax

of an episode, however minor."78

Across time the theatrical stanzas came increasingly

to reflect the more complex and linguistically involved

padyas of the mahākāvya and the more restrained forms.

That where in the relatively earlier kāvis such as Bhāsa

and Kālidāsa we find stanzas "qui sont aussi dénuées de

recherche que la prose environnante," with Bhavabhūti the

"strophe emphatique, grandiloquente, riche en allitérations

et en mot rares" makes its appearance.79

What exactly Daṇḍin means when he speaks of the campū

-- a term of unknown origin -- is unclear. The first

extant example is the Nalacampū (Damayantīkathā) of

Trivikramabhaṭṭa [10th century], which relates in seven

ucchvāsas an isolated portion of the story of King Nala and

Queen Damayantī.

The reaction of various writers to this work is

instructive, a chronological progression of primarily

paraphrase yet each with their individual turn. Keith

displays the characteristic "block" toward the complex

style: "The story is elaborated with the usual defects of

long sentences, consisting of epithets heaped on epithets

in long compounds, with double meanings, alliterations and

jingles complete."80

S. K. De, being pulled perhaps in two directions (as is

apparent in the approach of a number of Indian scholars

educated and productive in the closing decades of the Raj

and which is not necessarily worse that certain undiluted

variants of either extreme), affirms that the author

believes in the display of verbal complexities

after the manner of Bāṇa and Subandhu, and

deliberately, but wearisomely, imitates their

interminably descriptive, ingeniously recondite

and massively ornamented style. He has a decided

talent in this direction, as well as skill in

metrical composition, and elegant verses from his

campū are are culled by the Anthologists, but

beyond this ungrudgingly made admission, it is

scarcely possible to go further in the way of

praise.81

And Lienhard, who shows a greater degree of openness

to the text, notes that it "is written in difficult prose

full of erudition and paronomasia. The fact that the

stanzas from the Nalacampū are included in various

anthologies shows that Trivikrāma's poetry won the approval

of the critics."82

It should not be surprising then to find, in the case

of De and Lienhard, evaluations of the campū as such in

connsonance with these judgments. For De "The Campū . . .

shares the features of both Sanskrit prose and poetry, but

the mosaic is hardly of an attractive pattern. . . . The

Campū has neither the sinewy strength and efficiency of

real prose, nor the weight and power of real poetry; the

prose seeking to copy ex abundanti the brocaded stateliness

of the prose Kāthā and the verse reproducing the

conventional ornateness of the metrical Kāvya."83 Where

for Lienhard the "true campū" is a "calculated balance

between prose that is as perfect as possible and stanzas in

the genuine kāvya style."84

On the origins and development of the campū prior to

Dandin we have only speculation. Louis Renou would see the

antecedents of the campū in the Buddhist Jātakamālā and

perhaps in the inscription of Hariṣeṇa at Allahābad [4th

century] -- "sorte de campū épigraphique"-- a panegyric to

Samudragupta: "un texte mi-versifié mi en prose, se

prétendant un kāvya, qui condense en une seule phrase

interminable le généalogie et les mérites du souverain . . .

le tout dans une langue pleine d'artifices. . . ."85

Both De and Lienhard do agree that the campū developed

from the extended gadya kāvya, but again their views of the

motivation and processes involved are something else again.

Thus De would see a "disregard for form" arising out of

"stress": "Its formlessness, or rather disregard of a

strict form, shows that [the campū] developed quite

naturally, but haphazardly, out of the prose Kāvya itself,

the impetus being supplied by the obvious desire of

diversifying the prose form freely by verse as an

additional ornament under the stress or the lure of the

metrical Kāvya."86 Where Lienhard would see "harmony and

balance" arising in stately and methodical sequence from

the quest for "a new and more difficult form":

In just the same way as the prose novel arose

[where above he terms this form the "prose

poem"(?)] as a deliberate contrast to metrical

kāvya, which was then fully developed, so campū

was created when Sanskrit poets had mastered all

the fine points and difficulties of prose. Only

when the need was felt to find a new and more

difficult form of kāvya did poets endeavor to

develop a new genre -- campū -- in which both

poetry and prose combined to produce a harmonic

and balanced whole.87

As they say, one tends to see what one wishes to see.

I have included these excerpts to once again emphasize the

extreme critical caution with which one must approach this

material. At least writers with clearly biased views or

perceptible orientations usually tend to judge and evaluate

in character; these views will tend to reoccur, albeit

perhaps in strange and mysterious shape. Armed with this

awareness, the validity or feasibility of a given author's

speculations -- usually curiously disguised as absolute

truth -- may be weighted.

Given Daṇḍin's awareness of the campū, the fact that

the first extant example does not appear until the 10th

century hardly allows one to affirm or present such linear,

progressive development -- whether unbalanced or harmonious

as you wish. Of course, the form that the Nalacampū

displays may not be exactly what Daṇḍin had in mind, but

what exactly he did is inaccessible. What is clear,

however, is that the campū as an established form of kāvya

was recognized as such by the latter 7th to early 8th

centuries. But that the potentialities of this form were

explored at a much earlier date, in the story literature

and in the nāṭya kāvyas as well. And in this regard I

would certainly accept that "it is inadvisable to use the

term campū indiscriminately of any mixture of prose and

verse or to define as campū works like Ārya Śūra's

Jātakamālā or, still less, a book of fables like the

Hitopadeśa, as is often done [as, for example, by Louis

Renou and A. B. Keith88]. Although these books do contain

both prose and verse, their authors did not write them as

campūs."89

I tend to feel that whenever the gadya and padya forms

were established in and as kāvya, their mixture -- as kāvya

-- in whatever varying degrees would follow soon after.

130

Notes [1.10] - [1.31]

1. Charles P. Brown, Sanskrit Prosody and Numerical Symbols; Reprint (New Delhi: Asian Publication Services, 1981 (1869), p. v.

2. Hank Heifetz, "Issues of Literary Translation from Sanskrit and Tamil," Ph.D. dissertation, University of California, Berkeley, 1983, p. 171.

3. See: Vaman Shivaram Apte, "Appendix A: Sanskrit Prosody," in The Practical Sanskrit-English Dictionary, Revised and enlarged edition (Poona:, 1957 (1890)); Reprint (Kyoto: Rinsen Book Co., 1978). Charles P. Brown, Sanskrit Prosody and Numerical Symbols (London:, 1869); Reprint (New Delhi: Asian Publication Services, 1981). Michael Coulson, Sanskrit (London: Hodder and Stoughton, 1976), pp. 21-22; 249-55. Sheldon I. Pollock, Aspects of Versification in Sanskrit Lyric Poetry (New Haven: American Oriental Society, 1977). Louis Renou, "Notions de Métrique," in L'Inde Classique, vol. 2, pp. 713-19. H. D. Velankar, "Prosodial Practice of Sanskrit Poets," Journal of the Bombay Branch of the Royal Asiatic Society, vols. 24-25 (1948-49), 49ff.

And also: Edward A. Bloom, et al., "Versification," in The Order of Poetry, pp. 105-38 (New York: The Odyssey Press, 1961). Jean Cohen, "Niveau Phonique: La Versification," in Structure du Langage Poétique, pp. 53-104 (Paris: Flammarion, 1966). Paolo Valesio, "On Poetics and Metrical Theory," Poetics, 2 (1971), 36ff.

4. H. D. Velankar, "Prosodial Practice of Sanskrit Poets," Journal of the Bombay Branch of the Royal Asiatic Society, 24-25 (1948-49), 50.

5. Michael Coulson, Sanskrit (London: Hodder and Stoughton, 1976), 250.

6. Louis Renou, "Notions de Métrique," in L' Inde Classique, vol. 2, p. 717.

7. V. S. Apte, "Appendix A: Sanskrit Prosody," in The Practical Sanskrit-English Dictionary, p. 11 (of appendix).

8. Hermann Jacobi, "On Indian Metrics," Wiener Zeitschrift für die Kunde des Morgenlandes, 5 (1891), p. 153.

9. Louis Renou, "Notions de Métrique," p. 716.

10. A. B. Keith, A History of Sanskrit Literature, p. 47.

11. As, for example, Hermann Jacobi, "Miscellen: Die Musterverse der Metriker," Zeitschrift der Deutschen Morgenländischen Gesellschaft, 40 (1886), p. 100.

12. P. V. Kane, "The Chandovicitti," Indian Actiquary, 40 (1911), p. 177. See: The Chandas Śastra by Śrī Pingalanāga, with the commentary Mrtasamjīvanī by Śrī Halāyudha Bhaṭṭa, edited by Paṇḍit Kedāranāth, 3rd edition (Bombay: Nirnaya Sāgar Press, 1938 (1908). And with a German translation, Piṅgala, Das Chandahśūtram des Piṅgala, translated by Albrecht Weber, in Indische Studien, vol. 8: Ueber die Metrik der Inder (Berlin: Harrwitz und Gofsmann, 1863), Reprint (Hildesheim: Georg Olms Verlag, 1973), pp. 157-462.

13. Cited in Hari Chand, Kālidāsa, p. 79.

14. Chandoviciti: Texte zur Sanskritmetrik, edited by Dieter Schlingloff (Berlin: Akademie-Verlag, 1958).

15. See: Piṅgala's Chandas Sūtra [3.30], Albrecht Weber's edition, (1863), p. 243 and pp. 244-47. And as Dieter Schlingloff notes, "Ob dieser Yāska mit dem berühmten Verfasser der Nirukti identisch ist, ist unsicher (Chandoviciti, edited by Dieter Schlingloff, (1958), p. 20, n. 5.

16. Albrecht Weber, Das Chandaḥsūtram des Piṅgala, Indische Studien, vol. 8. Reprint (1973), p. 193ff.

17. Chandoviciti, edited by Dieter Schlingloff, (1958), pp. 14-15.

18. Chandoviciti, edited by Dieter Schlingloff, (1958), p. 26.: "These differences make it unlikely that the Naṭya-śāstra is the source of our text or that our text is the source of the Nāṭyaśāstra."

19. John Brough, "Review: Dieter Schlingloff, editor, Chandoviciti, Texte zur Sanskritmetrik (Berlin: Akademie-Verlag, 1958)," Bulletin of the School of Oriental and African Studies, University of London, 22 (1959), p. 192.

20. Siegfried Lienhard, History of Classical Poetry, pp. 67-68.

21. Siegfried Lienhard, History of Classical Poetry, p. 75.

22. Louis Renou, "Sur la Structure du Kāvya," Journal Asiatique, 247 (1959), p. 63, n. 3.

23. Siegfried Lienhard, History of Classical Poetry, p. 75.

24. A. B. Keith, A History of Sanskrit Literature, p. 376.

25. Ludwik Sternbach, Subhāṣita, Gnomic and Didactic Literature (Wiesbaden: Otto Harrassowitz, 1974), p. 2.

26. Hāla, Das Saptaśatakam des Hāla, edited by Albrecht Weber, Abhandlungen für die Kunde des Morgenlandes, vol. 7, no. 4. (Leipzig, 1881); Reprint (Nendeln, Liechtenstein:

Kraus Reprint, 1966). The Prakrit Gāthāsaptaśatī, Compiled by Sātavāhana King Hāla, edited with English translation by Radhagovinda Basak (Calcutta: Asiatic Society, 1971).

Albrecht Weber. Über das Saptaśatakam des Hāla: Ein Beitrag zur Kenntniss des Prākrit, Abhandlungen für die Kunde des Morgenlandes, vol. 5, no. 3. (Leipzig, 1870); Reprint (Nendeln, Liechtenstein: Kraus Reprint, 1966). G. Garrez.

"Nouvelles et Mélanges: Ueber das Saptaśatakam des Hāla. Ein Beitrag zur Kenntniss des Prākrit, von Albrecht Weber. Leipzig, 1870," Journal Asiatique, 20 (1872), 197-220.

Albrecht Weber, Ueber Bhuvanapālā's Commentar zu Hāla's Saptaśatakam, Indische Studien, 16 (Leipzig, 1883); Reprint (Hildesheim: Georg Olms Verlag, 1973), pp. 1-204.

27. A. B. Keith, A History of Sanskrit Literature, p. 224 ; V. V. Mirashi, "The Date of Gāthāsaptaśatī," Indian Historical Quarterly, 23 (1947), 300-10 ; Ludwik Sternbach, Subhāṣita, Gnomic and Didactic Literature, p. 11.

28. Hāla, The Prākrit Gāthāsaptaśatī, edited and translated by Radhagovinda Basak (Calcutta: The Asiatic Society, 1971), p. 1.

29. Ludwik Sternbach, Subhāṣita, Gnomic and Didactic Literature, p. 12.

30. A. B. Keith, A History of Sanskrit Literaure, 224.

31. Hāla, The Prākrit Gāthāsaptaśatī, edited by Radhagovinda Basak (Calcutta: The Asiatic Society, 1971), p. 13.

32. V. V. Mirashi, "The Original Name of the Gāthā saptaśatī," Proceedings and Transactions of All-India Oriental Conference. 13th Session. Nagpur, 1945, 370-74.

33. Jayavallabha, Vajjālaggam: A Prakrit Anthology with Sanskrit Version by Julius Laber (Calcutta: The Royal

Asiatic Society of Bengal, 1944). Jayavallabha,

Jayavallabha's Vajjälaggam, with the Sanskrit Commentary of

Ratnadeva and English translation by M. V. Patwardhan

(Ahmedabad: Prakrit Text Society, 1969).

34. Ludwik Sternbach, Subhāsita, Gnomic and Didactic

Literature, p. 14.

35. Jayavallabha, Jayavallabha's Vajjälaggam, with the

Sanskrit Commentary of Ratnadeva and English translation by

M. V. Patwardhan (Ahmedabad: Prakrit Text Society, 1969),

p. xxi.

36. Vidyākara, The Subhāṣitaratnakoṣa, compiled by

Vidyākara, edited by D. D. Kosambi and V. V. Gokhale

(Cambridge: Harvard University Press, 1957). Daniel H. H.

Ingalls, trans., An Anthology of Sanskrit Court Poetry:

Vidyākara's Subhāṣitaratnakoṣa (Cambridge: Harvard

University Press, 1965); and also, Sanskrit Poetry from

Vidyākara's Treasury, Reprint (Cambridge: Harvard

University Press, 1972 (1965).

37. A sketch of the more important Sanskrit kośas would

include : Subhāṣitaratnakoṣa of Vidyākara (c. 1100– 1130);

Saduktikarnāmṛta of Srīdhardāsa (1205); Sūktimuktāvalī of

Bhagavata Jalhaṇa (1258); Śārṅga-dharapaddhati of

Śārṅgadhara (1363); Subhāṣitasudhānidhi of Śāyaṇa (15th

century); Sūktiratnahāra of Sūrya (15th century); Prasannasāhitya-

ratnākara of Nandana (15th century); Padyāvalī of Rūpa

Gosvamin (15th-16th centuries); Subhāṣitahārāvalī of

Harikavi (17th century); Śṛṅgārālapa of

Lakṣmaṇabhaṭṭa Āṅkolakāra (1625-1650); Rasikajīvana of

Gadādharabhaṭṭa (17th century); Sabhyālaṅkaraṇa of

Govindajit (after 1656); Padyaveṇī of Veṇīdatta (1644 or

1701); Sūktisundara of Sundaradeva (1644-1710);

Padyāmṛtatarāṅgiṇī of Haribhāskara (1674);

Subhāṣitasārasamuccaya (end of the 17th century);

Subhāṣitasavaskrta (?)-śloka (18th or 19th centuries);

Vidyākarasahasraka of Vidyākara Miśra (19th century).

(Ludwik Sternbach,. Poésie Sanskrit dans les Anthologies

et les Inscriptions, Vol. 1 (Paris: College de France

Institut de Civilisation Indienne, 1980), pp. xviii-xix.

38. Siegfried Lienhard, History of Classical Poetry, p. 66.

39. Daniel H. H. Ingalls, Sanskrit Poetry from Vidyākara's

"Treasury," pp. 37-38.

40. caturvarga (or puruṣārtha) / The "Four Goals or Ends" of

human life:

(1) dharma focuses on harmonious, proper and

efficatious order, and on the necessary and correct

behavior of the individual or group within this

all-embracing scheme of things. In its ultimate sense

"Dharma is the foundation of the whole universe. . . . Upon

dharma everything is founded" (Taittirīya Āranyaka [10.79])

(V. Raghavan, "The Four Ends of Man," in Sources of Indian

Tradition, edited by Wm. Theodore de Bary, vol. 1 (New

York: Columbia University Press, 1964), p. 244).

It is ultimately personal for it delineates the

appropriate means of integration within this order. And it

is perhaps the most usual example of a conceptual technical

term in Sanskrit that defies translation (and where although

any number of writers may remark upon this particular case,

the distorted translations of any number of similarly

resistant terms are offered). Biardeau most appropriately

notes,

Mais il faut rejecter tous les équivalents anglais

ou francais du terme dharma proposés par les

traductions courantes qui cherchent à rapprocher

la réalité indienne du lecteur occidental. Ce

n'est ni la morale, ni le bien, ni le droit, ni la

justice, ni la loi. C'est l'ordre socio-cosmicque,

dont on peut dire qu'il est bon simplement dans la

mesure où il est nécessaire au maintien de

l'existence heureuse du tout constitué par les

'trois-mondes'. . . . (Madeleine Biardeau, "Les

Quatire Buts de L'Homme," p. 49 ).

(2) artha marks the quest for material satisfaction

to whatever degree, for mundane necessities or "material

advantage, social preferment, wealth, power" (Daniel H. H.

Ingalls, "Authority and Law in Ancient India," Supplement

No. 17 to the Journal of the American Oriental Society,

1954, p. 1).

(3) kāma is the pursuit of love, physical pleasure,

and sensual enjoyment. "Kāma is the enjoyment of

appropriate objects by the five senses of hearing, feeling,

seeing, tasting and smelling, assisted by the mind together

with the soul. The ingredient in this is a peculiar

contact between the organ of sense and its object, and the

consciousness of pleasure which arises from that contact is

called Kāma" (Vātsyāyana, The Kāma Sūtra of Vātsyayana,

translated by Sir Richard Burton and F. F. Arbuthnot (New

York: Capricorn Books, 1963 (1883), p. 65).

(4) mokṣa is the ultimate goal of final "release" and

freedom from all mundane attachment, culminating in a

cessation of the cyclic alternation of birth and death.

41. Margaret and James Stutley, Harper's Dictionary of

Hinduism, p. 121.

42. V. Raghavan, "The Four Ends of Man," in Sources of

Indian Tradition, edited by Wm. Theodore de Bary (New York:

Columbia University Press, 1960 (1958)), p. 244.

43. Sylvain Levi, Le Théatre Indien, vol. 1 (Paris: College de France. 1963), p. 62.

44. Sylvain Lévi, Le Théatre Indien, vol. 1, pp. 66-67. In the following sketch the French of Sylvain Lévi will generally be translated.

45. Sylvain Lévi, Le Théatre Indien, vol. 1, p. 64.

46. Sylvain Lévi, Le Théatre Indien, vol. 1, pp. 65-66.

47. A. B. Keith, The Sanskrit Drama in its Origin, Developnent, Theory and Practice (London: Oxford University Press, 1924, p. 305.

48. Dandin, Dandin's Dashakumara-charita: The Ten Princes, translated by Arthur W. Ryder (Chicago: The University Press, 1927), p. 32.

49. Dandin, Dandin's Dashakumara-charita, (1927), pp. 42-43.

50. Dandin, Dandin's Dashakumara-charita, (1927), p. 58.

51. Dandin, Dandin's Dashakumara-charita, (1927), p. 80.

52. Dandin, Dandin's Dashakumara-charita, (1927), pp. 60-61.

53. Sylvain Lévi, Le Théatre Indien, vol. 1, p. 72.

54. Siegfried Lienhard, "Typen der Nāyikā im Indischen Kāvya," Wiener Zeitschrift für die Kunde des Morgenlandes, 52 (1955), p. 389.

55. Dandin, Dandin's Dashakumara-charita: The Ten Princes, translated by Arthur W. Ryder (Chicago: The University of Chicago Press, 1927), pp. 46-47.

56. kalpa /"eon," a vast measure of time and essential

component of the Indic view of human existence as a linear

progression of decay and cyclical renewal moving through the

four yugas or fundamental ages: kṛta yuga, "the Golden age,

without envy . . . pride, hatred, cruelty or other vices.

All people belong to one caste . . . worship one deity,

have a single Veda . . . and are, without exception,

brāhmins. . ."; tretā yuga, "its chief virtue is

knowledge. The need for sacrifices and rituals begins to be

felt, and men now seek reward for their work"; dvāpara yuga,

"the main virtue is sacrifice, and only few adhere to duty

or truth for its own sake. Disease, misery and calamity

begin and the castes come into existence"; and kali yuga

(the present age), "true worship and even sacrifice have

ceased. . . . Men live to variable ages and few see a

century of summers. It is a time of anger, hatred, lust,

greed, passion, pride, strife, discord. There is universal

viciousness and weakness" (the kali yuga ends in fire and

flood, the cycle after a time recommencing with the kṛta

yuga).

According to one of the more common methods of

calculation, the duration of all four yugas = 1 mahā yuga

= 12,000 "god years" (where one god-year = 360 solar years)

= 4,320,000 solar years ; 1,000 mahā yugas = 1 ardha kalpa

(one-half of a kalpa) = 4,320,000,000 solar years = 1 day

or 1 night of Brahmā. Two ardha kalpas = 1 kalpa =

8,640,000,000 solar years = 1 "day" of Brahmā.

Brahmā lives for one hundred of his "years" (the

present Brahmā apparently being now 51). His death sets

off an apocalypse embracing the universe, the mahā

pralaya/"great chaos," which destroys all gods, demons, and

the whole cosmos." A period of emptiness, wherein chaotic

energy is gradually dissapated, follows equivalent to a

lifetime of Brahmā's, at the end of which another Brahmā

appears -- "And so the cycles are continued, ceaselessly

and without end" (This sketch of time and its divisions is

drawn from Benjamin Walker, Hindu World, vol. 1, pp. 6-8).

Margaret and James Stutley comment (with that

wonderful English touch for understatement), "Owing to the

development of two separate time-systems, attempts to

coordinate them has led to some confusion over the division

and duration of the successive periods of the world's

existence" (Margaret and James Stutley, Harper's Dictionary

of Hinduism, p. 139).

57. S. K. De, "The Ākhyāyikā and the Kathā in Classical

Sanskrit," Bulletin of the School of Oriental Studies,

London Institution, 3 (1923-25), p. 512.

58. A. K. Warder, Indian Kāvya Literature, vol. 4, (1983).

p. 165; and "Classical Literature," in A Cultural History

of India, edited by A. L. Basham (Oxford: Clarendon Press,

1975), p. 185.

59. Sylvain Lévi, Le Théatre Indien, vol. 1, (1890), p.

61. Louis Renou, "La Réflexion sur la Poésie dans L'Inde, "

in Sanskrit et Culture: L'Apport de l'Inde a la

Civilisation Humain (Paris: Payot, 1950), p. 140.

61. Sylvain Lévi, Le Théatre Indien, vol. 1, (1890),

pp. 140-45.

62. Sylvain Lévi, Le Théatre Indien, vol. 1, (1890),

pp. 140-41.

63. Sylvain Lévi, Le Théatre Indien, vol. 1 (1890),

p. 143.

64. Bharata, The Nāṭyaśāstra, translated by Manomohan

Ghosh, vol. 1, rev. second edition (Calcutta: Manisha

Granthalaya, 1967 (1951), p. 366.

140

65. Sylvain Lévi, Le Théatre Indien, vol. 1, (1890), p. 143.

66. Sylvain Lévi, Le Théatre Indien, vol. 1, (1890), p. 145.

67. Bharata, The Nāṭyaśāstra, translated by Manomohan Ghosh, vol. 1, (1967), p. 370.

68. Sylvain Lévi, Le Théatre Indien, vol. 1, (1890), p. 144.

69. A. B. Keith, History of Sanskrit Literature, (1920), p. 985.

70. H. Oldenberg, "Das altindische ākhyāna, mit besondrer Rücksicht auf das Suparnākhyāna," Zeitschrift der Deutschen Morgenländischen Gesellschaft, 37 (1883), pp. 54-86; "Ākhyāna-Hymnen im Ṛgveda," Zeitschrift der Deutschen Morgenländischen Gesellschaft, 39 (1885), pp. 52-90.

71. A. B. Keith, "The Vedic Akhyana and the Indian Drama," Journal of the Royal Asiatic Society, 1911, p. 983.

72. Sylvain Lévi, Le Théatre Indien, vol. 1 (1890), p. 307.

73. Sylvain Lévi, Le Théatre Indien, vol. 1 (1890), pp. 301-8.

74. Louis Renou, L'Inde Classique, vol. 1 (1947), p. 260.

75. A. B. Keith, "The Vedic Akhyana and the Indian Drama," (1911), pp. 984-85.

76. Louis Renou, L'Inde Classique, vol. 2 (1954), p. 240.

77. Louis Renou, Histoire de la Langue Sanskrit (Lyon: Editions IAC, 1956), pp. 150-51.

78. J. A. B. van Buitenen, "Classical Drama: Background and Types," in The Literatures of India, (1974), p. 89.

79. Louis Renou, Histoire de la Langue Sanskrit, (1956), p. 161.

80. A. B. Keith, History of Sanskrit Literature, p. 333.

81. S. K. De, "The Campū," The Journal of the Ganganatha Jha Research Institute, 1 (1943), p. 58.

82. Siegfried Lienhard, A History of Classical Poetry, p. 267.

83. S. K. De, "The Campū," (1943), p. 57.

84. Siegfried Lienhard, A History of Classical Poetry, p. 266.

85. Louis Renou, L'Inde Classique, vol. 2, (1953), p. 203.

86. S. K. De, "The Campu," ((1943), p. 56.

87. Siegfried Lienhard, A History of Classical Poetry, p. 266.

88. Louis Renou, L'Inde Classique, vol. 2, (1953), p. 258; A. B. Keith, History of Sanskrit Literature, (1920), p. 332; M. K. Suryanarayana Rao, "Origin and Development of Campūs," in Felicitation Volume Presented to V. V. Mirashi, pp. 175-88, edited by G. T. Deshpande, et al. (Nagpur: Vidarbha Samshodhan Mandal, 1965); and so on.

89. Siegfried Lienhard, A History of Classical Poetry, (1984), p. 265.

Language and Kāvya

"The authorities declare that literature is

alternately fourfold -- whether in Sanskrit, Prākṛta / and

similarly in Apabhraṃśa or a mixture" [ tadetadvāñmayam

bhūyaḥ saṃskṛtam prākṛtaṃ tathā | apabhraṃśaśca misraṃ

cetyāhurāryāścaturvidham ||]. [1.32]

"Sanskrit is the language of the gods / later employed

by great sages / The classification of Prākṛta is

threefold: Indirectly derived; directly borrowed; or

indigenous" [ saṃskṛtaṃ nāma daivī vāganvākhyātā maharṣibhiḥ

| tadbhavastatstamo deśītyanekah prākṛtakramaḥ || [1.33].

Dandin would seem to be giving "Prākṛta" a rather wide

sense. And it is unclear if he is referring (as he seems to

be) to the various Prākṛtic dialects, or to the origins of

the Prākṛtic lexicon. As the previous verse [1.32] shows,

he is not writing of the Prākṛta vernaculars, but of Prākṛta

as a vehicle for literature. Although I reject the use of

the term "artificial," Jules Bloch is most probably correct

in seeing an increasing divergence between the "literary"

Prākṛtas and the vernaculars:

Prakrit literature was, from the start, a relatively learned production and continued till a very late date, becoming more and more artificial. It is not yet dead, any more than Sanskrit. It is easy to imagine that its deviation from the tongues in general currency became more and more noticeable. Normally the forms of the words could be taken from the Sanskrit, source of all culture, but gradually words of ungrammatical meaning or appearance had, as in Sanskrit, slipped in among them.1

This divergence would imply not only that these Prākṛtas would have to be consciously learned (as with Sanskrit), but also that they would become more of an "object" of study, amenable to this type of analysis. "The term deśī is applied to those words in Prakrit which are derived from no Sanskrit equivalent. The number of such words which can be explained out of Dravidian or some other source is comparatively small and will probably always remain so. . . . On the whole classical Sanskrit avoids

such words, but a number are incorporated, and in

particular the Jain writers have adopted a fair number."2

Yet this group is perhaps more accurately seen as a

catch-all category, where the status of an individual word

may primarily depend on the etymological expertise (or

creativity) of the individual(s) actually doing the

categorizing:

The Indians include under the Deśya or Deśī class

very heterogeneous elements. They consider all

such words to belong to this class that they

cannot trace back to Sanskrit either in form or in

meaning. It depends upon their knowledge of

Sanskrit and ability in etymologising that some of

them call a word to be Deśya, while others include

it either among the tatsamas ["directly borrowed"]

or among the tadbhavās ["indirectly derived"]. So

we have many words that are classed as Deśī, even

though they go back to genuine Sanskrit roots,

simply because they do not have closely

corresponding Sanskrit

words. . . .3

In Bharata's Nāṭyaśāstra, for example, we find a

threefold classification similar to that of Dandin's, but

with different names, and appearing to pertain to words

145

(śabdas): [18.3] "One should realize that this [recitation

of prākṛt] is threefold in theatrical practice: samāna

śabda [tatsama]; vibhraṣṭa [tadbhava]; and deśī" [trvidha m

tacca vijñeyam naṭyayoge samāsataḥ | samānaśabdaṃvibhraṣṭaṃ

deśīgatamathāpi ca ||]. That we are presented with

equivalents of Daṇ̣in’s three categories is expressed by

Luigia Nitti-Dolci in his comments on Bharata’s verse, "Les

noms ordinaires des trois c tégeries de mots respetives

sont chez les grammairiens tatsama, tadbhava et desya."4

Recent writers have understood these three terms to

pertain primarily to the word borrowing of "modern Indo-

Aryan." Thus J. F. Staal, for example, sees these as: (1)

tadbhava /"words which have developed from Sanskrit via

Middle Indo-Aryan into Modern Indo-Aryan; (2) tatsama

/"words which are borrowed directly from Sanskrit"; and (3)

deśī /"words of Indian but non-Indo-Aryan origin."5

Thomas Burrow, however (in light of Daṇ̣in’s verse),

would seem to be incorrect in rejecting the tatsama

category of Prākṛta word incorporation: "An important new

feature in the modern languages, as opposed to the earlier

Middle-Indo Āryan, was the introduction, on an extensive

scale, of Sanskrit loanwords. In Prākrit, even at the

Apabhramśa stage, words might in fact be derived from

Sanskrit, but they always appeared disguised as Prākrit by

the operation of phonetic rules."6

"The speech of Mahārāṣṭra is known as the best Prākṛta

/ Its nature is seen in such texts as the Setubandha -- an

ocean of jewels of beautiful expressions" [ mahārāṣṭrāśrayāṃ

bhāṣāṃ prakṛṣṭaṃ prākṛtaṃ viduḥ | sāgarah sūktiratnānāṃ

setubandhādi yanmayam || [1.34].

The specific basis for the elevation of Māhārāṣṭrī as

the "best" (prakṛṣṭa) of Prākṛtas is unresolved. Two

viewpoints are generally held. Richard Pischel considers

Māhārāṣṭrī to be linguistically closer than any of the

other Prākṛtas to Sanskrit, the "perfected" standard, and

thus proportionately superior. This basis "is explained by

the fact that Māhārāṣṭrī is considered to be nearest to

Sanskrit. When the Indians speak simply about Prākrit,

they almost always thereby mean Māhārāṣṭrī. According to

them, Māhārāṣṭrī has the credit of being the basis of the

other Prākrit languages [he cites the Prākrtasarvasva of

Mārkaṇḍeya Kavīndra], and in the manuals written by the

native grammarians Māhārāṣṭrī occupies the first place."7

He notes that the Prākrtaprakāśa of Vararuci, one of the

oldest Prākṛta grammarians, devotes nine chapters to

Māhārāṣṭrī, and but one to each of the three other Prākṛtas

examined (Śaurasenī, Māgadhī, and Paiśācī).

Alternately, Luigia Nitti-Dolci, citing this verse in

support, affirms that "Daṇḍin does not consider giving a

linguistic classification: Māhārāṣṭrī is the best Prākṛt

because it is the one that has the richest literature. As

for the explanation that Māhārāṣṭrī was the best Prākṛt

because it was closer to Sanskrit, it is frankly

unacceptable; no Indian grammarian has ever expressed such a

heresy. On the contrary, Śaurasenī was for them, as for

us, closer to the source. . . ."8

148

Although not questioning the importance of Māhārāṣṭrī,

he notes that Vararuci considers Sanskrit to be the basis of

Śaurasenī (Prākṛtaprakāśa [12.2]), and Śaurasenī to be the

basis of both Paiśācī [10.2], and Māgadhī [11.2].

Nitti-Dolci stumbles, however, in his interpretation of

Daṇḍin, and provides yet another example of mistranslation

either generating or stemming from a predisposed position.

He translates Kāvyādarśa [1.34] as: "They consider that

the best prākṛt is the language spoken in the Mahārāṣṭra

country: an ocean of beautiful expressions -- such pearls!

-- in which the Setubandha and other poems have been

composed."9

The mistranslation lies in applying the "ocean of

beautiful expressions" to Māhārāṣṭrī, thus giving greater

stress than is warranted to his position. One could infer

that Daṇḍin considers Māhārāṣṭrī to be the "best" of

Prākṛtas due to the existence of such examples as the

Setubandhu and other such literary works, but there is no

certainty.10

Setubandha / The Setubandha ("The Building of the

Bridge") or Rāvanavaha ("The Killing of Ravana"), also

called in manuscripts the Daśamuhavaho or Rāmasetu, is

written as Danḍin indicates in Māhārāṣṭrī Prākṛta and in the

style of a padya mahākāvya. It is divided into fifteen

āśvāsas (or chapters) running to 1362 stanzas, and exists in

two primary recensions.11

The text relates the tale of Rāma and his monkey army,

their glorious bridge spanning the ocean to Śrī Laṅkā,

their fight and victory over Ravana and his demon hordes,

and the eventual rescue of Sītā. It is usually attributed

to a Pravarasena, who is in all probability the Vākāṭaka

King Pravarasena II [5th century] (and who may have written

the work with the assistance of Kālidāsa, though this is

highly speculative).12 I would tend to disagree with those

scholars who believe that the author of the Setubandha was

rather a King Pravarasena of Kashmir.13

"Śauraseni, Gauḍī, Lāṭī and others similar / Come to

mind when one says 'Prākṛta' [ śaurasenī ca gauḍī ca lāṭī

cānyā ca tādrśī | yāti prakṛtamityevam vyavahāreṣu samnidhim

||] [1.35].

"In kāvyas the speech of cowherders and so on is

referred to as 'Apabhraṃśa' / In śāstras however anything

other than Sanskrit is called Apabhraṃśca" [ābhīrādigiraḥ

kāyeṣvapabhraṃśa iti smṛtāḥ | śāstreṣu saṃskṛtādanyad

apabhraṃśatayoditam ||] [1.36].

"Sargabandhas and so on are in Sanskrit. . . . Nāṭakas

and so on employ a mixture [ of Sanskrit, Prākṛt, and

Apabhraṃśa" [ saṃskṛtaṃ sargabandhādi . . . | . . . nāṭakādi

tu miśrakam ||] [1.37].

"A kathā is composed in any spoken language (bhāṣā)

and in Sanskrit. . . ." [kathā hi sarvabhāṣābhiḥ saṃskṛtena

ca badhyate |] [1.38ab].

Notes [1.32] - [1.38]

1. Jules Bloch, Indo-Aryan: From the Vedas to Modern

Times, English editon revised by the author and translated

[from the French] by Alfred Master (Paris: Librairie

D'Amérique et D'Orient, 1965), p. 20.

2. Thomas Burrow, The Sanskrit Lanuage, 3rd rev. edition

(London: Faber and Faber, 1973 (1955), p. 57.

3. Richard Pischel, A Grammar of the Prākrit Languages,

translated from the German by Subhadra Jhā, 2nd rev. edition

(Delhi: Motilal Banarsidass, 1981), pp. 7-8.

4. Luigia Nitti-Dolci, Les Grammairiens Prakrits, (Paris:

Libraire D'Amerique et D'Orient, 1938), p. 70, n. 6.

5. J. F. Staal, "Sanskrit and Sanskritization," Journal of

Asian Studies, vol. 22, n. 3 (1963), pp. 261-75.

6. Thomas Burrow, "Ancient and Modern Languages," in A

Cultural History of India, edited by A. L. Basham (Oxford:

Clarendon Press, 1975), p. 166.

7. Richard Pischel, A Grammar of the Prākrit Languages, p.

9. From the French of Luigia Nitti-Dolci, Les Grammairiens

Prakrits, p. 2.

9. From the French of Luigia Nitti-Dolci, Les Grammairiens

Prakrits, p. 2.

10. It is curious that Siegfried Lienhard also

mistranslates this verse, arriving at yet another reading

and one that again provides presumed evidence for a

preconceived view. One of Lienhard's central assumptions

is that the individual stanza is essentially the source or

ground for all later developments in kāvya. His mistranslation is thus used to support the view that even in the sargabandha the stanza is "the centre of interest to poet and reader or listener" (Siegfried Lienhard, A History of Classical Poetry, p. 160).

Thus, although correctly having the Setubandha as the focus in the second half of the verse, he incorrectly translates sāgarah sūktiratnānam as "a sea of jewel-stanzas" (p. 161, n. 5). I feel that "su (+)ukti" simply means "beautiful" or "well-turned expressions", rather than indicating the padya (stanzaic) form as such.

11. (1) Rāvanavaha oder Setubandha: Prākrt und Deutsch Herausgeben, by S. Goldschmidt, 2 vols. (Strassburg and London, 1880 and 1884). (2) Edited by Pt. Śivadatta and K. P. Parab (Bombay: Nirnaya Sāgar Press, 1895).

12. See: S. K. Aiyangar, "The Vākāṭakas and their Place in the History of India," Annals of the Bhandarkar Institute, Poona, 5 (1925), pp. 31-54.

K. S. Ramaswami Shastri, "King Pravarasena and Kālidāsa," Proceedings and Transactions of the Seventh All-India Oriental Conference, Baroda (12/1933) (Baroda: Oriental Institute, 1935),`pp. 99-108. Shastri's article floating along in a logical wonderland, provides a wonderful circular argument in support of the view: "The validity of the tradition which is recorded by a recent commentator [Rāmadāsabhūpati, writing some 300 years previously], and which seems to contradict the statement of early authors such as Dandin, Bāṇa and Kṣemendra, cannot be questioned, because in reality the work was attributed to Pravarasena by Kālidāsa at the request of Vikramāditya" [as recorded by a recent commentator].

V. Raghavan, "Kālidāsa's Kuntalesvara Dautya," in B. C. Law Volume, edited by D. R. Bhandarkar, et al., Part 3 (Poona: The Bhandarkar Oriental Research Institute, 1946), pp. 191-97. V. V. Mirashi, "The Vakataka Chronology," Indian Historical Quarterly, 24 (1948), pp. 148-55. A. D.

PusalKer, "Identity and Date of Pravarasena, the Author of

the Setubandha," Journal of the Asiatic Society of Bombay,

31 and 32 (1956-57), pp. 212-17.

13. Eugeniusz Sluszkiewicz, for example, states, "le

Rāvaṇavaha ou Setubandha épopée prākrite composée par

Pravarasena II du Cachemire ou par un poète de sa cour. . . .

." ("La Rāvaṇavaha et le Rāmāyaṇa," Rocznik

Orientalistyczny, 16 (1950), p. 545); or most recently with

Siegfried Lienhard, "King Pravarasena II was probably the

successor of King Mātrgupta" of Kashmir (A History of

Classical Poetry, (1984), p. 197).

The Ten Gunas (or "Qualities") and the Mārgas (or "Styles")

It is essential to realize that Daṇḍin's conception of

alaṃkāra goes beyond his extensive, "figurative"

presentation. That paralleling the "conceptual" or artha

alaṃkāras of the Second Chapter, and the "phonemic" or

śabda alaṃkāras of the Third, we have a third category

whose members are at once more specific in their range of

application and yet more vague in their mode of operation.

That just as the artha and śabda alaṃkāras are applicable

to both of the primary (and extreme) "modes" or mārgas of

literary expression -- and thus "general"/sādhāraṇa -- so

the guṇa alaṃkāras and their transformations may be

considered "specific"/viśiṣṭa to a particular mode.

Daṇḍin's presentation of the ten guṇas or "qualities"

and their transformations as characteristic of either the

Vaidarbha (that is, "Southern") or Gauḍīya (that is,

"Eastern") mārgas ("paths," "styles") compries most of the

Kāvyādarśa's first chapter. The discussion is initiated in

155

[1.40]: "The mārgas are manifold / with branches mutually

and subtly distinct / Among these the Vaidarbha and the

Gauḍīya -- distinct extremes / shall be described"

[ astyāneko girāṃ mārgah sūkṣmabhedaḥ parasparam | tatra

vaidarbhagaudīyau varṇyete prasphuṭān-tarau ||].

"śleṣa prasāda samatā mādhurya sukumaratā / arthavyakti

udāratva ojas kānti and samādhi" [ śleṣaḥ prasādaḥ samatā

mādhuryaṃ sukumaratā | arthavyaktirudāratvamojahkānti

samādhayaḥ ||] [1.41].

"These ten guṇas are traditionally accepted as the

life breaths of the Vaidarbha path / Transpositions of

these are often seen in the Gauḍa mode" [ iti vaidarbha-

mārgasya prāṇā daśa guṇāḥ smṛtāḥ | eṣāṃ viparyayaḥ prāyo

dṛśyate gaudavartmani ||] [1.42].

Viparyaya ("transposition") is taken by some to mean

"opposite," but our analysis of these two mārgas in

relation to the guṇas will only support this reading

perhaps in two or three cases. Upon examining the

characteristics of the Gauḍīya style it is clear that

viparyaya does not mean strictly "opposite" [ < vi (+) pari

(+) *ī /"reverse," "opposite of"]; but also "transposi-

tion," "alteration of." S. K. De considers that viparyaya

"does not mean vaiparitya or contrariety (as the

Hrdayamgama commentary takes it), but athātva or

divergence."1

Mārga (literally, "path" or "way") is frequently

translated as "style," with the Vaidarbha and the Gauḍīya

thus assumed to represent kāvya's two primary and distinct

styles. "The denseness and elaborateness which Daṇḍin

associates with the Gaudi riti, the relative simplicity and

directness of the Vaidarbhī riti are constant poles in the

best Sanskrit writing. . . ."2 The very vagueness of

"style" does indeed allow it to approximate "mārga." Here

we have a not quite technical term where the degree of

conceptual overlapping is perhaps sufficient to allow

translation, but we should be aware of the limitations.

Very loosely, "style" generates two broad connotations:

style as the distinctive and (perhaps) unique expression of

a given writer (de Buffon's "Le style c'est l'homme même);

and style as a "characteristic mode of construction and

expression" (nicely marked by De Quincey as "the management

of language"). S. K. De's objection to this presumed

equivalence reduces style to but one of these senses: "It

should be observed that the term Riti [which De, among

others, rather loosely considers interchangable with mārga3]

is hardly equivalent to the English word "style," by which

it is often rendered but in which there is always a distinct

subjective valuation."4

With with rejection of "style" envisioned only in its

"subjective" sense, De's conception of mārga as "objective"

follows with seemingly inescapable -- however illusionary

-- logic. Thus rīti becomes "the outward presentation of

[kāvya] called forth by a harmonious combination of more or

less fixed literary 'excellences' [guṇas] ."5

And we should be wary of placing too much emphasis upon

Dandin's usage of "mārga" as a technical term. Further

paralleling "style" with its somewhat indeterminate sense,

it is really more of a convenient pointer to a conceived

way of doing things. Daṇdin readily interchanges words

whose connotations are fundamentally the same. In [1.42],

for example, we find the "vaidarbha mārga" but the "gauda

vatman" (vatman similarly meaning "path" or "way"); or

again in [1.50], where the Gauḍīya is referred to as "the

path of kāvya pertaining to the East"/paurastyā

kāvya-paddhatiḥ |

It is the Vaidarbha mārga then that Daṇdin sees

displaying -- as its "life breaths" -- the ten guṇas, and

is thus presumed to be a favored standard. Alternately

(and perhaps just as evident as the Vaidarbha in practice),

the Gauḍīya style "often"/prāyas -- not exclusively --

displays what may be considered "transpositions"

(viparyayāḥ). "The word prāyas . . . is important in this

connection. The characteristics of these two types of

poetry often differ but sometimes they agree. The Gauḍa

Mārga sometimes presents opposites of and deviations from

159

the excellences prevailing in the Vaidarbha, but qualities

[gunas] such as Samādhi, Arthavyakti, Audārya, Mādhurya

and Ojas are more or less common to both the Mārgas."6

There is a further and not necessarily correct

conclusion which might be drawn given a conception of guṇa

as "excellence": "If it is asked what constitutes the

essential characteristics of the Gauḍa Mārga, we cannot

reasonably answer that the opposites of these excellences

(which would really be Doṣas or faults) do it. . . . A

transposition of a given guṇa does not -- necessarily --

entail fault. Daṇḍin, whatever we might infer of his own

stylistic preference, certainly views the Gauḍīya style as a

valid mode of kāvya. As Gerow remarks, "The importance of

the guṇas lies in their service as characteristics, as

'plus-features,' of poetry whose alternate is not

necessarily non-poetry. In other words, the contrary of a

guṇa may be and usually is another feature whose presence

marks another kind of poetry."8 Let us consider then

Daṇḍin's elucidation of the ten guṇas.

(1) [1.43-44] śliṣṭa (śleṣa) /"compactness":

"Śliṣṭa [śleṣa] is devoid of looseness / This laxity

is marked by a profusion of non-aspirated syllables

(alpaprāna-aksara)" [ śliṭamasprṣṭasaithilyamalpa-

prāṇākṣarottaram | śithilam. . . . ||] [1.43]. These are

"Unaspirated letters which require little effort in

pronouncing, or more technically, the first [ k / c / ṭ /

t / p ] and third [ g / j / ḍ / d / b ] (non-conjunct)

letters of each varga, and the semivowels [ y / r / l / v ]

and nasals [ ṅ / ñ / ṇ / n / m ], the rest being mahāprāna

syllables."9 As in, for example, [1.43cd] "mālatīmālā

lolālikalilā" ("The garland of Mālatī flowers covered with

swarming bees").

"This is accepted by the Gauḍas in light of anuprāsa

("sound manipulation") [ anuprāsadhiyā gauḍaistadiṣṭam . . .

.| ] [1.44ab]. Thus [śleṣa] "to the Gauḍas is a preferable

excellence of diction inasmuch as it gives more scope to

alliteration [anuprāsa]."10 "And [it is accepted by the

161

Vaidarbhas due to the density of construction" [ bandha-gauravāt | vaidarbhair. . . .||] [1.44bc].

(2) [1.45-46] prasāda /"clarity," "lucidity":

"[A phrase] possessing prasāda displays a meaning commonly known . . ." [ prasādavat prasiddhārtham . . . .

|] [1.45a].

"The Gauḍīyas accept even [words] not commonly known / whose meanings reflect their etymology" [vyutpannamiti

gauḍīyairnātirūḍhamapiṣyate |] [1.46ab].

As Belvalkar and Raddi comment, "Its requires a very great self-restraint . . . not to let one's learning in the

śāstras unseasonably intrude itself into poetry" (Notes 1/45).

(3) [1.47-50] samatā/ "smoothness":

"Sama [samatā] is the absence of disparity in syllabic collocations / These constructions are soft (mrdu), harsh (sphuṭa), or in-between (madhyama) as their bases are an

ordering of letters that themselves are soft, harsh, or in-between" [ samaṃ bandheṣvaviṣamaṃ te mrduṣphuṭamadhya-

162

māḥ | bandhā mr̥dusphuṭonmiśravarṇavinyāsayonayaḥ ||] [1.47].

As in, for example, (harsh:) spardhate ruddhamaddhairyo

/ (soft:) vararāmāmukhānilaiḥ ("The Malay breeze blocking

my courage competes with the breath from the mouth of that

choice lady") [1.49cd].

"Not considering this disparity / and with an eye to a

show of meanings and alamkāras / The path ["paddhatih"

rather than "mārga" is employed] of Eastern kāvya [the

Gauḍīya] has grown" [ ityālocya vaiṣamyamarthālamkā

raḍambarau | avekṣamānā vavṛdhe paurastyā kāvyapad-

dhatīḥ ||] [1.50]. "The Gauḍas, we are told, admit such

compositions (even though they lack uniformity) for the

sake of richness of ideas and Alamkāras. . . ."11

(4) [1.51-68] mādhurya/ "elegance," "sweetness":

"Madhura [mādhura] reflects the possession of rasa /

and rasa exists in both sound (vāk) and sense (vastu) /

[Rasa] through which the connoisseur becomes drunk / like

the bee through honey" [ madhuram rasavadvāci vastunyapi

163

rasasthitih | yena mādyanti dhīmanto madhuneva madhuvratāḥ

|| [1.51].

Dandin's conception of rasa within the guṇas is quite

specific and should be distinguished from that of rasa in

its more usual technical role (see under [2.279]): "It

appears that Dandin means by the term Rasa in the mādhurya

guṇa to connote the absence of vulgarity . . . and does not

contemplate the inclusion of Rasa in the technical

sense."12

This is clearly indicated during his later

presentation of rasavat alaṃkāra where he remarks in

[2.292]: "Rasa was presented in the context of mādhurya

guṇa / as the absence of vulgarity in expression / Yet here

the fact that the words display rasa / stems from the eight

rasas themselves." Yet its range of meaning here is

probably somewhat wider. Dandin's presentation of mādhurya

guṇa, the role that rasa (in this vague alternate sense)

plays, its relationship to anuprāsa (repetition of sound

patterns), and the rationale for the latter's inclusion

here rather than among the śabda alamkāras are ample

grounds for bemusement.

We might add that Dandin's inclusion of vāk and vastu

in this verse is his only explicit acknowledgement of a

distinction between śabda/"sound" and artha/"sense" within

the guṇas -- a distinction formally developed later by

Vāmana and incorporated within the tradition from that

point onwards.

Dandin basically categorizes mādhurya from the

perspective of both vāk rasa or "sound," and vastu rasa or

"meaning" - the commentator Taruṇavācaspati glosses vāk

rasa as "śabda mādhurya" (under KD [1.52]), and vastu rasa

as artha mādhurya" (under KD [1.62]).13 In [1.52-60] we

have verses concerned with the realization of mādhurya

through sound, in this case, through anuprāsa.

"Whatever one experiences as similar in sound -- / a

juxtaposition of words displaying this feature / and

possessed of anuprāsa / generates rasa" [ yayā

kayācicchrutyā yat samānamanubhūyate | tadrūpā hi padāsattiḥ

sānuprāsā rasāvahā ||] [1.52].

Dandin distinguishes two varieties of anuprāsa (and

thus of vāk rasa), śruti anuprāsa and varṇāvṛtti anuprāsa.

Anuprāsa is usually treated by later authors as one of the

primary śabda alaṅkāras. Given that Daṇḍin (or again,

perhaps the ambient tradition from which he drew) would

choose to include it among the guṇas, it is clear that

anuprāsa is seen as one of the most evident and important

variables that may serve to distinguish mārga as such.

Anuprāsa is generally translated as "alliteration," and

although here the technical correspondence is close, the

English term does not cover one of the two primary usages

of the term (for Daṇḍin). And also, once having used

"alliteration" in this sense, one will be left rather up in

the air when attempting to translate (if one feels it

necessary) its near relative, yamaka (as reflected, for

example, in Gerow's "cadence"). The conceptions of both of

these terms are quite straitforward, and again I feel less

distortion will occur if we stay with their actual names.

In verse [1.52] and the immediate verses following,

Dandin is concerned with śruti anuprāsa, that is, the

repetition of sounds categorized according to their place

of physical articulation (sthāna). There are five primary

sthānas,14 and thus five groups or "vargas" into which

consonants or vyañjanas (requiring a vowel to be

pronounced), and vowels or svaras (which do not require any

other letter to be pronounced) are divided. The five

vargas are: kaṇṭhya (guttural), tālavya (palatal), mūrdhanya

(cerebral), dantya (dental), and oṣṭhya (labial).

Dandin provides an example in [1.53] (which I have

broken into pādas) :

(1) eṣa RāJa YaDā Lakṣmīṁ

(2) pRāptavān bRāhmaṇapriyaḥ

(3) TaDāPraBHṛTi DHarmaSya

(4) LokeSminnuTSavobhavat

167

("When this king -- dear to brahmins -- realized

prosperity / From then on there was a festival of dharma in

the world").

Śruti anuprāsa is thus displayed in the first pāda by

the repetition of [ṣ] and [r] as mūrdhanya consonants, by

[j] and [y] as tālavya consonants, and by [d] and [l] as

dantya consonants; in the second by [r] and [ṇ] as

mūrdhanya consonants; in the third by [t], [d], [dh], and

[ṣ] as dantya consonants, and by [p] and [bh] as oṣṭhya

consonants; and in the fourth by [t], [l], and [ṣ] as

dantya consonants. Obviously the English "alliteration"

does not cover this primary usage (the semantic fit is too

small), nor would "repetition" be appropriate (the fit is

too large).

Dandin then draws in the attitude of the practicioners

themselves to this aspect of mādhurya guna: "This [śruti

anuprāsa] is not respected by the Gauḍas / although

anuprāsa is dear to them / Due to the presence of anuprāsa /

This is generally accepted by the Vaidarbhas" [ itīdam

nādṛtaṃ gauḍairanuprāsastu tatpriyaḥ | anuprāsādapi prāyo vaidarbhairidamīpsitam || [1.54].

P. C. Lahiri comments, "It involves an economy of

effort in articulation, and thereby gives a special pleasure

to the Vaidarbhas, who avoid, for fear of incurring

monotony, mere varṇānuprāsa [1.55ff.] or the alliteration

consisting of repetition of similar [the same] letters."15

Alternately, varṇavṛtti anuprāsa is the repetition of

identical, isolated letters, and may be accomplished either

by letters in two different pādas (pāda varṇāvṛtti), or

between letters in the same word or different words within

the same pāda (pada varṇāvṛtti).

"Varṇāvṛttianuprāsa (the "repetition of letters")

whether at the level of the pādas ("quarter-verses"), or at

that of the pada ("word") is anuprāsa provided there is

sufficient juxtaposition of sounds [such that the latter

sound] arouses the impression (saṃskāra) left by the

previous [sound]" [ varṇāvṛttiranuprāsaḥ pādeṣu ca | pūrvānubhavasamskārabodhinī yadyadūrata ||] [1.55].

169

Varnāvṛtti anuprāsa between pādas is illustrated in

[1.56]:

(1) caNDre śaranniśottamse

(2) kuNDastabakavibhrame

(3) NDraniLanibhaṃ lakṣma

(4) saNDadhátyalinaḥ śriyam

("That mark like a saphire on the moon -- the crown

ornament of the autumn night displaying the brilliance of

bunches of Kunda flowers -- has the charm of the

bumblebee"). Here the conjunct [ -nd- ] in each pāda marks

anuprāsa.

Where varnāvṛtti anuprāsa at the level of the pāda or

word is illustrated in the following [1.57]:

(1) Cāru Candramaśambhiru

(2) Bimbaṃ paśyaitadambare

(3) ManMano ManMathākrāntam

(4) NirDayaṃ haNtumuDyatam

("Timid one! See this delightful lunar disc / arisen in

the sky to mercilessly torture my mind / laid low by love").

And in the first-half of [1.59] we have an example of

anuprāsa involving excessive "harshness" (pāruṣya) :

(1) smaraḥ KHaraḥ KHalaḥ Kāntāḥ

(2) Kāyaḥ Kopaśca naḥ Kṛśaḥ

Here there is excessive use of the visarga [ -ḥ ],

which is considered to be hard, as well as a number of

repetitions of [ k ] and [ kh ], both of which are harsh.

Where in the second-half we find excessive "slackness"

(śaithilya) :

(3) cYuto MāNodhiko RāGo

(4) MoHo JātosaVo Gatāḥ

Now the visarga is replaced by the "soft" vowel [ o ]

and joined by number of soft consonants, [ y ], [ m ], [ n

], [ r ], [ g ], [ h ], [ j ], and [ v ]. And as Daṇḍin

points out in [1.60], such extremes of anuprāsa are not

employed by the "Southerners" (dākṣiṇātyas), that is, by

the Vaidarbhas.

Thus in a general sense both mārgas accept mādhurya

guṇa from the perspective of vāk rasa (śabdamādhurya). The

Vaidarbhas favor its realization through śruti anuprasa and

reject excessively harsh or loose collocations. The

Gaudīyas, however, favor varṇa anuprāsa provided there is

an effective balance, that is, with the repetitive letters

neither too close nor too far apart.

Before presenting vastu rasa, Daṇḍin briefly mentions

yamaka. Yamaka as with anuprāsa involves repetition, but

repetition rather of units or groups of letters. [1.61]:

"Repetition involving groups of letters is known as

yamaka, but as it is not exclusively sweet (madhura) it

will be covered later" (in [3.1-76]) [ āvṛttiṃ varṇasaṃghā

tagocarāṃ yamakam viduḥ | tattu naikāntamadhuramataḥ paścā

dvidhāsyate ||] [1.61]. As a basis for distinction this is

certainly vague -- again we have repetition and what we are

to understand by "madhura" is not at all clear. As a śabda

alamkāra, yamaka is more structured in its varied,

predesigned patterning, and thus perhaps more amenable to

general use irrespective of the individual mārga.

In verses [1.62-68] on the second primary category of

mādhurya guṇa, vastu rasa (or "artha mādhurya") we are

concerned with the sense conveyed, or more specifically,

with the degree of elevation and the avoidance of mundane

vulgarity that kāvya demands.

"Surely every alamkāra sprinkles rasa upon the meaning

/ Even so, just the absence of vulgarity carries this

burden to a large extent" [ kāmaṃ sarvopyalamkāro

rasamarthe niṣiñcati | tathāpyagrāmyataivainaṃ bhāraṃ

vahati bhūyasā ||] [1.62].

"Vulgarity (grāmyatā) results from stating the

opposite of what is refined. . . ." [ grāmyatā . . . sā

samyetarakīrtanāt |] [1.65ab]. Daṇḍin offers examples with

alternatives, upon which P. C. Lahiri comments, "In 1.63-57

two kinds of indecorous expression are distinguished. The

proposal in 1.63 is direct and therefore vulgar; in 1.64 it

is reached by implication and therefore taken as quite

decorous. In 1.66 words are used which, if united

together, give rise to a new word in Sanskrit by

combination, which conveys a vulgar meaning. In 1.67 the

words used, possessing more than one meaning, give rise to

an undesirable and indecorous suggestion."16 As this

attempt at refined or polished expression lies at the heart

of kāvya it is not surprising that "Even in both the mārgas

such [vulgar examples preceding] are not praised" [evamadi

na śaṃsanti mārgayorubhayorapi !] [1.67cd] .

As Belvalkar and Raddi conclude: "All definitions of

mādhurya from Bharata downwards agree in regarding it as a

subtle quality which one can feel but which defies all

analysis. . . . Vāgbhaṭa [II./14th century] in his

Kāvyānuśāsana tells us that mādhurya is what causes the

heart to melt in joy (yatha ānandamandam mano dravati)

(Notes 1/47-48).

(5) [1.69-72] sukumāratā/ "tenderness," "softness":

"[A phrase] that abounds in non-harsh letters is

174

considered to reflect sukumāra / Yet the defect of

looseness (śaithilya) in syllabic collocations -- where

all [letters] are soft -- has been shown" [ aniṣṭhurākṣara-

prāyam sukumārameṣyate | bandhaśaithilyadoṣopi darśitaḥ

srvakomale ||] [1.69].

Dandin has previously indicated [1.43] "looseness" as

the viparyaya (and here we may accept the meaning

"opposite") of śleṣa guṇa -- "where all letters are soft

(komala) -- that is, where there is a profusion or excess

of non-aspirated syllables (alpaprāna akṣara). And

concrete examples of both "harshness" (pāruṣya) and

"looseness" (śaithilya) have been offered in [1.59].

Sukumāratā, as with śleṣa guṇa, is generated and

marked by vocaic balance. The elements for each, although

to a degree overlapping, may yet be distinguished:

"Sukumāratā might have a chance of being confused with

śleṣa. To meet such an objection the commentator

Tarkavāgīśa remarks (under KD [1.69]) that the admixture of

alpaprāna and mahāprāṇa syllables constitutes śleṣa, whereas

175

sukumāratā consists in tenderness as a total effect arising

from the admixture of soft (komala) and harsh (paruṣa)

letters."17 With the balance in śleṣa tipped by an excess

of non-aspirated consonants we have "looseness"; with an

excess of strong or harsh consonants the balance of

sukumāratā fails and we have "harshness."

And just as the Gauḍīyas employ a degree of looseness

in view of anuprāsa, so they admit what might be seen as an

excessive element of harshness: "Whereas the Vaidarbhas

accept Sukumāratā in which expressions consisting of

unharsh vocables generally predominate, the Gauḍas have an

eye to a 'glaring composition,' and consequently they do

not mind if their poetry involves harsh vocables requiring

much strain for pronouncing them."18

(6) [1.73-75] arthavyakti /"explicit meaning":

"Arthavyakti reflects the absence of conjecture

over the meaning . . . ." [ arthavyaktiiraneyatvamarthasya

|] [1.73ab].

"Even both the mārgas do not think much of such a

phrase [a preceding example of neyatva or "opaqueness"

expressed in [1.74]] / For certainly an idea that leaps

beyond the principles of words is unfortunate" [ nedṛśam

bahu manyante mārgayorubhayorapi | na hi pratītiḥ subhagā

śabdanyāyavilāṅghini ||] [1.75].

(7) [1.76-79] udāratvam (udāra) /"magnificence":

"When [a phrase] is expressed a quality of

magnificence is perceived -- This is termed Udāra / The

path of kāvya has a protector in this" [ utkarṣavān guṇaḥ

kaścid yasminnukte pratīyate | tadudārāhvayaṃ tena sanāthā

kāvyapaddhatịḥ ||] [1.76].

And further, "Some accept that [phrases] displaying

praiseworthy (ślāghya) attributes reflect udāra"

[ ślāghyairviśeṣanairyuktamudāraṃ kaiścidīṣyate |] [1.79ab].

And as arthavyakti is accepted by both the Vaidarbha

and Gauḍīya, so "we can take it that this Guṇa is

entertained in both types of poetry in the absence of any

mention of the corresponding characteristics prevalent in

the Gauḍa mode."19

(8) [1.80-84] ojas /"power," "intensity":

"Ojas [stems from] an abundance of compounds / This is

the life of the prose form (gadya) / Yet even in verse

(padya) / it is the singular refuge of the non-Southerners

[the Gauḍīyas]" [ ojahaṃ samāsabhūyastvametadgadyasya jīvitam

| padyepyadākṣiṇātyānāmīdamekan parāyaṇam ||] [1.80].

"It displays a number of varieties / through the

profusion, lack and mixture / of either heavy or light

syllables. . . ." [ tad gurūṇāṃ laghūnāṃ ca bāhulyālpatva-

miśranaiḥ | uccāvacaprakāraṃ tad dṛśyamākhyāyikādiṣu ||]

[1.81abc].

"Thus even in verse / the Easterners [Gauḍīyas] employ

phrases abounding in ojas / But the others [Vaidarbhas]

desire ojas / in phrases where it is harmonious and

captivating" [ iti padyepi paurastyā badhnantyojasvīrīrgiraḥ

| anye tvanākulaṃ hrdyamicchantyojo giraṃ yathā ||][ 1.83].

"Ojas is one of the key-words of Indian culture. The

general idea expressed by this word is that of power, or . . .

. of 'power substance,' of a vital and magnetic energy

present in beings, in phenomena or things. . . . Ojas is

inherent in literary compositions rich in lofty and

sonorous words, with a stringent and compact rhythm, bound

together in long compounds. . . ."20

With ojas guna the usual roles have been somewhat

altered. Both of the primary styles employ it, but it is

especially characteristic of the Gaudīyas: "Ojas is

particularly a characteristic excellence with the Gauḍa

poets, who use it to any degree in any composition, while

the Vaidarbhas employ it with greater discretion. . . ."21

(9) [1.85-92] kānti /"grace":

"Kāvya possessing kānti -- an element seen even in

statements of fact (vārtā) and descriptions (varṇanā) --

without transgressing conventional meaning / is precious to

all the world" [ kāntaṃ sarvajagatkāntaṃ laukikārthā

natikramāt | tacca vārtābhidhāneṣu varṇanāsvapi dṛśyate || ]

[1.85].

In the introduction to Daṇḍin's first alaṃkāra,

svabhāvoḳti [2.8-13], we shall consider vārtā -- its

179

relationship to poetic language and the confusion

surrounding it -- at some length. At this point it is

important to recognize that there is no contradiction. As

Dandin comments after presenting two examples, "These are

certainly quite plausible / yet are polished through

distinguished expression (viśeṣa-ākhyāna) [ iti

saṃbhāvyamevaitadviśeṣākhyānasamskṛtam | ] [1.88ab]. It is

not the case that vārtā somehow appears as kāvya simply

because kānti may be involved. Kāvya reflects the organic

integration of a number of elements, a reality quite easy

to forget as we practice the illusionary surgery of

analysis.

The practice of the Vaidarbhas and the Gauḍīyas again

diverge, and in Dandin's verses marking this distinction

note that there is no mention of fault. If accepted by the

connoisseurs of kāvya, a distinctive and perhaps unusual

linguistic feature may serve to mark a given style:

"The learned take pleasure in meaning being excessively

superimposed apparently transcending the conventional -- no

one else" [ lokātīta ivātyarthamadhyāropyā vivakṣitaḥ |

yorthastenātivyanti vidagdhā netare janāḥ ||] [1.89].

And following two examples Daṇḍin writes: "These

reflect exaggeration (atyukti) which is favored by the

Gauḍas / But the way described previously [kāṇti as such,

1.85-87] is the essence of the other path [the Vaidarbhas]"

[idamatyuktirityuktametadgaudopalālitam | prasthānaṃ

prākpraṇītam tu sāramanyasya vartmanaḥ ||] [1.92].

(10) [1.93-100] samādhi /"transfer":

"Where a kavi -- observing conventional limits --

appropriately transfers (ādhīyate) a distinctive feature

(dharmaḥ) of one thing to another -- This is considered

samādhi" [ anyadharmastatonyatra lokasīmānurodhinā |

samyagādhīyate yatra sa samādhiḥ smr̥to yathā ||] [1.93].

In the following verses three types of "transfer" are

found: (1) transfer of an action (kriyā adhyāsa) [1.94];

(2) transfer of a word in a figurative or "secondary" sense

(gauna vṛtti) [1.95–97ab]; and (3) the simultaneous

transfer of a number of features (yugapadnaika dharmāṇām

adhyāsaḥ) [1.97cd-99]. "This is the well-known guṇa termed

samādhi -- the essence (sarvasva) of kāvya / Every group

of kavis -- without exception --/ accepts this" [ tadetat

kāvyasarvasvaṃ samādhirnāṃ yo guṇaḥ | kavisārthaḥ samagropi

tamenāmanugacchati ||] [1.100].

The kavis' ability to transfer, to shift levels of

meanings, to imagine one thing as though displaying the

actions or possessing the attributes of another thing, must

certainly be one of the most pervasive elements of creative

expression (if not, as many would posit, of language and

thought). Samādi guṇa appears closely related to the

extremely common artha alaṃkāra "rūpaka" [2.66-96], whose

distinctive feature is the "transfer of form." In view of

the above types of transfer that samādhi entails, and in

consideration of what rūpaka entails (as we shall see) we

might speculate on the grounds of their differentiation.

As S. K. De comments, "It is quite possible that from

Dandin's point of view, the difference between the samādhi

guṇa and the rūpaka alaṃkāra may consist in the fact that in

the Guṇa there is a transference only of the qualities or

actions of one thing to another, while in the Alamkāra

either one dharmin [the "possessor"] itself is substituted

for another, or the new dharma [that "possessed"] entirely

supplants the existing dharma."22

Dandin concludes this section with an explicit

recognition of the infinite possibilties stemming from the

individual, creative predilections of the poet -- yet again

a counter to those who would freeze the mārgas (and much

else) as "prescriptions":

"Thus the two Paths are distinguished through a

description of their individual natures / Yet of further

subvariations -- displayed by the individual kavis -- it's

impossible to speak" [ iti mārgadvayaṃ bhinnam tatsvarūpa-

nirūpaṇāt | tadbhedāstu na śakyante vaktum pratikavi

sthitiāḥ" ] [1.101].

"The difference in sweetness -- of sugar cane, milk,

brown sugar, and so on -- is great / Yet even by Sarasvatī

it cannot be described" [ ikṣukṣīrgudādīnāṃ mādhuryasyān-

taraṃ mahat | tathāpi na tadākhyātuṃ sarasvatyāpi śakyate

||] [1.102].

184

Notes [1.40] - [102]

1. S. K. De, "A Note on the Gaudī Rīti," in Some Problems of Sanskrit Poetics, Reprint (Calcutta: Firma KLM, 1981 (1959)), p. 62, n. 2.

2. Henry S. Heifetz, "Issues of Literary Translation from Sanskrit and Tamil," p. 202.

3. The "rīti of Vāmana is not interchangeable with Dandin's "mārga." Vāmana (KAS [1.2.6]) elevates and focuses on rīti as "the soul of kāvya"/rītirātmā kāvyasya. Dandin's usage of the term mārga is really quite loose; mārga rather providing a somewhat general context for the primary play of the various alamkāras.

4. S. K. De, History of Sanskrit Poetics, vol. 2, 2nd edition ; Reprint, 1976, p. 92.

5. S. K. De, History of Sanskrit Poetics, vol. 2, p. 92.

6. Prakash C. Lahiri, Concepts of Rīti and Guna in Sanskrit Poetics in their Historical Development (Dacca: The University of Dacca, 1937), p. 63, n. 12.

7. Prakash C. Lahiri, Concepts of Rīti and Guna in Sanskrit Poetics, p. 60.

8. Edwin Gerow, Indian Poetics, p. 23.

9. P. C. Lahiri, Concepts of Rīti and Guna, p. 63.

10. P. C. Lahiri, Concepts of Rīti and Guna, p. 64.

11. P. C. Lahiri, Concepts of Rīti and Guna, p. 66.

12. S. K. De, History, vol. 2, Reprint 1976, p. 110.

13. Cited by P. C. Lahiri, in Concepts of Rīti and Guṇa, p. 67.

14. The vocalic sthānas are: kanṭha/"throat"; tālu/"palate"; mūrdhan/"top of the palate"; danta/"teeth"; oṣṭha/"lips"; kanṭha-tālu/"throat and palate"; kanṭha-oṣṭha/"throat and lips"; danta-oṣṭha/"teeth and lips"; nāsika/"nose"; and uras/"chest."

15. P. C. Lahiri, Concepts of Rīti and Guṇa pp. 68-69.

16. P. C. Lahiri, Concepts of Rīti and Guṇa, p. 71, n. 33.

17. P. C. Lahiri, Concepts of Rīti and Guṇa, p. 72.

18. P. C. Lahiri, Concepts of Rīti and Guṇa, p. 73.

19. P. C. Lahiri, Concepts of Rīti and Guṇa p. 75, n. 38.

20. Paolo Daffina, "Review: Jan Gonda, Ancient Indian ojas, Latin augos, and the Indo-european nouns in -es -os (Utrecht: N. V. A. Ousthock's Uitgevers, 1952), in East and West, 5 (1954), pp. 142, 143.

21. P. C. Lahiri, Concepts of Rīti and Guṇa, p. 77.

22. S. K. De, History of Sanskrit Poetics, vol. 2, 2nd edition, Reprint 1970 (1960), pp. 81-82.

The Kavi and the Generation of Kāvya

"Natural creative imagination / Extensive flawless

erudition / Vigorous application -- These are the cause of

the excellence that is kāvya " [ naisargikī ca pratibhā

śrutam ca bahunirmalam | amandaścābhiyogosyāḥ kāraṇam

kāvyasampadah ||] [1.103].

"Even if there is not marvelous creative imagination

-- endowed with qualities stemming from impressions of

previous births -- Speech worshipped with erudition and

application will certainly grant a degree of favor" [ na

vidyate yadyapi pūrvavāsanāgunānubandhi pratibhānam-

adbhutam | śrutena yatnena ca vāgupāsitā dhruvam karotyeva

kamapyanugraham ||] [1.104].

"Therefore those wishing fame should continuously and

strenuously serve Sarasvatī with sloth cast aside / For

although poetic skill be slight those who make the effort

may yet sport in the gatherings of the wise" [ tadastat-

andrairaniśam sarasvatī śramādupāsyā khalu kīrtimīpsubhiḥ |

187

kr̥ṣe kavitvepi janāḥ kr̥taśramā vidagdhagoṣṭhīṣu vihartum-iśate || [1.105].

The "fullness," "wealth," "excellence" (sampada) which

the best kāvya displays stems from the conjunction and

integration of three factors: pratibhā ("creative

imagination"), śruta ("erudition"), and abhiyoga

("application").

Pratibha [ < prati (+) bha ] /"to shine upon; come

into sight, present oneself to,' but also 'to appear to the

mind, to flash upon the thought, occur to, become clear or

manifest'. . . . It usually denotes 'a sudden thought . . .,

a quick understanding or insight,' then also 'presence of

mind, wit, genius,' 'boldness, audacity,' 'fancy,

imagination'."1

Pratibhā appears in early Buddhist literature in one

sense as "eloquence," "fluency in improvisation." In the

Aṅguttara Nikāya [3.195] "The brahman Piṅgiyāni sees the Buddha approaching in all the brilliance of his superhuman

beauty and at this sight is seized with enthusiasm; he

cries out: 'O Bhagavat, I am inspired! O Sugatā I am

inspired!' 'Then may you be inspired,' responded the

master."2

In the yoga system of Patañjali "pratibhā is

synonymous with an aspect of Prajñā. It is said to be the

supreme faculty of omniscience which is evolved through a

continued practice of concentration on the self, not in its

absolute and transcendent nature, but as appearing in the

form of the phenomenal ego. . . . It is, so to speak, the

vision of the many as reflected in the mirror of the one. .

. ."3

Pratibhā retained these shades of meaning upon its

incorporation into kāvya śāstra as the primary term used to

mark the basis or source of kāvya within the kavi.4 "Si la

vicchitti fait la poésie, c'est la pratibhā qui fait le

poète. La pratibhā est un don naturel qui participe du

génie, de l'inspiration et de l'imagination."5

Its essentially ineffable nature hardly dissuaded

attempts to delineate it. "Cette inspiration, qui par sa

fantaisie même semble défier l'analyse, les critiques

indiens ont pourtant cherché à la définir."6 Or if not in

every writer's case to attempt a definition, certainly to

indicate its importance.

Dandin's position is essentially generous. This

"creative imagination" alone it would seem is not

sufficient to generate kāvya at its best. Through

extensive learning -- of kāvyas, kāvya śāstras, and all

ancillary disciplines -- the medium of its expression is

tempered and given depth; through application and practice

it is honed and perfected. And further we find that

pratibhā is innate, "natural" (naisargikī),7 and "endowed

with qualities stemming from impressions (vāsanās) of

previous births" -- it is seen as developing across time,

beyond the boundaries of any given limited lifetime.

When by the false notions of associations of body

and soul there is the feeling of a concrete

individual as "I," it is called ahamkāra. When

there is reflective thought associated with the

memory of the past and the anticipations of the

future, it is called citta. When the activity is

taken in its actual form as motion or action

towards any point, it is called karma. When,

leaving its self-contained state, it desires

anything, we have kalpana. When the citta turns

itself to anything previously seen or unseen, as

being previously experienced, we have what is

called memory (smṛti). When certain impressions

are produced in a very subtle, subdued form,

dominating all other inclinations, as if certain

attractions or repulsions to certain things were

really experienced, we have the root inclinations

(vāsanā). 8

One of the older commentators on the Kāvyādarśa

Tarunavācaspatī glosses, "'Natural inspiration' (naisargikī

pratibhā) indicates the origin of this gift: 'natural

inspiration' is an intelligence which is refined thanks to

the effect of knowledge acquired when instructing itself in

previous existences."9 And we may note Lienhard's

comments, "It is an acquired faculty gained by merit of

acts (karman) performed in previous existences which have

influenced the poet's mind in such a way that they have

left behind a residue of latent mental impressions

(saṃskāra) which has matured in his present life to genuine,

innate pratibhā."10

Yet even if pratibhā should be lacking, Daṇḍin avows

that wide learning and diligent application will "certainly

grant a degree of favor," which will allow one to

participate in and to enjoy "the gatherings of the wise,"

the practitioners and connoisseurs of kāvya. He is not

saying that "poetic talent, even when it is not innate, can

be acquired to a certain extent."11

Poetic "talent" falls within the realm of pratibhā --

one is born with it (or perhaps more correctly re-born with

it). With learning and practice one might achieve a

semblance of kāvya, and might be able to speak of it

intelligently, but this is not to be equated with the

"excellence that is kāvya" as such.

Vāmana [8th-9th centuries] stresses the importance of

pratibhā in the Kāvyālaṅkāra [1.3.16]: "Pratibhāna

[pratibhā] is the 'seed' of the kavi's creativity

[literally, "kavi-ness"]" /kavitvabĪjam pratibhānam. He

continues, "The 'seed' of the kavi's creativity is a

distinctive saṃskāra [an "impression" (of prior experience)]

derived from previous births. Without the seed, kāvya

cannot arise; if then effected, it would only be grounds

for laughter" [ kavitvasya bījam. . . . | janmāntarāgata-

saṃskāraviśeṣaḥ kaścit | yasmādvinā kāvyam na niṣpadyate

niṣpannam vā hāsyā 'yatanam syāt ||].

This centrality of pratibhā [or for some, the

synonymous "śakti"] remained throughout the tradition.

Rājaśekhara [10th century] writes with flourish in the

Kāvyamīmāṃsā [chapter 4], "Pratibhā is that which causes

the mass of words, the caravan of meanings, the weave of

alaṃkāras, the styles [mārgas] of expression and such

similar things to blaze in the spirit" (and we note his use

of mārga as with Daṇḍin, rather than rīti) [ ā śabdagrāmam-

arthasaṅkētāratantramuktimārgamanyadapi tathāvidham-

adhiḥṛdayam pratibhāsayati sā pratibhā ||].12

And Abhinavagupta in the Dhvanyālokocana, under

[1.6] of Ānandavardhana's Dhvanyāloka, writes: "Pratibhā is

a wisdom capable of creating novel things. It is

distinguished by the capacity to create all forms of rasa,

brilliance, beauty, kāvya. The sage [Bharata] designated

it as 'the interior disposition of the kavi'" [ pratibhā

apūrvavastunirmānakṣamā prajñā | tasyā viśeṣo rasāveśa-

vaiśadya saundaryaṃ kāvyanirmāṇakṣamatvaṃ | yadāha muniḥ

kaverantargataṃ bhāvam iti |.13

Bhāmaha's view is perhaps stricter, but I do not feel

in opposition to Dandin's. In Kāvyalaṅkāra [1.5] we have:

"Even a fool is capable of learning śāstra from the

teachings of master / But kāvya is born -- perhaps -- in

those who possess pratibhā" [ gurūpadeśādhyetum śāstraṃ

jaḍadhiyo 'pyalam | kāvyaṃ tu jāyate jātu kasyacitpratibhā

vataḥ ||]. Bhāmaha then would consider pratibhā a necessary

rather than a sufficient cause for the creation of kāvya.

He then expatiates on what he sees as additionally

necessary, and what I feel we may subsume within Dandin's

śruta ("erudition," "learning").

"Words, metres, meanings, kathās based on itihāsas,

worldly convention, the arts and various skills -- These

are considered the foundation of kāvya" [ śabdaścandobhi-

dhānārthā itihāsā śrayāḥ kathāḥ | loko yuktiḥ kalāśceti

mantavyā kāvyavaikharī || (KA [1.9]).

"Familiarizing oneself with words and meanings,

learning from masters versed in these, examining other

compositions -- one should then devote oneself to the

creation of kāvya" / śabdābhidheye vijñāya kṛtvā tad-

vidupāsanam | vilokyānyanibandhāmśca kāryah kāvyakriyādarah

|| (KA [1.10]).

The would-be kavi must devote him- or herself to all

of the language skills (grammar, metrics, lexicology and

etymology, phonetics, and so on); the associated śāstras

(such as kāmaśāstra ("erotics"), arthaśāstra ("statecraft,"

"economic polity"), nyāyaśāstra ("logic"), dharmaśāstra

(law, ritual, religious and social duties)); and the sixty-

four kalās or "arts and skills."

Among the kalās we find, for example: singing (gītam);

playing of instruments (vādyam); dance (nṛtyam); drawing

and painting (ālekhyam); also the making of ear-ornaments

(karṇapatrabhāgāḥ); and perfumes (gandhayuktiḥ); the

proper arrangement of jewels and adornments

(bhūṣanayojanam); the making of garlands and wreaths

(mālyagrathana-vikalpāḥ); magic and illusion (aindrajālālạḥ);

and the knowledge of omens (nimittajñānam); carpentry

(takṣaṇam); and building (vāstuvidyā); knowledge of coins

and precious stones (rūpyaratnaparikṣā); culinary skills

(vicitraśākayūṣabhakṣavikārakriyā); the preparation of

juices, liquors and spirits (pānakarasarāgāsavayojanam);

organizing cock, quail and ram fights (meṣakukkuṭalīcakayuddhavidhīḥ); dice-play (ākarṣakrīdā); word games in verse

(pratimalā); and spontaneous kāvya (mānasī kāvyakriyā);

knowledge of the various languages of foreigners

(mlecchitavikalpāḥ); and the regional dialects

(deśabhāṣāvijñānam); and (overlapping with the śāstras) such

skills as) lexicography (abhidhānakośa); prosody

(chandojñānam); and even the composition of kāvya replete

with alamkāras (kriyākalpaḥ); and so on.14

Thorough familiarity with worldly, conventional

knowledge was expected; and with the various literary

stories (kathās) based on the itihāsas; the Purāṇas; the

Mahābhārata and Rāmāyana; and of course with other kāvyas

and kāvya śāstras.

And somewhat later, Rudraṭa in the Kāvyalaṅkāra

mirrors Daṇḍin’s three fundamental factors (in [1.14]):

śakti (pratibhā), vyutpatti (śrutam), and abhyāsa

(abhiyoga). And in [1.18-19] discusses just how wide the

range of the kavi’s learning should be: "In a restricted

sense, vyutpatti refers to the discrimination of what is

appropriate or inappropriate, due to the knowledge of

prosody, grammar, the arts and skills, world affairs, words

and meanings. But in a wider sense, is there anything

other than this? In this world there is no topic or

expression that may not be an element of kāvya -- Thus this

is complete knowledge" [chandavyākaraṇakalālokasthitipada-

padārthavijñānāt | yuktāyuktaviveko vyutpattiriyam samāsena

|| vistaratastu kimanyattata iha vācyam na vācakam loke | na

bhavati yatkāvyaṅgam sarvajñatvam tato 'nyaiṣā ||.

Where Abhinavagupta, now in his commentary on the

Ghatakarpakavya (and utilizing the same terminology as

Rudrata), subsumes vyutpatti ($rutam) within the sakti

(pratibha) of the kavi: "The (imaginative) power (sakti) of

the kavi is certainly the most important thing. This

indeed is known as the erudition that transcends the world.

For there is no other erudition [of value] apart from the

(imaginative) power of the kavi" [ kavinam saktir eva

baliyasī sā eva lokottarā vyutpattirityabhidhīyate na tu

anyā kaviśakter vyutpattir nāma kācit ].15

And finally we seem to return to Dandin yet some five

centuries later with Vāgbhaṭa (I., the son of Soma) [12th

century] who writes in his Vāgbhaṭālaṅkāra [1.3]:

"Inspiration makes the poet; instruction adorns him;

practice gives him facility: it is these which mark the best

of poets" [ ratibhākarāṇaṃ tasyā vyutpattistu vibhūṣaṇam |

bhrśotpattikrd abhyāsa ityādyakavisamkathā ||].16

198

Notes [1.103] - [1.105]

1. Jan Gonda, "Pratibhā," in The Vision of the Vedic Poets (The Hague: Mouton, 1963), p. 318.

2. Hari Chand, Kālidāsa et L'Art Poétique de L'Inde (Paris: Librairie Ancienne Honore Champion, 1917), p. 65.

From the Angguttara Nikāya, Part 3: Pañcaka Nipāta and Chakka Nipāta, edited by E. Hurdy (London, 1897); Reprint (London: Pāli Text Society, 1958), p. 239: paṭibhāti maṃ bhagavā paṭibhāti maṃ sug·ta ti paṭibhatu taṃ piṅgiyāni ti bhagavā avoca.

3. Gopinath Kaviraj, "The Doctrine of Pratibhā in Indian Philosophy," Annals of the Bhandarkar Institute, 5 (1924), p. 6.

4. Note: Siegfried Lienhard's assertion (A History of Classical Poetry, (1984), p. 309, n. 1) that "This [pratibhā] is the most common term, but Daṇḍin and Vāmana use pratibhāna. . . ." is incorrect -- Daṇḍin uses [1.103] pratibhā.

5. Hari Chand, Kālidāsa, p.65.

6. Hari Chand, Kālidāsa, p. 66.

7. And again Lienhard is incorrect (A History of Classical Poetry, (1984), p. 311), in stating "Daṇḍin, Rudraṭa, Rājaśekhara and others let it clearly be understood that acquired pratibhā is definitely inferior to natural talent." There is no "acquired" pratibhā for Daṇḍin -- by its very nature it is "natural."

8. Surendranath Dasgupta, A History of Indian Philosophy, vol. 2 (Cambridge: Cambridge University Press, 1932), p. 239.

9. Cited in Hari Chand, Kālidāsa, p. 67: pūrva janmakṛtavidyāyāsotpannajñānajanitasamskāratāgatā yā buddhiḥ sā naisargikī pratibhā.

10. Siegfried Lienhard, A History of Classical Poetry, (1984), p. 311.

11. Siegfried Lienhard, A History of Classical Poetry, p. 311.

12. Rājaśekhara, Kāvyamīmāṃsā, Chapter 4, p. 11.

13. Ānandavardhana, The Dhvanyāloka of Śrī Ānandavardhana, with the Lochana and Bālapriyā commentaries by Śrī Abhinavagupta and Pandit Śrī Mahādeva Śāstri (Benares: Chowkhambā Sanskrit Series Office, 1940), pp. 92–93.

14. See: Vātsyāyana, Kāmasūtram, edited by Devduṭṭa Śāstri (Varanasi: The Chowkhamba Sanskrit Series Office, 1964), pp. 83-84. Louis Renou and Jean Filliozat, L'Inde Classique, vol. 2, (1953), Appendice 11. The Kama Sutra of Vatsyayana, translated by Sir Richard Burton and F. F. Arbuthnot, edited by W. G. Archer, Reprint (New York: Capricorn Books, 1963 (1883)), pp. 70–74.

15. The Ghaṭakarpakāvya, with the commentary by Abhinavagupta, edited by Madhusudan Kaul Shastri (Srinagar: The Mercantile Press, 1945), p. 21. Cited in J. L. Masson, "When is a Poem Artificial?" -- A Note on the Ghaṭakarpara-vivṛti," Journal of the American Oriental Society, 95 (1975), pp. 265-65.

16. As cited in Hari Chand, Kālidāsa, p. 66.

Chapter Three

Yamaka or Variations of "Phonemic Repetition"

"Yamaka is the repetition of groups of letters --

contiguous or discontiguous / Its range pertains to the

initial, medial, and final parts of pādas" [ avyapetavya-

petātmā vyāvṛttirvarṇasamhateḥ | yamakam tacca pādānāmādi-

madhyāntagocaram || ] [3.1].

"The varieties of yamakas may appear in one of the

four pādas [of the conventional padya or "stanza"], in all,

or in any combination: in the beginning; middle; end; the

middle and end; the middle and beginning; the beginning and

end; and throughout" [ ekavitracatuṣpādayamakānām vikal-

panāḥ | ādimadhyāntamadhyāntamadhyādyādyantasarvataḥ || ]

[3.2].

"These varieties -- arising from such combinations --

are both easy and difficult to compose / From among them a

few will be shown" [ atyantabahavasteṣām.bhedāḥ saṃ.bhedayo-

200

201

nayaḥ | sukarā duṣkarāścaiva darśyante tatra kecana || ]

[3.3].

Dandin develops a highly structured evolving pattern

of sound or phonemic repetition, progressing in a series of

increasingly complex steps. Essentially we have two modes

of progression. Given our initial pattern we will progress

horizontally, where each new variation operates within the

given framework, with all other elements but the varying

feature held constant. When the logical possibilities of

this mode have been demonstrated, ·e jump vertically a

short distance without breaking contact with the general

forward line of progression, with new parameters explicitly

announced. One should keep in mind that although these

patterns may be abstractly sketched, they are realized in

language -- we really have an incredibly detailed,

acrobatic demonstration of what the Sanskrit language is

capable of in this regard.

Dandin now proceeds to illustrate the various and

numerous permutations that yamaka may display. In our

first series (to [3.18]) the elemental unit is a single,

"contiguous" (vyapeta) repetition, not of individual

letters as in anuprāsa [1.51-68], but of groups of letters.

Initially we have three variable features: (1) the

repetitive block may occur in any of the four pādas ; (2)

it may occur in either the initial, medial, or final

position within that given pāda; and (3) the number of

distinct blocks may vary from one to four.

Thus in [3.4] we have a single block in initial

position in the first pāda: (1) mānena mānena sakhi ("Oh

friend! With this anger, let there not be. . . .").

Alternately, the block may be placed in (2) the initial

position in the second pāda / madano madano . . . | [3.5];

(3) in the third / caturam caturam . . . [3.6]; and (4) in

the fourth / rahitai(r)-ahitai(s) . . . || [3.7].

These variations may be abstractly sketched, where a

single capital letter stands for a specific group or block

of letters -- [ AA ___ ], for example, would represent

mānena mānena placed in initial pāda position -- with the

stanza broken into the four pādas.

[3.4] [ AA___ / ___ / ___ ; ___ ]

[3.5] [ ___ / AA___ / ___ / ___ ]

[3.6] [ ___ / ___ / AA___ / ___ ]

[3.7] [ ___ / ___ / ___ / AA___ ]

Maintaining the same initial position within the pāda,

the number of repetitive blocks may be increased. Two now

distinct blocks might appear, for example, in the first and

second pādas [3.8]; in the first and third pādas [3.9]; in

the first and fourth pādas [3.10]; in the second and third

pādas [[3.11]; in the second and fourth pādas [3.12]; in

the third and fourth pādas [3.13], and so on.

[3.8] [ AA___ / BB___ / ___ / ___ ]

[3.9] [ AA___ / ___ / BB___ / ___ ]

[3.10] [ AA___ / ___ / ___ / BB___ ]

[3.11] [ ___ / AA___ / BB___ / ___ ]

[3.12] [ ___ / AA___ / ___ / BB___ ]

[3.13] [___ / ___ / AA___ / BB___]

Extending the process a third repetitive block may

appear:

[3.14] [AA___ / BB___ / CC___ / ___]

[3.15] [AA___ / BB___ / ___ / CC___]

[3.16] [AA___ / ___ / BB___ / CC___]

[3.17] [___ / AA___ / BB___ / CC___]

And logically concluding this series, contiguous,

distinctive repetitive groups in initial position may

appear in all four pādas:

[3.18] [AA___ / BB___ / CC___ / DD___]

Yet the repetitive elements forming a contiguous pair

may be "discontiguous" (vyapeta), introducing a fourth

variable modality. Thus maintaining the initial pāda

position, we may have a given group of letters in the first

pāda repeated in the second, as in [3.20] (1) madureṇadrśam

205

māṇam (2) madhureṇa sugandhinā | ("Spring with but the

sweet and fragrant [Sahakāra shoots will turn ] the anger

of the doe-eyed ones [into fading sound]").

[3.20] [ A___ / A___ / ___ / ___ ]

And varying only which pādas the repeated elements

appear in, five additional possibilities are generated:

[3.21] [ A___ / ___ / A___ / ___ ]

[3.22] [ A___ / ___ / ___ / A___ ]

[3.23] [ ___ / A___ / A___ / ___ ]

[3.24] [ ___ / A___ / ___ / A___ ]

[3.25] [ ___ / ___ / A___ / A___ ]

Alternately, as a fifth variable feature, the number

of repetitions of the one element may be increased

(maintaining all other features constant):

[3.26] [ A___ / A___ / A___ / ___ ]

Their positions may vary:

[3.27] [ A___ / ___ / A___ / ___ ]

[3.28] [ ___ / A___ / A___ / ___ ]

And their number may extend to all four pādas:

[3.29] [ A___ / A___ / A___ / A___ ]

As in [3.8] a new and distinct repetitive block may be introduced, yet which is now, extending the current pattern, composed of two discontiguous, identical elements:

[3.30] [ A___ / A___ / B___ / B___ ]

[3.31] [ A___ / B___ / A___ / B___ ]

[3.32] [ A___ / B___ / B___ / A___ ]

Dandin initiates yet a third series, "There is also a variety [of yamaka] whose form is both contiguous and discontiguous" / avyapetāvyapetātmā vikalpopyasti [3.33cd].

That is, repetition is considered from two perspectives simultaneously. Again, as in the first series, we have a block consisting of two adjacent elements [ AA ], which are

thus "contiguous." Yet our perspective also shifts to this

block as a whole, which itself is repeated -- the two

blocks themselves are thus considered "discontiguous." We

have moved vertically, but Dandin now immediately adds an

additional feature to this paradigm -- two pairs of

repeating blocks:

[3.34] [AA___ / AA___ / BB___ / BB___ ]

Dandin's Sanskrit example thus reads:

(1) sālam sā lambakalikā

(2) sālam sālam na vīkṣitum

(3) nālinā līnābakulā

(4) nā lī nā lī kinīrapi

Varying this sequence we have (note that we are still

in pāda initial position):

[3.35] [ AA___ / BB___ / BB___ / AA___ ]

208

And filling out the template with but one repeated

block:

[3.36] [ AA___ / AA___ / AA___ / AA___ ]

Dandin now shifts and varies a feature that has thus

far remained constant -- position within the pāda.

"Such is the way of yamakā's varieties in pāda initial

position / In this same way other yamakas may be formulated"

[ iti pādādiyamakavikalpasyedṛśi gatir | evameva vikalpyānī

yamakānītarānyapi ||] [3.37].

"For fear of over elaboration there is no intention to

exhaustively describe these varieties / Rather some of

those considered difficult to compose will now be

described" [ na prapañcabhayādbhedān kārtsyenākhyātum-

īhitāḥ | duṣkarābhimatā ye tu varṇyante tetra kecana || ]

[3.38].

Maintaining the same previous paradigm, although

reducing it to one block, Dandin now shifts to pāda medial

position:

209

[3.39] [ AA , AA / AA / AA ]

These blocks may be broken up with the repetitive block

itself thus discontiguous:

[3.40] [ _A_A / A_A / A_A / A_A ]

One element in each separated pair may be dropped and

again these may shift, now to final position (mirroring

[3.29]):

[3.41] [ __A / __A / __A / __A ]

Maintaining the same paradigm, each element may be

doubled (mirroring [3.39] with a contiguous/discontiguous

pattern), but with the identical blocks now in final

position:

[3.42] [ __AA / __AA / __AA / __AA ]

And mirroring the pattern developed from [3.39], the

elements of each pair may be separated, one remaining in

pāda final position, one moving to the medial position:

210

[3.43] [ _A_A / _A_A / _A_A / _A_A ]

These elements, maintaining their positions, may be doubled yet again:

[3.44] [ _AA_AA / _AA_AA / _AA_AA / _AA_AA ]

Or mirroring [3.43], the elements of a single pair may be separated, now in initial and medial positions:

[3.45] [ A_A / A_A / A_A / A_A ]

Again these may be doubled, as in [3.44], but now only half of the immediately preceding positions are held constant. We thus have separated identical blocks in initial and final pāda position in pādas one and three, and in initial and medial position in pādas two and four:

[3.46] [ AA_AA / AA_AA / AA_AA / AA_AA_ ]

Dandin again alternates patterns, returning to single separated repetitive elements (mirroring [3.43]), although

211

now in pāda initial and final position:

[3.47] [ A__A / A__A / A__A / A__A ]

And again these may be doubled, holding pāda position

constant:

[3.48] [ AA_AA / AA_AA / AA_AA / AA_AA ]

We may return to single elements, yet adding a third

in each pāda, thus logically completing the patterns of

[3.45] and [3.47]:

[3.49] [ A_A_A / A_A_A / A_A_A / A_A_A ]

And once again each of these may be doubled in place,

completing and concluding this series:

[3.50] [ AA_AA_AA / AA_AA_AA / AA_AA_AA ]

Dandin's example (and it should go without saying that

although sound is repeated the meaning is each case is not)

of this more difficult pattern is:

(1) kālakālagalakālakālamukhakālakāla-

(2) kālakālapanakālakālaghanakālakāla-

(3) kālakālasitakālakā lalanikālakāla-

(4) kālakālagatu kālakāla kalikālakāla

("Oh You who fulfill like the Embellisher of Alakā

[Kubera], You seasons of seasons, You capable of

embellishing all buds / Let those beautiful women -- with

heads embellished with black curls, dark as the neck of the

Destroyer of Time [Śiva], as a swarm of bees, as the black-

faced monkies, as Kālá [God of Death], as Time, as the

black-clouded season which causes the black-headed ones

[Peacocks] to cry out -- embrace me!")

Samdasta Yamaka

Dandin now introduces and provides but a single

example for samdasta yamaká, that is, where repetitive

sound elements are (literally) "bitten or held between the

teeth."

213

"The position of samdasta yamaka is the end and the

beginning of two pādas / Although this is included in the

preceding it is mentioned here independently" [ samdastaya-

makasthānamantādī pādayordvayoh | uktāntaragatamapyetat

svātantryenātra kīrtyate ||] [3.51]. Elements thus meet

only at the boundaries of two pādas.

Dandin's example displays three distinct pairs of

single elements:

[3.52] [A / A___B / B___C / C]

Samudga Yamaka

In Dandin's third mode of yamaka, samudga, the

repeated element is extended to the pāda as a whole.

"Samudga is repetition involving one-half [of a

stanza, that is two pādas] / Its varieties are three"

[ardhābhyāsah samudgah syādāsya bhedāstra yo matāḥ |]

[3.53ab].

We may have two pairs of matching pādas:

[3.54] [ __ / ..... / __ / ... ]

[3.55] [ __ / __ / ..... / ... ]

[3.56] [ __ / ..... / ..... / __ ]

Pāda Abhyāsa or the "Repetition of Pādas"

Yet the repetition of pādas need not be restricted to

two distinct pairs. "And further, the repetition of pādas

(pādābyāsa) of numerous variations will be illuminated with

examples" [ pādabhyāsopyanekātmā vyajyate sa nidarśanaiḥ

||] [3.53cd]. Thus the number of identical pādas may

extend to three, with varying placement within the stanza:

[3.57] [ __ / __ / -- - / .... ]

[3.58] [ __ / -- - / __ / .... ]

[3.59] [ __ / -- - / ..... / __ ]

[3.60] [ -- - / __ / __ / .... ]

[3.61] [ -- - / __ / ..... / __ ]

[3.62] [ -- - / ..... / __ / __ ]

[3.63] [ __ / __ / __ / -- - ]

[3.64] [ __ / __ / -- - / __ ]

[3.65] [///]

Or to the extreme of four:

[3.66] [///]

"The repetition of a pāda once, twice, thrice is thus illustrated" [sakṛdadvistriśca yobhyāsah pādasyaivam pradarśitaḥ |] [3.67ab].

Śloka Abhyāsa or "Stanzaic Repetition"

At the level of the entire pāda, repetition may extend to the entire śloka (padya) or stanza. That is, two contiguous ślokas may be phonemically identical -- and with the further stipulation that their meanings must be related.

"The repetition of ślokas whose meanings are related is considered śloka abhyāsa" [ślokadvayam tu yuktārtham ślokābhyāsaḥ smṛto yathā ||] [3.67cd]. Thus two contiguous

stanzas appear identical, whose meanings although related

yet vary:

[3.68] [ ___ / ___ / ..... / ..... ]

[3.69] [ ___ / ___ / ..... / ..... ]

Mahā Yamaka

"Four identical pādas within which repetition is seen

is termed mahā yamaka -- This is the highest formulation of

yamaka" [ ekākaracatuṣpādam tanmahāyamakāhavyam | tatrāpi

drśyatebhyāsaḥ sa parā yamakakriyā ||] [3.70]. Each pāda

as a whole is identical, yet now each may be broken evenly

into two matching halves -- we have a stanza composed of

eight identical groups of letters:

[3.71] [ AA / AA / AA / AA ]

Thus in Daṇḍin's example the pāda "samānayāsamānayā" is

repeated four times ("Unequaled one! Unite me with this

lady -- angry, without compare -- whose measured distress

is equal [to my own] yet who is not without splendor and

style).

Samsrṣṭi or the "Combination of Yamakas"

Dandin gives a single example in [3.72] of what may be

considered samsrṣṭi yamaka, that is, the display of a

number of specific types of yamaka. This stanza thus

presents an avyapeta/vyapeta ("contiguous"/

"discontiguous") yamaka in the first pāda; three distinct

vyapeta yamakas in each of the three following pādas; and

three distinct samdaṣṭa ("bitten") yamakas at the three

pāda boundaries.

This pattern is realized in the Sanskrit verse as:

(1) dharādharākārādharā dharābhujām

(2) bhujā mahīm pātumahīnavikramām

(3) kramāt sahante sahasā hatārayo

(4) rayoddhurā mānadhurāvalambinah

Which we may abstract and picture as:

(1) [ AA ___ AAB

(2) BC ___ CD

(3) DE ___ DF

(4) F_G_G_ ]

Pratiloma Yamaka or "Repetition in Reverse"

"The yamaka displaying repetition in reverse /

involving either a pāda, one-half of or an entire śloka /

is considered -- due to the reversal -- pratiloma" [ āṛttiḥ

prātilomyena pādārdhaślokagocarā | yamakaṃ pratilomatvāt

pratilomamiti smṛtam ||] [3.73].

We now move yet another step up in complexity. In

pratiloma (literally, "against the hair"; "against the

grain") yamaka the first pāda read backwards -- right-to-

left, syllable by syllable, with the syllables themselves

unchanged -- will generate the second pāda (as read left-

to-right); and the second pāda read backwards will generate

the first. The same relationship holds for the third and

fourth pādas. The pattern displayed in Daṇḍin's Sanskrit

example in [3.74] is:

(1) yāmatāśca kṛtāyāsā

(2) sā yātā kṛśatā mayā

(3) ramaṇārakatā testu

(4) stutetākaraṇāmara

Thus in reading, for example, the first pāda

backwards, we would not read it letter by letter

(āsāyātiṛkaśātamāy ], but rather syllable by syllable, which

would generate the second pāda (sā yā kṛi śā tā ma yā).

With the individual numbers marking a complete pāda,

and with the pointers indicating which direction one should

read to equalize the two pādas, we might graphically

represent this pattern as follows:

[3.74] [ < (1) = (2) >

(1) > = < (2)

< (3) = (4) >

(3) > = < (4) ]

And the focus might shift to the level of the half-

stanza, where a reading of the first-half (first two pādas)

-- right-to-left, syllable by syllable -- will generate the

second-half of the stanza (the last two pādas); and

similarly for the reverse:

[3.75] < (1) (2) = (3) (4) >

(1) (2) > = < (3) (4) ]

And carrying the process to its logical extreme, our

focus may extend to the entire stanza. Now with two

adjacent stanzas, reading the first right-to-left, syllable

by syllable, will generate the complete following stanza, as

read left-to-right, syllable by syllable -- and again the

reverse holds:

[3.76] and [3.77]

[ < (1234) = (5678) >

(1234) > = < (5678) ]

221

Duṣkara Śabda Alamkāras -- Those "Difficult to Construe"

(1) Gomūtrikā or "Cow Piss"

"Having syllables of the same form in both half-

stanzas, but with a gap [of differing syllables] in between

-- This -- difficult to compose -- the wise call "gomūtrikā"

[ varṇānāmekartūpatvaṃ yattvekāntaramardhayoḥ | gomūtriketi

tat prāhurduṣkaram tadvido yathā ||] [3.78].

With the picturesque gomūtrikā or "cow piss," we again

have repetition, the medium -- the range of syllables that

may be employed -- is open, and once again we have

attempted conformity to a predetermined syllabic

"potentiality," but now we move another step further. The

syllabic arrangement itself is now but a means to a

preconceived "pictoral" template. As a cow urinating when

walking will create a zig-zag pattern in the dust, so in

gomūtrikā with the two half-stanzas aligned vertically,

beginning with either initial letter, one must be able to

generate alternately each of the half-stanzas when

proceeding in a "zig-zag" fashion. And this given that

every other pair of vertically aligned syllables (beginning

with the first) will be the same.

Graphically displaying Dandin's example in [3.79]

might make this somewhat clearer:

Pādas (1) and (2)

ma da no ma di rā kṣi nā ma pā ṅgā

2 1 2 1 2 1 2 1 2 1 2

Pādas (3) and (4)

ma de no ya di tat kṣi na ma na ṅgā

1 2 1 2 1 2 1 2 1 2 1

Each of the two half-stanzas are aligned vertically,

syllable by syllable. We note that beginning with the

first column, in every other column forward, both syllables

are identical. Between them we have a "gap" of differing

syllables. Thus reading from the lower initial syllable,

if we proceed in a zig-zag fashion following the "1's" we

will generate the first half-stanza; similarly, following

the "2's" we have the second half-stanza.

(2) Ardhabrahma or the "Half-Rotation" /

Sarvatobhadra or the "Full-Rotation"

"They call it Ardhabrahma if there is one-half (ardha)

rotation (bhramana) of the stanza / If there is complete

(sarvatas) rotation it is considered Sarvatobhadra" [ prā-

hurardhabhramaṃ nāma ślokārdhabhramaṇam yadi | tadiṣṭaṃ

sarvatobhadraṃ bhramaṇam yadi sarvataḥ || [3.80].

(3) Ardhabrahma

There would appear to be three possible interpretations

of ardhabrahma. In each case the first step is to align the

four pādas of the given stanza vertically. One view (as

that of the commentator Premachandra Tarkabāgīśa1) would be

to then create a matching "block" and, while leaving the

first in place, turn this duplicate "one-half" (that is,

180 degrees along the horizontal axis, and 180 degrees

along the vertical axis) and then place it underneath (we

might say that the original stanza is turned upside-down

and rolled over left-to-right).

Dandin's example in [3.81] laid out in this manner,

would be thus represented:

(1) ma no bha va ta va nī kam

(2) no da yā ya na mā ni nī

(3) bha yā da me yā mā mā vā

(4) va ya me no ma yā na ta

(4) ta na ya ma no me ya va

(3) vā mā mā yā me da yā bha

(2) nī ni mā na ya yā da no

(1) kam nī va ta va bha no ma

We may note then, that given this pattern we have four

ways of generating the original stanza: (1) from the top

left, forward left-to-right down the four rows; (2) from

the bottom right, right-to-left up the four rows; (3) from

the top left, top-to-bottom across to the right four

columns; and (4) from the bottom right, bottom-to-top and

across to the left four columns.

The second method (as presented by Gerow, Glossary/179)

is easier to conceptualize. Now we have only have the

original stanza arranged in four horizontal rows:

(1) ma no bha va ta va nī kam

(2) no da yā ya na mā ni nī

(3) bha yā da me yā mā mā vā

(4) va ya me no ma yā na ta

From the top left, we read in essentially a large

counter-clockwise spiral, down the first column on the

left, up the last column on the right; down the second

column from the left, up the second column from the right,

and so on. Following this movement we generate one reading

of the original stanza.

Yet we may also consider that ardhabraḥma involves the

following. The four pādas of the original stanza are again

stacked vertically. When this is done we find that if we

begin at the top left and move down we generate only one-

half of the first pāda; and further we cannot do a complete

reversal -- we must skip to the top of the second column

and move down again to generate the first-half of the

second pāda and so on to the right until but the first-half

of each pāda appears. At this point we repeat the process

but from the bottom right, moving up each column proceeding

to the left, generating the second-half of each of the

original four pādas in order as we go. We are limited to

but a "half-reversal" in our movements, and can but

generate one-half of a pāda as we proceed.

(3) Sarvatobhadra

With sarvatobhadra Premachandra again generates a

second block of four pādas, but there is really no sense of

logical extension from the preceding. In Gerow's case

there is no mention of a "complete" helical movement,

rather of "a verse, having the same number of lines as

syllables, which can be read backwards and forwards both

vertically and horizontally" (Glossary/189).

Let us lay out in four rows stacked vertically the pā

das of Dandin's example of sarvatobhadra presented in

[3.82]:

(1) sā mā yā mā mā yā mā sā

(2) mā rā nā yā yā nā rā mā

(3) yā nā vā rā rā vā nā yā

(4) mā yā rā mā mā rā yā mā

With sarvatobhadra we have "complete" movement, that

is, we have the same cyclical movement as in ardhabrahmā --

down the first column, up the last, and so on -- but now we

may also generate the original four pādas through a

corresponding reverse helical movement. Thus, beginning at

the top right syllable we proceed down that column, then up

the first column on the left, down the second column in

from the right, and so on.

And from another perspective, where in ardhabrahmā we

were capable of only generating one-half of a pāda by

moving down a given column, now we may do a complete

reversal -- moving down and then immediately back up in

each of the first four columns from the left will yield the

four original pādas; where moving down and then immediately

back up in the first four columns from the right will yield

the original four pādas yet with each in reverse syllabic

order.

229

Niyama or "Phonemic Restriction"

We have seen in the preceding the extreme variations

in the four pāda stanza that may be achieved in Sanskrit

given any number of "formatting" constraints. Hardly

content to rest here, kavis sought the challenge of

composing when the medium itself was constrained.

"The restriction of vowels, sthānas, or consonants to

four or less is considered difficult to achieve -- These

will now be shown / Otherwise a stanza is easy to compose"

[ yaḥ svarasthānavarnānāṁ niyamo duṣkarosvasau |

iṣṭaścatuḥprabhṛtyeṣa darśyate sukaraḥ paraḥ ||] [3.83].

That is, one must compose a stanza given a specific

restriction (niyama) on either the number of svaras

("vowels"), sthānas (physical points of verbal

articulation), or varṇas (literally, "letters";

"consonants").2 If the number allowed for the given

category is five or more, the task is considered "easy"

(sukara) -- four or less are considered another matter

(duṣkara).

Dandin presents a series of twelve examples, four for

each category of niyama, as follows:

(1) Svāra or "Vowel Restriction"

[3.84] Restriction to four vowels: [ ā / ī / o / e ].

In this case, each of the four pādas displays only one

of these four vowels.

[3.85] Restriction to three vowels: [ a / i / u ].

[3.86] Restriction to two vowels: [ ī / e ]

(With [ ī ] appearing alone in the first two pādas,

[ e ] appearing alone in the last two pādas).

[3.87] Restriction to one vowel: [ ā ].

(2) Sthāna or "Articulatory Restriction"

[3.88] Restriction to four sthānas: [ danta ("teeth" /

tālu ("palate") / mūrdhan ("top of the palate") / kanṭha

("throat") ].

[3.89] Restriction to three sthānas: [ danta / tālu /

kanṭha ].

231

[3.90] Restriction to two sthānas: [ danta / kanṭha ].

[3.91] Restriction to one sthāna: [ kanṭha ].

(3) Varna or "Consonant Restriction"

(Where the vowel added to realize the consonants does

not itself appear to be restricted.)

[3.92] Restriction to four consonants: [r / g / k / m].

[3.93] Restriction to three consonants: [ d / n / v ].

[3.94] Restriction to two consonants: [ r / s ].

[3.95] Restriction to one consonant: [ r ].

We might add that Danḍin provides an excellent example

of sthāna niyama in Chapter One of his Daśakumāracarita.

The entire chapter is written without employing any labial

(oṣṭhya) letters -- reflecting the state of the protagonist

Mantragupta's lips, sore from excessive love play.

232

Prahelikā or the "Riddle"

Daṇḍin follows his various types of duṣkara śabda

alaṃkāras, which focus essentially on the syllable --

whether in restriction (niyama), or in selective

arrangement in the service of a preconceived pattern of

reading movement (gomūtrika) and/or repetition (ardha-bhrama

and sarvatobhadra) -- with an entirely new category.

"Thus in the path of those duṣkara varieties a method

is demonstrated / The method of the varieties of Prahelikā

will now be explained" [ iti duṣkaramārgopi kaśvidādarśitaḥ

kramaḥ | prahelikāprakārāṇām punaruddiśyate gatiḥ̣ || ]

[3.96]. The prahelikā is a "riddle" or "literary puzzle,"

a question and answer happily couched in resolvable

ambiguity.

Evident throughout Indian literature, riddles appear

in the Vedas as brahmodya or brahmavadya "désigne dans le

rituel védique un échange de questions et de réponses entre

les participants du culte, échange qui se situe à certaines

moments essentiels de la liturgie."3 And even here their

entertainment value is evident, "[The Brahmans] employ a

very interesting form of poetic riddle or charade to

enliven the mechanical and technical progress of the

sacrifice by impressive intellectual pyrotechnics."4

As Dandin continues, "Prahelikās are useful in the

entertainments of playful gatherings [of literary

connoisseurs] (goṣṭhīs) / for private conversation between

those familiar with these when in public / and for the

confusion of others" [ krīḍāgoṣṭhīvinodeṣu tajjñair-

ākīrṇamantrāṇe | paravyāmohane cāpi sopayogāḥ prahelikāḥ ||

[3.97].

Dandin itemizes sixteen varieties of prahelikā, yet it

is extremely important to note the verses immediately

preceding his various examples. [3.106] "These are the

sixteen prahelikās indicted by previous teachers / yet

fourteen other defective (duṣṭa) prahelikās were also

taught by them" [ etāḥ ṣoḍaśa nirdiṣṭāḥ pūrvācāryaiḥ

prahelikāḥ | duṣṭaprahelikāścānyāstairadhītāścatūrdaśa ||]

234

[3.106]. "We however -- assuming the defects to be

innumerable -- shall speak only of the good ones / The

defective ones will be left without characterization "

[ doṣānaparisamkhyeyān manyamānā vayam punah | sādhvīr-

evābhidhāsyāmas tā duṣṭā yāstvalakṣaṇāḥ ||] [3.107]. Once

again, we have evidence of an active earlier formal

tradition from which Dandin drew.

And we should be aware that the various varieties of

prahelikā involve an array of patterned technique and

"riddle" to be. That "although riddle poetry belongs to

the short form of kāvya and, frequently being

composed on the spur of the moment, has often not

been preserved, there can be no doubt that it was

one of the most popular forms of Indian lyrical

poetry. In many respects it conformed to the

requirements of kāvya: it was written in many

different metres, some of them difficult, it made

use of an unusual vocabulary comprehensible only

to the connoisseur and, like so many other poems,

it was two-dimensional in that behind the meaning

first perceived, in this case the question, there

lay a second, hidden meaning in the poem; the

answer, which the reader or listtener had to

decipher for himself.5

235

Dandin's sixteen varieties of prahelikā are the

following:

(1) samāgatā / "Where meaning is hidden through the

coalescence of words" [ āhuḥ samāgatām nāma gūḍhārthā

padasaṃdhinā 1] [3.98ab], with an example in [3.108].

(2) vañcitā / "Where there is deception through a

word whose usual denotation is other [than that intended]"

[ vañcitānyatra rūḍhena yatra śabdena vañcanā 1] [3.98cd],

with an example in [3.109].

Dandin's first two varieties of prahelikā reflect his

two primary categories of śleṣa alaṃkāra [2.310-22]. In

samāgatā, bhinna śleṣa is involved. The ease of word

"coalescence" in Sanskrit may easily be employed to create

an intentional ambiguity. Here a unitary string of

syllables may be variously broken up, yielding respectively

varying meanings.

With vañcitā prahelikā, abhinna śleṣa is displayed.

Now the words as such are clearly integral, but a given

word may express more than one meaning. In each case,

"hesitation between two meanings, both of them possible, at

first tends to throw the hearer off. But secondary factors

allow him to choose between the two and, in case of

multiple meanings, to determine a hierarchy."6

(3) vyutkrāntā /"creates confusion through the

employment of [related words] excessively separated"

[ vyutkrāntātiyavahitaprayogān mohakāriṇī || ] [3.99ab], with

an example in [3.110]. Ludwik Sternbach comments, "Today,

this would not be considered as a riddle sensu stricto but

as . . . not well construed and because of that difficult

to understand. The difficulty in understanding . . .

depends on using wit and intelligence and therefore it was

considered in ancient India as a riddle."7

In vyutkrāntā prahelikā there is really only one

correct interpretation. Through separating words which

would otherwise be ordered quite closely (to aid the under-

standing), confusion arises.

(4) pramuṣitā / "Where the succession of words

contains a meaning difficult to understand" [ sā syāt

pramuṣitā yasyāṃ durbodhārthā pāḍāvalī || ] [3.99cd], with

an example in [3.111]. As in the preceding vyutkrāntā,

confusion arises. Yet now it is due to the employment of

rare and obscure words, to an unusual semantic presentation

rather than an unusual syntactic arrangement.

(5) samānarūpā / "The one strewn with words employed

with indirect (gauna) meanings" [ samānarūpā gaunārthā-

ropitairgrathitā padaiḥ | ] [3.100ab], with an example in

[3.112]. Going beyond the literal meanings of a number of

the words actually presented, the solution of samānarūpā is

to be found in the realization of various indirect or

figurative meanings. Samānarūpā is an extension of the

previous [3.98cd] vañcita prahelikā, where put a single word

is to be taken in a secondary or figurative sense. Clearly

the focus is upon artha ("meaning") rather than upon śabda

("sound" or the phonemic entity) in such varieties as

vañcitā, pramuşitā and samānarūpā. Indeed Marie-Claude

Porcher would see samānarūpā reflecting atiśayokti alamkāra

[2.214-20]:

The classification of the metaphorical process

within the prahelikā should not overshadow the

fact that this same process gives rise to the

figure of speech atiśayokti. . . . Thus [this]

prahelikā does not differ -- linguistic-

ally from the figure of speech atiśayokti, which

belongs to the arthālan̐kāra. The disappearance of

one term of the comparison [a beautiful woman is

compared to a creeper or vine in the example of

[3.112] casts a doubt in the mind of the reader

and results in an enigma: thus the process itself

has been classified as a part of the prahelikā.8

(6) paruşā / "With a word etymologically derived

merely due to the existence of grammatical rules (lakşana)

[ paruşā lakşan̐āstitvamātravyutpāditaśrutih ||] [3.100cd],

with an example in [3.113]. Now the solution of the

prahelikā lies in the correct application of derivational

grammatical rules to a given word -- whose usual meaning is

evident -- generating a second meaning which the speaker or

writer wishes to convey. In the example [3.113], "surāh"

239

clearly means "gods," yet through the application of

Pāṇinian rules [3.1.21] and [3.1.134] the additional and

desired meaning of "drinkers," "drunkards" is revealed.9

(7) saṃkhyātā / "where enumeration (saṃ-khyāna) is

the cause of perplexity" [ saṃkhyātā nāma saṃkhyānaṃ yatra

vyāmohakāraṇam |] [3.101ab], with an example in [3.114].

Although clues are given, the solution now depends upon the

correct application of enumerated attributes. The example

for this prahelikā is held to be evidence for Daṇḍin's

habitation in the South: "There is a city with a nasal

(nāsikya) in the middle / embellished on the sides with [a

total of ] four letters / wherein there are kings whose

names have eight letters" [ nāsikyamadhyā paritaścatur-

varṇavibhūṣitā | asti kācitpurī yasyāmaṣṭāvarṇāhvayānṛpāḥ

|] [3.114]. The city is thus Kāñcī and with Pallavah as

the name of its kings.10

(8) prakalpitā / "Where the meaning of a sentence

(vākya) appears other [than what one intends]" [ anyathā

240

bhāsate yatra vākyārthah sā prakalpitā || [3.101cd], with

an example in [3.115]. The focus is now on the meaning

conveyed at the sentence level. A word appears to be

ambiguous, but the context of the whole indicates the

correct solution. In the example [3.115] the word "vrddhe"

(as the vocative of vrddhā) would initially lead one to

assume that the verse is addressed to an "old woman" -- in

conflict with the male speaker's evident physical

agitation, one "waiting with stumbling words, bowed head,

pathetic glance, and trembling" [ girā skhalantyā

namreṇaśirasā dīnayā drśā | tiṣṭhantamapi sotkampaṃ vṛddhe

māṃ nānukampase ||]. Yet vṛddhā may also refer to the

goddess of wealth Lakṣmī, and given the context this sense

should be selected.11

(9) nāmāntaritā / "Where in regard to a name there is

the postulation of various meanings" [ sā nāmāntaritā

yasyāṃ nāmni nānārthakalpanā | ] [3.102ab], with an example

in [3.116]. Again we have multiple possible meanings, yet

241

now resolution of ambiguity rests on the correct

identification of a specific nominal or "name." [3.116] "Oh

You of unsteady eyes! Something well-known on the earth

(pārthivah) is at first called a king (rājā) and eternal /

But this one is neither a king nor eternal" [ ādau

rājetyadhīrākṣi pārthivaḥ kopi gīyate | sanātanāśca naivāsau

rājā nāpi sanātanah || ]. From this one should infer that

the alternate reading of pārthivaḥ, "one born from the

earth" or "tree" is correct. And combining disparate

elements explicit albeit "hidden" in the verse -- [ rājā

(+) (sana-) tanas ] -- the name of a particular tree, the

Rājatana, is found.

(10) nibhṛtā / "The one that has another meaning

concealed (nibhrta) in words that touch on common

attributes (dharmas)." [ nibhṛtā nibhṛtārthā tulya-

dharmasprśā girā ||] [3.102cd], with an example in [3.117].

The denial of what one would initially assume to be the

subject of a series of attributes leads to the inference of

the correct subject -- and the reinterpretation of the

attributes to respectively correspond.

As Gerow notes (Glossary/212), nibhṛtā may be compared

to samānarūpa [3.100ab, 112]. Here we have a series of

"hidden" attributes, "adjectival comparability"; in the

latter a "similarity of form," or "nominal comparability."

(11) samānaśabda / "The one realized through synonyms

(paryāya) of the words actually expressed" [ samāna-

śabdopanyastasābdaparyāyasādhītā ] [3.103ab], with an

example in [3.118]. One must now derive appropriate

synonyms for certain explicit words, and apply them to the

verse as a whole.

We find in the example [3.118], "Sweet speaker! That

of yours whose name is 'non-earth' (a-bhūmi) [ bhūmi = dhara

adhara = "lower lip" ] / which has conquered that whose

name is 'extensive hair' (pra-kṛṣṭa-keśa; [ keśa = vāla >

pravāla = "tender bud" ]. . . ." That is, "That lower lip

of yours, which has conquered the tender bud [the lower lip

is similar in shape to the tender bud yet conquers it

surpassing beauty] now generates a great desire in me"

[ jitaprakṛṣṭakeśākhyo yastavābhūmisāhvayah | sa māmadya

prabhūtotkaṃ karoti kalabhāṣiṇī ||].

As nāmāntaritā prahelikā [3.102ab, 116] focused on the

correct resolution of homonyms, now success depends upon

the correct identification of synonyms.

(12) sammūḍhā / "Where perplexity appears despite a

meaning directly stated" [ sammūḍhā nāma yā sākṣānir-

diṣṭārthāpi mūḍhaye ||] [3.103cd], with an example in

[3.119]. In the previous vyutkrāntā prahelikā [3.99ab,

110] confusion arises from syntactical displacement. In

sammūḍhā the evident meaning makes little sense.

In the first half-stanza of Daṇḍin's example [3.119],

two lovers on a bed turn away from each other out of anger.

Yet the second-half reads, "Lying down in a similar way out

of passion they freely kissed [each other's] mouths"

[ śayanīye parāvṛttya śayitau kāminau krudhā | tathaiva

śayitau rāgāt svairam mukhamacumbatām ||.

Following the

action of the first-half, "Lying down in a similar way"

would initially seem to indicate that they again "turned

away." Yet given the contradiction, "in a similar way" may

simply mean "again they turned over."

(13) parihārikā / "Whose form is a succession [of

words] from the combination of which another word is

derived" [ yogamālatmikā nāma yā syāt sā parihārikā | ]

[3.104ab], with an example in [3.120]. In pra muṣitā

prahelikā [3.99cd, 111] the meaning of obscure words must

be found. Parihārikā is similar -- now a continuous string

of words, conjoined through compounding, is in fact a

series of epithets which (1) must be correctly broken up,

and (2) the meaning must be correctly grasped.

(14) ekacchannā / "Where the dependent (āśrita) is

evident and the basis of dependence (āśraya) is hidden"

[ ekacchannāśritam vyaktam yasyā māśrayagopan am || ]

[3.104cd], with an example in [3.121]. Here but part of a

subsuming whole is directly provided as a clue. Yet

further, the "whole" is present as well -- but one step

removed. This is clarified by the example in [3.121].

"This hand (hasta) of something which is not human

(āmanuşya) never touches a weapon nor a women's breasts /

Yet surely it is not without fruit" [ na sprśatyāyudhaṃ jātu

na strīṇāṃ stanaṃandalam | amanuṣyasya kasyāpi hastoyaṃ na

kilāphalạḥ ||]. "Fruit" is provided as the dependent part

(āśrita). Given this and the further clues of the verse,

one might recognize that a synonym of āmanuşya ("not human")

is "gandharva" (a celestial being). And this combined with

hasta ("hand") would indeed generate "gandharvahasta," a

specific species of tree, the Eranda -- thus the "basis of

dependence" (āśraya).

Ekacchannā prahelikā thus not only incorporates a

specific type of relationship (part/whole) in varying

degrees of exposure, but it also utilizes a technique that

we have seen in samānaśabda [3.103ab] -- pointing to a

"hidden" element through explicitly including a relevant

synonym (paryāya).

246

(15) ubhayacchannā / "Where there is concealment of

both [the dependent (āśrita) and the basis of dependence

(āśraya) ]" [ sa bhavedubhayacchannā yasyāmubhayagopanam | ]

[3.105ab], with an example in [3.122]. Ubhayacchannā

("with both hidden") merely extends the process of the

preceding.

In the example [3.122] we find, "Who (kaḥ) joining

along with whom (kena) participate in all actions but if

seen together at the time of eating one of them is cast

away?" [ kena kaḥ saha sambhūya sarvakāryeṣu sam-

nidhim | labdhvā bhojanakāle tu yadi drṣṭo nirasyaṭe ||].

Given the clues provided, one might see the synonyms of or

double meanings reflected by the words kaḥ/kena. Kaḥ

itself may mean (among other things) "head" or "hair"; with

kena then seen as simply the instrumental singular pronoun.

Alternately, kaḥ could be taken as a synonym of "keśaḥ" or

"hair" (the one "joining along with"), with kena

specifically marking the instrumental singular as synonym

of "mastaka" or "with the head."

The Ten Doṣas or "Faults" and their Positive Transformations

The final section of Daṇḍin's Third Chapter is devoted

to an exposition of the ten doṣas or "faults" that may mar

kāvya and thus inhibit the proper generation of śobhā.

Daṇḍin at the beginning of the Kāvyādarśa [1.7] stresses the

extreme importance of avoiding defects: "Therefore a flaw

in kāvya however slight should not be neglected -- A body

however beautiful would become ugly through a single

blemish."

The importance of avoiding and eliminating doṣas in

kāvya is indeed a constant element throughout the

literature. "Whatever controversy might have existed

amongst theorists of different ages and schools regarding

the character and relative importance of . . . [the]

embellishing elements in their theory of poetry, they have

all agreed upon one fundamental point, namely, . . . they

have insisted upon the avoidance . . . of Doṣas or poetic

flaws. . . ."12

248

Although we grant the above, we should further

immediately note that there has hardly been an equality of

opinion, for the "Theorists themselves have hardly

concurred with regard to the nature and scope of the

individual Doṣas -- their classification, number and

nomenclature, and their relationship with other poetic

factors."13 A doṣa for one writer might be a guṇa or

"excellence" for another. Thus perhaps the most important

point to realize is that a doṣa as such is not necessarily

absolute -- one of the most distinctive features of

Daṇḍin's presentation is his indication of how, with an

alternate situation or desired effect, a doṣa may be

transformed into a positive factor.

Daṇḍin lists the ten doṣas in [3.125-26ab], and

remarks before proceeding with his presentation, "There

are only ten doṣas -- These should be avoided by the wise"

[ iti doṣā daśaivaite varjyāḥ kāyeṣu sūribhiḥ ||]

[3.126cd]. "Whether a deficiency in pratijñā, hetu, or

drṣṭānta is a doṣa or not -- The analysis of this is

generally difficult / What's the use of merely touching upon it?" [ pratijñāhetudṛṣṭāntaḥānirdoṣo na vetyasau |

vicāraḥ karkaśaḥ prāyastenālīḍhena kiṃ phalam ||] [3.127].

Daṇḍin is referring here to those who would apparently pass somewhat lightly over a consideration of faults in

logical reasoning. Specifically, this entails deficiencies in the nyāya or -- etymologically -- the "argument that

leads one to the establishment of intended meaning."14 And

the commentator Vātsyāyana writes under Nyāya Sūtra

[1.1.1], "What then is this nyāya? Nyāya is the examination

of an object with the help of the instruments of valid

knowledge (pramāṇas)."15

"Nyāya" also refers to the structured and formal

logical argument which comprises the five following

components (according to the Nyāya Sūtra)-- three of which

Daṇḍin mentions above, and all of which then must be free of

fault:

(1) pratijñā (Nyāya Sūtra [1.1.33]) / "The statement

of what is to be proved" (sādhya nirdeśah); the thesis or

probandum.

(2) hetu [1.1.34-35] / The basis for the

establishment of the thesis "through similarity or

dissimilarity of the subject [pakṣa] with the instance,"16

that is, the example (udāharaṇa).

(3) udāharaṇa [1.1.36-37] / The exemplification or

example which is an instance (drṣṭānta -- the term Daṇḍin

employs) "similar or dissimilar to the subject either

possessing its characteristics or being opposite to it in

nature." Drṣṭānta is defined in [1.1.25] as "An object with

regard to which the layman and expert hold the same

opinion" [laukikaparīkṣakāṇāṃ yasminnarthe buddhisāmyam sa

drṣṭāntaḥ |].

(4) upanaya [1.1.38] / "Application is the

proposition which characterises the subject as 'this is

similar' (tathā) or as 'this is not similar' (na tathā)

according to the nature of the instance cited."17

(5) nigamana [1.1.39] / The conclusion, a restatement

of the thesis prefaced by the statement of the reason

(hetu).

Some would take Dandin’s verse [3.127] as a direct

reference to Bhāmaha’s discussion of logic and its possible

defects in Chapter Five of the Kāvyalaṅkāra. Given the

fifth chapter’s first verse, one’s speculations are easily

led in this direction, "Now the fault of a deficiency in the

pratijñā, hetu and so on will be described / It will be

brief and according to [the tenets of] nyāya -- The purpose

is to indicate but an idea of these" [ atha pratijñā-

hetvādihīnam duṣṭaṃ ca varnyate | samāsena yathānyāyaṃ

tanmātrārtha pratītaye ||] (KA [5.1]). As A. K. Warder

remarks, "Bhāmaha . . . proposes to compose kāvya in the

guise of logical propositions and arguments based on

experience. There must be verisimilitude, probability and

agreement with both reason and the ways of the world"

(although Bhāmaha viewing kāvya as a "guise" for logical

proposition is surely overstated).18

Dandin devotes the remainder of Chapter Three to a

presentation of the ten doṣas, their definitions and

potentialities for transformation:

(1) apārtha [3.128-30] / "The one considered to be

collectively void of meaning is apārtha / This is defective

except in the exclamations of the insane, the intoxicated,

and children" [ samudāyārthaśūnyam tadapārthamitiṣyate |

unmattamattabalānāmukteranyatra duṣyati || ] [3.128].

As in, for example, "The ocean is being drunk by devas

/ I am afflicted with old age / These clouds are thundering

/ Airāvana is dear to Indra" [ samudrah pīyate devair-

ahamasmi jarāturaḥ | amī garjanti jīmūtā harerairāvanaḥ

priyaḥ ||] [3.129]. There is no semantic coherence or

connection between these four pādas -- the verse as a whole

is "void of meaning."

(2) vyartha [3.131-34] / "Where there is

inconsistency between earlier and later sections in either a

single sentence or a text / Due to displaying contradictory

meaning. . ." [ ekavākye prabandhe vā pūrvāparāhatam |

viruddhārthatayā vyarthamiti doṣeṣu paṭhyate ||] [3.131].

253

Yet "There is a certain state displayed by the mind when

deeply preoccupied / wherein even an expression whose

meaning is contradictory would be accepted" [ asti kācid-

avasthā sā sābhiṣaṅgasya cetasaḥ | yasyāṃ bhavedabhimatā

viruddhārthāpi bhāratī ||] [3.133].

As in, for example, "How is desire for another's wife

worthy of me -- a noble man? / Oh when can I drink from her

shining lips?" [ paradārābhilāṣo me kathamāryasya yujyate |

pibāmi tarakaṃ tasyāḥ kada nu dāsanacchadam ||] [3.134].

In the initial section of the verse a man is aware of

feelings improper for one of such noble birth; "but in the

next moment his mind is peculiarly engrossed by the over

powering influence of passion which drowns the logical

sense and moral propriety."19 When such seemingly

contradictory expressions stem from such a mind as this,

from one deeply engrossed or preoccupied due to an

overpowering emotion there is not necessarily a flaw.

(3) ekārtha [3.135-38] / "If what is stated earlier

is stated again / With reference to either meaning or the

254

words without any difference. . . ." [ aviśeṣeṇa pūrvoktam

yadi bhūyopi kīrtyate | arthataḥ śabdato vāpi tadekārthaṃ

matam yathā || ] [3.135].

As in, for example, "These deep water-bearers, posses-

sors of lightning, thunderous ones -- with a color like her

curls -- cause longing in that longing girl" [ utkāmun-

manayantyete bālām tadalakatviṣaḥ | ambhodharāstaditvanto

gambhīrāḥ stanayitnavaḥ || ] [3.136].

Ekārtha then refers to excessive repetition. In our

example the senses of utkam ("one with longing") and ud-

manayanti ("these causing longing") both overlap. And

further, ambhas-dharāḥ ("water-bearers"), taditvantaḥ

("possessors of lightning"), and stanayitnavah ("thunderous

ones") are all epithets for "clouds." Yet "If one wishes

to express an excess of compassion and so on / Then even

repetition is not a fault -- Rather this is an embellish-

ment" [ anukampādyatiśayo yadi kaścidvivakṣyate | na doṣaḥ

punaruktopi pratyuteyamalankriyā || ] [3.137].

(4) sasamsaya [3.139-43] / "If words employed for

the sake of precision cause doubt / This is certainly a

fault. . . ." [ nirṇayārthaṃ prayuktāni saṃśayaṃ janayanti

cet | vacāṃsi doṣa evāsau sasamśaya iti smrtaḥ || ] [3.139].

Yet "If this is sometimes employed with the intention [of presenting] doubt itself then it would surely be an

embellishment -- There is no fault in this case. . . ."

[ īdrśaṃ saṃśayāyaiva yadi jātu prayujyate | syādalamkāra

evāsau na doṣāstatra tadyathā || ] [3.141].

As in, for example, "I see her / That faultless lady

possessed by disease born of love / not born of love /

Captured by that hard-hearted death / season -- What's the

point of us [entertaining] hope for her?" [ paśyāmyan-

aṅgajātañkalañghitāṃ tamaninditām | kālenaiva kaṭhorena

grastāṃ kiṃ nastadāśayā || ] [3.142]. Ambiguity here arises

from the usage of [ an-aṅga-ja ], which may mean either

"born of love (anāṅgaja) or "not born of love" (an-aṅgaja);

and of [ kālena ], which may mean either "by death" (that

is, "love"), or "by that ((hot) season." Yet there is no

fault in this case, for as Daṇḍin explains, "Whether his

lady is afflicted by love or scorched by heat / A female

messenger has teasingly spoken words creating uncertainly in

order to confuse the young lover" [ kāmārtā gharmataptā

vetyaniścayakarạṃ vacah | yuvānamākulīkartumiti dūtyāha

narmanā ||] [3.143].

(5) apakrama [3.144-47] / "If a later formulation

referring to items is not made in sequential conformity

with their prior formulation. . . ." [ uddeśānuguno-

rthānāmanūddeśo na cet kṛtaḥ | apakramābhidhānāṃ taṃdoṣama-

cakṣate budhāḥ ||] [3.144].

As in, for example, "Those responsible for the

maintenance, creation and destruction of the worlds . . . /

May these - Śambhu [Śiva], Nārāyaṇa [Viṣṇu] and Ambhoja

[Brahmā] -- protect you!" [ sthitinirmāṇasaṃhārahetavo

jagatāmamī | śambhunārāyaṇāmbhojayonayaḥ pālayantu vah ||]

[3.145]. In conformity with the first formulation here --

the roles of maintenance, creation, and destruction -- the

latter formulation of the gods responsible should rather

be: Viṣṇu, Brahmā, and Śiva.

257

"But if an effort -- the cause of one clearly

realizing the relationship involved -- is made [by the

kavi] / Then the wise declare that even transgressing the

order in not a fault" [ yatnaḥ sambandhavijñānahetukopi

kr̥to yadi | kramalaṅghanamapyāhuḥ sūrayo naiva dūṣaṇam || ]

[3.146].

As in, for example, "Leaving one's relatives, leaving

one's body, and leaving one's country -- In these three the

first and the last excessive distress / The middle but a

momentary fever" [ bandhutyāgastanutyāgo deśatyāga iti triṣu

| ādyantāvāyatakleśau madhyamaḥ kṣaṇikajvaraḥ || ] [3.147].

(6) śabdahīna [3.148-51] / "In the usage of words

when the path between a rule and its range of application

is not discernible / and when unacceptable to the authori-

ties (śiṣṭa) this is Śabdahīna / Yet when acceptable to the

authorities this is not defective" [ śabdahīnamānālakṣya-

lakṣyalakṣaṇapaddhatīḥ | padaprayogośiṣṭeṣṭaḥ śiṣṭeṣṭastu na

duṣyati || ] [3.148].

Clearly ungrammatical usage is a flaw. Yet what at

first may seem unacceptable may indeed be permissible for

those whose knowledge of language runs deep. As in, for

example, "The breeze coming off the southern mountain makes

the Mango trees shine with gently trembling buds and

shoots" [ dakṣiṇādṛerupasaran mārutaścūtapādapān | kurute

lalitādhūtapravālānkurāśobhinah || ] [3.150].

In this case we have what might initially appear to be

two grammatical faults: (1) In upa-saran [ vartamāne krdanta

< *ṛ > , the verbal root [*ṛ] should be replaced by the

verbal root [*dhāu] in the present participle form

(according to Pāṇini [7.3.78]). But this should only apply

when [*ṛ] means "quick moving," "running." As here "slow

movement" is meant, upasaran is acceptable.

And (2) Where a verbal root may take both ātmanePada

and parasmaipada endings, such as the root [*kṛ] here >

kurute, the ātmanePada form should be used if the one

benefiting from or acting as the recipient of that verbal

action is the agent itself; if not then the parasmaipada

form should be employed. In this case one would initially

assume that kurute -- having an ātmanePada ending -- is

incorrect since the agent of the action, the "breeze," is

affecting others. But as it turns out, if the agent is

insentient -- as here -- this rule does not apply. It is

well to heed Daṇḍin's following remark, "Cases such as

these appear as solecisms to those whose minds are too lazy

to see into the vastness of the sūtras -- And these do not

relinquish beauty" [ ityādiśāstramāhātmyadarśanālasa-

cetasām | upabhaṣaṇavadbhati na ca saubhagyamujjhati || ]

[3.151].

(7) yatibhraṣṭā [3.152-55] / "A break between words

whose position is specified is known as yati / A deviation

from this -- jarring to the ear -- is yatibhraṣṭa"

[ ślokeṣu niyatasthānam padacchedam yatim viduḥ | tadapetam yatibhraṣṭam śravanodvejanaṃ yathā || ] [3.152].

An example follows in [3.153] which is in the mandā-

krāntā metre. This displays a "samavṛtta" padya where the

number and position of syllables in each pāda are

equivalent. It has seventeen syllables to the pāda and is

represented by the following ganas: [ ma / bha / na / ta /

ta / ga / ga ], that is, [ — — — u u u u u — — u — — u —

_ ]; and with yatis or "specified word breaks" after the

fourth, tenth, and seventeenth syllable of each pāda

(further breaks may occur, but the above are mandatory).

Yet here we find that a word breaks occur in the first pāda

after the fifth and eighth syllables; in the second pāda

after the fifth and seventh syllables; in the third pāda

after the fifth and seventh syllables again; and in the

fourth pāda after the fourth, which is correct, and after

the seventh. It is not that these breaks are necessarily

incorrect, but that, with the exception of the first

position in the fourth pāda, the word breaks specified by

yati do not occur and thus we have fault.

(8) bhinnavṛtta [3.156-58] / "A deficiency or excess

of syllables / Improper placement of heavy and light

syllables -- This is bhinnavṛtta / This doṣa is truly

censured" [ varṇānāṃ nyūnatādhikye gurulaghvayathā-

sthitīḥ | tatra tadbhinnavṛttam syādeṣa doṣaḥ sunindi-

taḥ ||] [3.156]. Again, a syllable where the vowel is long

[ a / ī / ū / ṛ / e / ai / o / au ], or where the vowel is

short but followed by either an anusvāra (-ṁ), visarga (-ḥ),

or a consonant cluster is considered guru or "heavy." A

syllable where the vowel is short [ a / i / u / ṛ ], and

not otherwise qualified is laghu or "light." And further,

the last syllable of any pāda may be considered long or

short -- regardless of its natural length -- depending on

the demands of the given metre.

Dandin's examples display both aspects of bhinnavṛtta.

In [3.157], in the common anuṣṭubh metre with eight

syllables to the pāda, we find that the first two pādas are

deficient by one syllable, where the last two pādas have one

extra syllable. In [3.158] the first pāda is in the

indravajrā metre which consists of eleven syllables to the

pāda, with the following ganas: [ ta / ta / ja / ga / ga ].

A single pāda in this metre would then be [ _ _ u _ _ u u _

u _ _ ]. Yet now the second syllable is short [ ḍa ] when

it should be long, and thus we have the improper placement

of a light syllable.

(9) visandhi [3.159-61] / "'I do not intend to speak

quickly' -- The failure to combine words [properly

resulting from this intention] is visandhi / But not where

the cause [of hiatus] is due to pragrhya and so on" [ na

samhitām vivakṣāmītyasamdhānam padeṣu yat | tadvisamdhīti

nirdiṣṭam na pragrhyādinetukam ||] [3.159]. With

"pragrhya" there is an exemption to the usual sandhi rules,

and thus the absence of what would otherwise be a fault. As

Pāṇini specifies [1.1.11-19] this involves (a) [i/ī], [u/ū],

and [e] when appearing as dual endings; (b) the [Ī] of the

pronoun amī; and (c) the vowels of particles (of a single

vowel) or of interjections. And further in Aṣṭādhyāyī

[6.1.125] he notes, "Pluta and pragrha vowels are not

altered when followed by [another] vowel," where "pluta"

vowels [8.2.82-108] are "protracted vowels possessed of

three mātrās," that is, they take longer to pronounce than

the regular long (dīrgha) vowels.

263

In [3.160] we have an example with the word "calatā

[ (m.) (intr.) vartamāne krdanta ] /"moving," separated from

the immediately following aṅganā- /"beautiful women" --

these should be combined. In the following example [3.161]

the dvandva compound in the nominative dual māna-īrṣye

/"anger and jealously" is separated from the adjacent

iha/"here" -- yet "This kind of hiatus is accepted by the

wise" [ mānerṣye iha śīryete strīṇām himartau priye | āsu

rātriṣviti prajñairamṛtam vyastamīdrśam | ] [3.161].

(10) deśādivirodhī [3.162-85] / That is, "deśa,"

"kāla," "kalā," "loka," "nyāya," and "āgama" virodhī.

"Deśa ["place"] refers to mountains, forests, kingdoms, and

so on / Kāla ["time"] refers to night, day, and the seasons

/ The Kalās [the "skills"] are dance, song and so on -- the

bases of kāma and artha" [ deśodrvanarāṣṭrādiḥ kālo

rātrimdivartavah | nrtyagītaprabhrtayah kalāḥ kāmārtha-

samśrayāḥ || ] [3.162]. "The behavior of mobile and immobile

beings is termed Loka / Nyāya refers to the branches of

knowledge based upon reasoning / And Śruti along with Smṛti

264

are Āgama"20 [ carācarāṇāṃ bhūtānāṃ pravṛttirlokasamjñitā |

hetuvidyātmako nyāyah sasmṛtih śrutirāgamaḥ ||] [3.163].

"If something is presented which is not in accord with

what is established regarding each of these -- due to an

error of the kavi. . . ." [ teṣu teṣvayathārūḍhaṃ yadi

kiṃcit pravartate | kaveḥ pramādaddeśādivirodhityetad-

ucyate ||] [3.164]. This doṣa thus subsumes all forms of

contradiction ("virodha") between what is evident in a

verse and what is conventionally established in a number of

areas. Daṇḍin lays out the following specific types:

(1) deśa virodhī [3.165-66] / "contradiction with

respect to place." As in, for example, "The Chola domains

are the lands along the banks of the Kāverī river / dark

with the black Aguru trees. . . ." [ colāḥ kālāguruśyāma-

kāverītīrabhūmayaḥ |] [3.166ab]. Yet at the probable time

of Daṇḍin's writing, at the height of Pallava rule, the

Cholas no longer controlled the banks of the Kāverī River,

nor do Aguru trees grow along its bank.

(2) kāla virodhī [3.167-169ab] / "contradiction with

265

respect to time." As in, for example, "The Padminī wakes-up

at night / The Kumudvatī blooms during the day / Spring

displays the blossomed Nicula / And summer has but cloud

days" [ padminī naktamunnidrā sphuṭatyahni kumudvatī |

madhurutphullaniculo nidāgho meghadurdinah |!] [3.167]. In

actuality, the "Padminī" opens during the day; the

"Kumudvatī" blooms at night; the "Nicula" grass blossoms

during the rainy season; and the "Summer days" are hot and

clear.

(3) kalā virodhī [3.169cd-171] / "contradiction with

respect to the "skills." As in, for example, "The

underlying bhāvas of the vīra and śṛṅgāras rasas are Anger

(krodha) and Wonder (vismaya) / The bhinnamārga proceeds

filled with the seven notes" [ vīraśṛṅgārayorbhāvau sthā-

yinau krodhavismayau | pūrṇasaptasvarah soyam bhinnamārgah

pravartate ||] [3.170].

In kalā virodhī we have contradiction with respect to

any of the sixty-four kalās or "skills." Thus the

underlying bhāva of vīra or the "heroic" rasa is not

"anger" but rather "resolve" (utsāha); the underlying bhāva

of śṛṅgāra or the "erotic" rasa is not "wonder" but rather

"love" (rati). Similarly, now with respect to classical

Indian music, the "bhinna mārga" ("broken path") utilizes

but one of the seven svaras or "notes" (ṣadja [sa] /

riṣabha [ri] / gāndhāra [ga] / madhyama [ma] / panchama

[pa] / dhaivata [dha] / and niṣāda [ni]) -- not all seven.

"In this way let contradiction within the sixty-four

kalās be properly imagined / Their character will become

clear in the kalā pariccheda" [ itthaṃ kalācatuhṣaṣṭi-

virodhaḥ sādhun nīyataṃ | tasyāḥ kalāparicchede rūpamāvir-

bhaviṣyati] [3.171]. This verse has been taken by some as

evidence for a fourth, lost chapter to the Kavyādarśa

("pariccheda" as "chapter"). Yet pariccheda also means

simply "section," "division" in general, and Daṇḍin may be

referring here to the various sections of other works that

present the various kalās. And too the kalās, especially

those concerned with music and the visual arts, are not

considered formally within the the extant central kāvya

śāstra texts -- but are rather found elsewhere in close

association with the exposition of drama, the theatre,

musical theory, and so on.

(4) loka virodhī [3.172-173ab] / "contradiction with

respect to conventional knowledge." As in, for example,

"The elephant has a wavy mane / The horse has sharp horns /

The Eraṇḍa tree has great strength / The Khadira tree is

without strength" [ ādhūtakesaro hastī tikṣṇaśṛṅgas-

turaṅgamaḥ | gurusāroyameraṇḍo niḥsaraṇ khadiadrumaḥ || ]

[3.172]. Loka virodhī is a doṣa stemming from error with

respect to conventional, "worldly" (loka) knowledge. Just

as the physical form of the elephant and the horse are

well-known, so would the Eraṇḍa ("Castor Oil") "tree" be

recognized as more or less a large shrub, hardly renowned

for strength. Where the Khadira tree, on the contrary, has

great strength, being used for the tips of ploughshares,

sword handles, pestles, the axle-pins of chariots, and so

on. "According to the Aitreya Brāmaṇa he who desires

heaven ought to make his sacrificial post of Khadira wood.

268

In the Aśvamedha yajña [sacrifice] of Daśaratha [ a

reference to the Ādikānda of the Rāmāyana, 14th sarga]

Khadīra wood was used in making a sacrificial post."21

(5) nyāya virodhī [3.173cd-176ab] / "contradiction

with respect to reasoning." "Contradiction with respect to

the branches of knowledge that are based upon reasoning

(hetu) will now be shown" [ virodho hetuvidyāsu nyāyākhyāsu

nidarśyate ||] [3.173cd]. As in, for example, "The Sugata

[Buddha] indeed spoke truly [in affirming] that the

saṃskāras are imperishable / Surely its so / For that

Cakora-eyed lady remains in my heart even now" [ satyamevāha

sugataḥ saṃskārānavinaśvarān | tathāhi sā cakorakṣī

sthitaivādyāpi me hṛdi ||] [3.174].

saṃskāras / memory or mental impressions, especially

those of previous states of reincarnation. In Buddhism, a

mental construction or image held to be real (unlike a

mirage) though in actuality without any true, inherent

existence.

269

cakora / a bird said to feed on moonbeams, and whose

beautiful eyes are said to turn red at the sight of

poisoned food.

Yet in truth the reasoning of the Sugata on the

"imperishability" of the saṃskāras is quite otherwise: "The

Tathāgata proclaims the truth (dharma) from the Middle

Position: Conditioned by ignorance (avidyā) are the

constructions (saṃskāras); the stopping of the constructions

is from the utter fading away and stopping of this

ignorance. . . . " (Samyuttanikāya [2.17]).22 And as Bechan

Jhā points out, drawing from the Sarvadarśanasaṅgraha of

Madḥava Āchārya, "The Buddhist's supposition is that all

things are momentary (sarvam kṣaṇikam) and all things are

transient (sarvam anityam)."23

(6) āgama virodhī [3.176cd-178] / "contradiction with

respect to āgama, that is here, "scripture" (śruti and smṛti

(see above). Āgama virodhī involves a contradiction with

respect to either śruti or smṛti. In [3.177] a violation of

the injunctions invoked by śruti is expressed -- a violation

of the proper sacrificial sequence, where the vaiśvānara

birth rite is being performed without the performance of the

agnyādhāra ceremony.

In [3.178] Dandin provides an example of fault arising

from a contradiction with smṛti: "Although not undergoing

the upanayana rite he studied the Vedas with his teacher / A

crystal -- naturally pure -- does not require further

refinement" [ asāvanupanītopi vedānadhījas gurōḥ |

svabhāvaśuddhaḥ sphāṭiko na saṃskāramapekṣate || ] [3.178].

Where according to the smṛti teachings, the performance of

the upanayana rite is mandatory before one commences the

study of the Vedas.

In the closing verses of the doṣa section (and of the

text itself) [3.179-85], Dandin offers a number of

exceptions where, as we have seen for all the preceding

doṣas, such contradiction need not be a defect given the

specific situation. "Through the skill of the kavi all of

these contradictions may sometimes go beyond being

considered faults / And enter the path of qualities (gunas)

[ virodhah sakalopyeṣa kadācit kavikauśalāt | utkramya

doṣagaṇanām guṇavīthīṃ vigāhate || ] [3.179].

In [3.181] we have an exception to kāla virodhī: "A

harsh wind -- harbinger of the destruction of kings -- is

shaking the pollen from the Kadamba flowers and the buds

from the Saptacchada trees" [ rājñāṃ vināśapiśunaścacāra

kharamārutaḥ | dhunvan kadambarājasaḥ saḥ saptacchadod-

gamān || ] [3.181]. Autumn is the time for military

expeditions and battle, a season whose implication in the

first two pādas of this verse is thus reinforced with the

"buds from the Saptacchada trees" -- a tree that blooms

only in the Autumn season. The anomaly or contradiction

here lies with the "pollen from the Kadamba flowers," which

should normally blossom only in the rainy season. Yet the

tone of this verse is a grim foreboding reflecting the

outset of a military campaign, stemming from the negative

omens marked by a "harsh wind" and plants blooming out of

season -- contradictions that appropriately develop the

situation and which are thus acceptable.

And in [3.183], for example, we have an exception to

loka virodhī: "A lover tormented with the distress of being

separated from his beloved / counts fire colder than

moonbeams" [ aindavādarciṣaḥ kāmi śiśiraṃ havyavāhanam |

abalāvirahakleśavihvalo gaṇayatyayam || ] [3.183]. To

emphasize the intensity of the suffering stemming from

separation, the kavi contradicts conventional knowledge of

the world -- any warmth from fire has died, a fire now felt

as colder than -- as traditionally considered -- moonbeams.

273

Notes [3.1] - [3.185]

1. Dandin, The Kāvyādarśa of Śrī Dandin, edited with a commentary by Premachandra Tarkabāgīśa (Calcutta, 1863); Reprint (Osnabrück: Biblio Verlag, 1981), pp. 377-78.

2. Siegfried Lienhard's approximation of "niyama" is thus incorrect: "The second of the two main types of citra poetry (literally, "variegated"; "difficult to compose") is poetry that limiting the number of phonetic classes employed. . . . In this sort of poem the author uses as many vowels as he wishes, but limits the choice of consonants to one, two, or only a few classes" (History of Classical Poetry, p. 157). As we see in Dandin's presentation, "restriction" applies to vowels and the points of articulation as well as to the consonants.

3. Louis Renou, "Sur la Notion de Brahman," Journal Asiatique, 237 (1947), p. 22.

4. Maurice Bloomfield, The Religion of the Veda (N. Y.: G. P. Putman's Sons, 1908), p. 215.

5. Siegfried Lienhard, History of Classical Poetry, pp. 150-51.

6. Marie-Claude Porcher, "On Prahelikā," in Ludwik Sternbach Felicitation Volume, edited by J. P. Sinha, vol. 1 (Lucknow: Akhila Bharatiya Sanskrit Parishad, 1979), p. 326.

7. Ludwik Sternbach, Indian Riddles: A Forgotten Chapter in the History of Sanskrit Literature (Hoshiarpur: Vishveshvaranand Vedic Research Institute, 1975), p. 41.

8. Marie-Claude Porcher, "On Prahelikā," p. 328.

9. Marie-Claude Pocher, "On Prahelikā," p. 327 and n. 6.

10. Porcher's and Sternback's analyses would appear doubtful. They take nr̥pā(h) ("kings") as singular and consider it to refer to a king "Pun̥d̥raka" [?]. (Marie-Claude Porcher, "On Prahelikā," p. 329; and Ludwik Sternbach, Indian Riddles, p. 44.)

11. I would question Gerow's analysis of this example [3.115]: "Vr̥ddhā is not a pun [śleṣa], for Lakṣmī bears that epithet in approximately the sense of 'the fully developed one.' The conundrum [Gerow's term for prahelikā] plays only on the legitimate connotations of the one word" (Glossary/213). Regardless of the etymologically derived meaning of vr̥ddhā, it stands as a "name" for Lakṣmī and thus is distinct from the nominalized "old woman." I would think that we have something other than connotations. We shall see in a number of Dan̥din's examples of śleṣa alamkāra that he considered this play between "nominal/Name" an instance of one word having more than one meaning and thus a legitimate instance of śleṣa.

12. P. C. Lahiri, Concepts of Rīti and Guṇa in Sanskrit Poetics in their Historical Development (Dacca: The University of Dacca, 1937), pp. 1-2.

13. P. C. Lahiri, Concepts of Rīti and Guṇa in Sanskrit Poetics, p. 3.

14. Anant Lal Thakur, "Members of an Indian Syllogism," in Ludwik Sternbach Felicitation Volume, edited by J. P. Sinha, part 1 (Lucknow: Akhila Bharatiya Sanskrit Parishad, 1979), p. 615.

15. Gautama, Nyāya: Gautama's Nyāya Sūtra with Vātsyāyana's Commentary, trans. by Mrinalkanti Gangopadhyaya (Calcutta: Indian Studies, 1982), p. 4.

16. Anant Lal Thakur, "Members of an Indian Syllogism," p. 615.

17. Gautama, Nyāya Sūtra, translated by Mrinalkanti Gangopadhyaya, (1982), p. 44.

18. A. K. Warder, Indian Kāvya Literature, vol. 1 (Delhi: Motilal Banarsidass, 1972), p. 85.

19. Bechan Jhā, Concept of Poetic Blemishes in Sanskrit Poetics (Varanasi: The Chowkhamba Sanskrit Office, 1965), p. 61.

20. śruti (literally, "heard") / The sacred literature held to have been "heard" by the ancient ṛsis, divinely revealed at the time of the world's creation. These include the four Vedas -- Ṛg, Yajur, Sāma, and Atharva -- and their primary categories -- Samhitā, Brāhmaṇa, Āraṇyaka, and Upaniṣad.

smṛti (literally, "remembered") / A group of texts "remembered" or passed on as traditional lore rather than divinely inspired. They include:

(1) sūtras (literally, "thread") / manuals of instruction in the form of prose aphorisms on ritual, law, and scriptural exigesis. These in turn include texts on: śrauta / community ritual; grhya / domestic ritual; dharma / law both religious and civil; śulva / the skills necessary in the building of altars (architecture, geometry, mathematics); pratiśākhya / the pronounciation of the Vedas; and vyākaraṇa / language and grammar.

(2) śāstras / post-Vedic compilations in verse explicating a given subject, often an explanation of an earlier sūtra. Such as, for example: the Dharmaśāstras, the Arthaśāstra of Kauṭilya, the Kāmaśasāstra of Vātsyāyana,

the Nītiśāstras, and indeed the various kāvya śāstras (the

nāṭyaśāstras and the alaṃkāraśāstras).

(3) the Mahābhārata and Rāmāyaṇa

(4) the eighteen major and the eighteen minor Purāṇas

(5) the various Tantras, and so on.

āgama (literally, "what has come (down)") / non-Vedic

religious texts; a traditional doctrine or discipline.

21. B. C. Law, "Ancient Indian Flora," Indian Culture,

vol. 15, n. 4 (1948-49), p. 132.

22. Padmanabh S. Jaini, "Śramaṇas: Their Conflict with

Brāhmanical Society," in Chapters in Indian Civilization,

edited by Joseph W. Elder, , rev. edition, vol. 1 (Joseph

W. Elder, 1970), p. 65.

23. Bechan Jha, Concept of Poetic Blemishes, p. 68.

See Madhava Āchārya, The Sarva-darśana-saṃgraha,

translated by E. B. Cowell and A. E. Gough, 6th edition

(Varanasi: The Chowkhamba Sanskrit Series, 1961 (1894)).

The

Verses

of

the

Second

Chapter

277

An Enumeration with English and Sanskrit Titles

2.1 Definition of the Alamkāra

[ Alamkāralakṣaṇam ]

2.2 The Intention of the Writer

[ Granthakaruh Āśayah ]

2.3 Indicating the Distinction between the Alamkāras

Previously Discussed and Those about to be Discussed

[ Uktavakṣyamānālamkārabhedanirūpaṇam ]

2.4- The Thirty-Five (Artha) Alamkāras.

2.7

2.8 Definition of Svabhāvokti Alamkāra

[ Svabhāvoktyalamkāralakṣaṇam ]

2.9 Example of the Svabhāvokti of Genus

[ Jāti Svabhāvoktyudāharaṇam ]

278

2.10 Example of the Svabhāvokti of Action

[ Kriyā Svabhāvoktyudāharaṇam ]

2.11 Example of the Svabhāvokti of Attribute

[ Guṇa Svabhāvoktyudāharaṇam ]

2.12 Example of the Svabhāvokti of an Individual

[ Dravya Svabhāvoktyudāharaṇam ]

2.13 Conclusion to Svabhāvokti Alaṅkāra

[ Svabhāvoktyalaṅkāropasaṃhāraḥ ]

2.14 Definition of Upamā alaṅkāra

[ Upamālakṣaṇam ]

2.15 The Upamā of Attribute

[ Dharma Upamā ]

2.16 The Upamā of Objects

[ Vastu Upamā ]

2.17 The Upamā of Transposition

[ Viparyāsa Upamā ]

2.18 The Upamā of Reciprocity

[ Anyonya Upamā ]

2.19 The Upamā of Restriction

[ Niyama Upamā ]

2.20 The Upamā of Non-Restriction

[ Aniyama Upamā ]

2.21 The Upamā of Conjunction

[ Samuccaya Upamā ]

2.22 The Upamā of Intensity

[ Atiśaya Upamā ]

2.23 The Upamā of Imagination

[ Utprekṣitā Upamā ]

2.24 The Upamā of the Wondrous

[ Adbhuta Upamā ]

2.25 The Upamā of Confusion

[ Moha Upamā ]

2.26 The Upamā of Doubt

[ Samśaya Upamā ]

2.27 The Upamā of Resolution

[ Nirṇaya Upamā ]

2.28 The Upamā of Multiple Embrace

[ Śleṣa Upamā ]

2.29 The Upamā of the Uniform

[ Samāna Upamā ]

2.30 The Upamā of Depreciation

[ Nindā Upamā ]

2.31 The Upamā of Appreciation

[ Praśaṃsā Upamā ]

2.32 The Upamā involving a Wish to Express

[ Ācikhyāsā Upamā ]

2.33 The Upamā of Rivalry

[ Virodha Upamā ]

2.34 The Upamā of Negation

[ Pratiṣedha Upamā ]

2.35 The Upamā of Flattery

[ Caṭu Upamā ]

2.36 The Upamā Expressing the Actual

[ Tattvākhyāna Upamā ]

2.37 The Upamā of the Unique

[ Asādhāraṇa Upamā ]

2.38 The Upamā of the Non-Existent

[ Abhūta Upamā ]

2.39 The Upamā of the Inconceivable

[ Asambhāvita Upamā ]

2.40 The Upamā of the Multiple

[ Bahu Upamā ]

2.41 The Upamā of Transformation

[ Vikriyā Upamā ]

2.42 The Upamā of the Interwoven

[ mālā Upamā ]

2.43 The Upamā of Complete Expressions

[ Vakyārtha Upamā ]

2.44 Example of the Upamā of Complete Expressions: I.

[ Vakyārtha Upamodāharaṇam: I. ]

2.45 Example of the Upamā of Complete Expressions: II.

[ Vākyārtha Upamodāharaṇam: II. ]

2.46 The Upamā of Parallel Objects

[ Prativastu Upamā ]

2.47 Example of the Upamā of Parallel Objects

[ Prativastu Upamā udāharaṇam ]

2.48 The Upamā of Equalization

[ Tulyayoga Upamā ]

2.49 Example of the Upamā of Equalization

[ Tulyayoga Upamodāharaṇam ]

2.50 The Upamā of Cause

[ Hetu Upamā ]

2.51 Exceptions to Faults in Upamās

[ Upamādoṣāpavadah ]

2.52 Examples of Exceptions to Faults in Gender and Number

[ Liṅgavacanadoṣāpavadodāharaṇāni ]

2.53 Examples to Exceptions to Faults in Inferiority/Superiority

[ Hīnādhikatādoṣāpavadodāharaṇāni ]

2.54 Conclusion to Exceptions to Faults in Upamās / Indicating Examples of Faults in Upamās

[ Upamādoṣāpavādopasam̐hāraḥ /

Upamādoṣodāharaṇasūcanam ]

285

2.55 Examples of Faults in Upamās

[ Upamādoṣodāharaṇam ]

2.56 Conclusion to Faults in Upamās

[ Upamādoṣopasam̐hāraḥ ]

2.57- Particles, Words, and Expressions Indicating

2.65 Similarity in Upamās

[ Upamāsādrśya sūcinaḥ śabdāḥ ]

2.66 Definition of Rūpaka Alamkāra / Examples of the

Compounded Rūpaka

[ Rūpakālamkāralakṣaṇam / Samasta

Rūpakodāharaṇāni ]

2.67 Example of the Uncompounded Rūpaka

[ Asamasta Rūpakodāharaṇam ]

2.68 Specification of the Compounded and Uncompounded

Rūpakas / The Compounded/Uncompounded Rūpaka

[ Samastavyastayoḥ Rūpakayoḥ Nirdeśaḥ /

Samastavyasta Rūpakam ]

2.69 Example of the Complete Rūpaka

[ Sakala Rūpakodāharaṇam ]

2.70 Explication of the Example of the Complete Rūpaka

[ Sakala Rūpakodāharaṇasvarūpaprakāśanam ]

2.71 Example of the Rūpaka of Attributes

[ Avayava Rūpakodāharaṇam ]

2.72 Explication of the Example of the Rūpaka of

Attributes

[ Avayava Rūpakodāharaṇasvarūpaprakāśanam ]

2.73 Example of the Rūpaka of the Aggregate

[ Avayavi Rūpakodāharaṇam ]

2.74 The Explication of the Example of the Rūpaka of

the Aggregate

[ Avayavi Rūpakodāharaṇasvarūpaprakāśanam ]

2.75 Example of the Rūpaka of One-Attribute

[ Eka aṅga Rūpakodāharaṇam ]

2.76 The Rūpaka of One-Attribute

[ Eka aṅga Rūpakam ]

2.77 The Rūpaka of Congruity

[ Yukta Rūpakam ]

2.78 The Rūpaka of Incongruity

[ Ayukta Rūpakam ]

2.79 The Rūpaka of the Uneven

[ Viṣama Rūpakam ]

2.80 Example of the Rūpaka of the Uneven

[ Viṣama Rūpakodāharaṇam ]

2.81 Example of the Rūpaka of Attribution

[ Saviśeṣaṇa Rūpakodāharaṇam ]

2.82 Explication of the Example of the Rūpaka of

Attribution

[ Saviśeṣaṇa Rūpakodāharaṇasvarūpaprakāśanam ]

2.83 Example of the Rūpaka of the Incongruous

[ Viruddha Rūpakodāharaṇam ]

2.84 Explication of the Example of the Rūpaka of the

Incongruous

[ Viruddha Rūpakodāharaṇasvarūpaprakāśanam ]

2.85 Example of the Rūpaka of Cause

[ Hetu Rūpakodāharaṇam ]

2.86 Explication of the Example of the Rūpaka of

Causality

[ Hetu Rūpakodāharaṇasvarūpaprakāśanam ]

2.87 The Rūpaka of Multiple Embrace

[ Śliṣṭa Rūpakam ]

2.88 The Rūpaka of Similarity and the Rūpaka of

289

Disparity

[ Upamā Rūpakam Vyatireka Rūpakam ca ]

2.89 Example of the Rūpaka of Similarity

[ Upamā Rūpakodāharaṇam ]

2.90 Example of the Rūpaka of Disparity

[ Vyatireka Rūpakodāharaṇam ]

2.91 The Rūpaka of Denial

[ Ākṣepa Rūpakam ]

2.92 The Rūpaka of Rationalization

[ Samādhāna Rūpakam ]

2.93 The Rūpaka of Transference

[ Rūpaka Rūpakam ]

2.94 Example of the Rūpaka Concealing the Actual

[ Tattvāpahnava Rūpakodāharaṇam ]

2.95 Explication of the Example of the Rūpaka

Concealing the Actual

290

[ Tattvāpahnava Rūpakodāharaṇasvarūpa-

prakāśanam ]

2.96 Conclusion to Upamā and Rūpaka Alamkāras

[ Upamārūpakālaṃkaropasaṃhāraḥ ]

2.97 Definition of Dīpaka alaṃkāra

[ Dīpakālaṃkāralakṣaṇam ]

2.98 Example of the Dīpaka of Genus (in Initial Position)

[ Jāti (Ādi) Dīpakodāharaṇam ]

2.99 Example of the Dīpaka of Action (in Initial Position)

[ Kriyā (Ādi) Dīpakodāharaṇam ]

2.100 Example of the Dīpaka of Attribute (in Initial Position)

[ Guṇa (Ādi) Dīpakodāharaṇam ]

2.101 Example of the Dīpaka of an Individual (in Initial Position)

[ Dravya (Ādi) Dīpakodāharaṇam ]

2.102 The Explicit Indication of the Initial, Medial, and Final Positions of Dīpaka

[ Ādimadyānta Dīpakasūcanam ]

2.103 Example of the Dīpaka of Genus (in Medial Position)

[ Jāti (Madhya) Dīpakodāharaṇam ]

2.104 Example of the Dīpaka of Action (in Medial Position)

[ Kriyā (Madhya) Dīpakodāharaṇam ]

2.105 Example of the Dīpaka of Genus (in Final Position)

[ Jāti (Anta) Dīpakodāharaṇam ]

2.106 Example of the Dīpaka of Action (in Final Position)

[ Kriyā (Anta Dīpakodāharaṇam ]

2.107 Example of the Interwoven Dīpaka

[ Mālā Dīpakodāharaṇam ]

2.108 The Interwoven Dīpaka

[ Mālā Dīpakam ]

2.109 Example of the Dīpaka of Opposite Meanings

[ Viruddha Artha Dīpakodāharaṇam ]

2.110 The Dīpaka of Opposite Meanings

[ Viruddha Artha Dīpakam ]

2.111 Example of the Dīpaka of Uniform Meaning

[ Eka Artha Dīpakodāharaṇam ]

2.112 The Dīpaka of Uniform Meaning

[ Eka Artha Dīpakam ]

2.113 Example of the Dīpaka of Multiple Embrace

[ Śliṣṭa Artha Dīpakodāharaṇam ]

2.114 Explication of the Example of Dīpaka of Multiple

Embrace

[ Śiṣṭa Artha Dīpakam ]

2.115 Conclusion to Dīpaka Alamkāra

[ Dīpakālamkāropasamhārah ]

2.116 Definition of Āvrtti Alamkāra

[ Āvrttyalamkāralakṣaṇam ]

2.117 Example of the Āvrtti of Sense

[ Artha Āvrttyudāharaṇam ]

2.118 Example of the Āvrtti of Word

[ Pada Āvrttyudāharaṇam ]

2.119 Example of the Āvrtti of Both Sense and a Word

[ Arthanadobhayoh Āvrttyudāharaṇam ]

2.120 Definition of Ākṣepa alamkāra

[ Ākṣepālamkāralakṣaṇam ]

2.121 Example of the Ākṣepa of the Past

[ Vṛtta Ākṣepodāharaṇam ]

2.122 The Ākṣepa of the Past

[ Vṛtta Ākṣepaḥ ]

2.123 Example of the Ākṣepa of the Present

[ Vartamāna Ākṣepodāharaṇam ]

2.124 The Ākṣepa of the Present

[ Vartamāna Ākṣepaḥ ]

2.125 Example of the Ākṣepa of the Future

[ Bhāviṣyat Ākṣepodāharaṇam ]

2.126 The Ākṣepa of the Future

[ Bhāviṣyat Ākṣepaḥ ]

2.127 Example of the Ākṣepa of Attribute

[ Dharma Ākṣepodāharaṇam ]

2.128 The Ākṣepa of Attribute

[ Dharma Ākṣepaḥ ]

2.129 Example of the Ākṣepa of the Basis of Attribution

[ Dharmin Ākṣepodāharaṇam ]

2.130 The Ākṣepa of the Basis of Attribution

[ Dharmin Ākṣepaḥ ]

2.131 Example of the Ākṣepa of Efficient Cause

[ Kāraṇa Ākṣepodāharaṇam ]

2.132 The Ākṣepa of Efficient Cause

[ Kāraṇa Ākṣepaḥ ]

2.133 Example of the Ākṣepa of Effect

[ Kārya Ākṣepodāharaṇam ]

2.134 The Ākṣepa of Effect

[ Kārya Ākṣepaḥ ]

2.135 Example of the Ākṣepa through Permission

[ Anujñā Ākṣepodāharaṇam ]

2.136 The Ākṣepa through Permission

[ Anujñā Ākṣepaḥ ]

2.137 Example of the Ākṣepa through Authority

[ Prabhutva Ākṣepodāharaṇam ]

2.138 The Ākṣepa through Authority

[ Prabhutva Ākṣepaḥ ]

2.139 Example of the Ākṣepa through Indifference

[ Anādara Ākṣepodāharaṇam ]

2.140 The Ākṣepa through Indifference

[ Anādara Ākṣepaḥ ]

2.141 Example of the Ākṣepa of Benediction

[ Āśīrvacana Ākṣepodāharaṇam ]

2.142 The Ākṣepa through Benediction

[ Āśīrvacana Ākṣepaḥ ]

2.143 Example of the Ākṣepa through Harshness

[ Paruṣa Ākṣepodāharaṇam ]

2.144 The Ākṣepa through Harshness

[ Paruṣa Ākṣepaḥ ]

2.145 Example of the Ākṣepa through Counsel

[ Ācivya Ākṣepodāharaṇam ]

2.146 The Ākṣepa through Counsel

[ Sācivya Ākṣepah ]

2.147 Example of the Ākṣepa through Effort

[ Yatna Ākṣepodāharaṇam ]

2.148 The Ākṣepa through Effort

[ Yatna Ākṣepah ]

2.149 Example of the Ākṣepa through Control of Another

[ Paravaśa Ākṣepodāharaṇam ]

2.150 The Ākṣepa through Control of Another

[ Paravaśa Ākṣepah ]

2.151 Example of the Ākṣepa through an Impossible

Expedient

[ Upāya Ākṣepodāharaṇam ]

2.152 The Ākṣepa through an Impossible Expedient

[ Upāya Ākṣepah ]

2.153 Example of the Ākṣepa through Anger

[ Roṣa Ākṣepodāharaṇam ]

2.154 The Ākṣepa through Anger

[ Roṣa Ākṣepah ]

2.155- Interpolations

2.156

2.157 Example of the Ākṣepa of Compassion

[ Anukrośa Ākṣepodāharaṇam ]

2.158 The Ākṣepa of Compassion

[ Anukrośa Ākṣepah ]

Note: The following three varieties are in what I

believe to be the more accurate order; the numbering of

Rangacharya Raddi's text is yet retained.

2.161 Example of the Ākṣepa of Regret

[ Anuśaya Ākṣepodāharaṇam ]

2.162 The Ākṣepa of Regret

[ Anuśaya Ākṣepah ]

2.163 Example of the Ākṣepa of Doubt

[ Samśaya Ākṣepodāharaṇam ]

2.164 The Ākṣepa of Doubt

[ Samśaya Ākṣepah ]

2.159 Example of the Ākṣepa of Multiple Embrace

[ Śliṣṭa Ākṣepodāharaṇam ]

2.160 The Ākṣepa of Multiple Embrace

[ Śliṣṭa Ākṣepah ]

2.165 Example of the Ākṣepa through Analogous

Corroboration

[ Artha antara Ākṣepodāharaṇam ]

299

2.166 The Ākṣepa through Analogous Corroboration

[ Artha antara Ākṣepah ]

2.167 Example of the Ākṣepa with Cause

[ Hetu Ākṣepodāharaṇam ]

2.168 The Ākṣepa with Cause / Conclusion to Ākṣepa Alamkāra

[ Hetu Ākṣepah / Ākṣepālamkāropasamhārah ]

2.169 Definition of Arthāntaranyāsa Alamkāra

[ Arthāntaranyāsālamkāralakṣaṇam ]

2.170 The Varieties of Arthāntaranyāsa Alamkāra

[ Arthāntaranyāsālamkārabhedāḥ ]

2.171 Illuminating the Varieties of Arthāntaranyāsa Alamkāra

[ Arthāntaranyāsālamkārabhedaprakāśanam ]

2.172 Example of the Arthāntaranyāsa involving

301

Universal Corroboration

[ Viśvavyāpī Arthāntaranyāsodāharaṇam ]

2.173 Example of the Arthāntaranyāsa involving

Specific Corroboration

[ Viśeṣasthaḥ Arthāntaranyāsodāharaṇam ]

2.174 Example of the Arthāntaranyāsa involving

Multiple Embrace

[ Śleṣaviddhaḥ Arthāntaranyāsodāharaṇam ]

2.175 Example of the Arthāntaranyāsa involving

Contradiction

[ Virodhavān Arthāntaranyāsodāharaṇam ]

2.176 Example of the Arthāntaranyāsa involving

Inappropriate Correspondence

[ Ayuktakārī Arthāntaranyāsodāharaṇam ]

2.177 Example of the Arthāntaranyāsa involving

Appropriate Correspondence

[ Yuktātmā Arthāntaranyāsodāharaṇam ]

2.178 Example of the Arthāntaranyāsa involving

Appropriate - Inappropriate Correspondence

[ Yuktāyukta Arthāntaranyāsodāharanam ]

2.179 Example of the Arthāntaranyāsa involving

Inappropriate - Appropriate Correspondence

[ Viparyaya Arthāntaranyāsodāharanam ]

2.180 Definition of Vyatireka Alamkāra

[ Vyatirekālamkāralakṣanam ]

2.181 Example of the Vyatireka involving a Single Object

[ Eka Vyatirekodāharanam ]

2.182 The Vyatireka involving a Single Object

[ Eka Vyatirekah ]

2.183 Example of the Vyatireka involving Two Objects

[ Ubhaya Vyatirekodāharanam ]

2.184 The Vyatireka involving Two Objects

[ Ubhaya Vyatirekah ]

2.185 Example of the Vyatireka of Multiple Embrace

[ Saśleṣa Vyatirekodāharaṇam ]

2.186 The Vyatireka of Multiple Embrace / Introduction

To the Vyatirekas of Denial and Cause

[ Saśleṣa Vyatirekaḥ / Sākṣepa Sahetu

Vyatirekopakramaṇ ]

2.187 Example of the Vyatireka of Denial

[ Sākṣepa Vyatirekodāharaṇam ]

2.188 Example of the Vyatireka of Cause

[ Sahetu Vyatirekodāharaṇam ]

2.189 Conclusion to the Varieties of Vyatireka where

Similarity is Explicit / Introduction to the

Varieties of Vyatireka where Similarity is

Implicit

304

[ Śabdopādānasādṛśya Vyatirekopasamhārah /

Pratīyamānasādṛśya Vyatirekopakramah ]

2.190 Example of the Vyatireka involving Difference Alone

[ Bhedamātra Vyatirekodāharaṇam ]

2.191 Example of the Vyatireka involving Superiority

[ Ādhikya Vyatirekodāharaṇam ]

2.192 The Vyatirekas involving Difference Alone

and Superiority / Introduction to the

Vyatireka involving Similarity in Difference

[ Bhedamātra Ādhikya Vyatirekau / Sadrśa

Vyatirekopakramah ]

2.193 Example of the Vyatireka of Similarity in

Difference with the Similarity Expressed

[ Śabdopādānasādṛśya Sadrśavyatirekodāharaṇam ]

2.194 Example of the Vyatireka of Similarity in

Difference with the Similarity Implicit

[ Pratīyamānasādṛśya Sadrśavyatirekodāharaṇam ]

2.195 Explication of the example of the Vyatireka of

Similarity in Difference with the Similarity

Implicit

[ Pratīyamānāsādrśya Sadrśavyatirekodāharaṇa-

svarūpaprakāśanam ]

2.196 Conclusion to and Explication of the Vyatirekas of

Similarity in Difference

[ Sadrśavyatirekaprakāśanopasaṃhāraḥ ]

2.197 Example of the Vyatireka of Species

[ Svajāti Vyatirekodāharaṇam ]

2.198 The Vyatireka of Species

[ Svajāti Vyatirekaḥ ]

2.199 Definition of Vibhāvanā Alamkāra

[ Vibhāvanālamkāralakṣaṇam ]

2.200 Example of the Vibhāvanā involving Another Cause

[ Kāraṇāntara Vibhāvanodāharaṇam ]

2.201 Example of the Vibhāvanā involving

Characteristic Condition

[ Svābhāvika Vibhāvanodāharaṇam ]

2.202 Explication of the Examples of the Vibhāvanās

Involving Another Cause and Characteristic

Condition

[ Kāraṇāntara Svābhāvika Vibhāvanodāharaṇa-

svarūpaprakāśanam ]

2.203 Example of the Vibhāvanā involving Explicit

Characteristic Condition

[ Śabda Svābhāvika Vibhāvanodāharaṇam ]

2.204 The Vibhāvanā involving Explicit

Characteristic Condition

[ Śabda Svābhāvika Vibhāvanā ]

2.205 Definition of Samāsokti [Samāsa] Alaṃkāra

[ Samāsoktyalaṃkāralakṣaṇam ]

2.206

Example of Samāsokti as Such

[ Samāsokti Svarūpodāharaṇam ]

2.207

Explanation of the Example of Samāsokti as Such

[ Samāsokti Svarūpodāharaṇasvarūpaprakāśanam ]

2.208

Introduction to the Samāsoktis of Equivalent

Application and Equivalent and Differential

Application

[ Tulyākāraviśeṣaṇa Bhinnābhinnaviśeṣaṇa

Samāsoktyupakramaḥ ]

2.209

Example of the Samāsokti of Equivalent Application

[ Tulyākāraviśeṣaṇa Samāsokyudāharaṇam ]

2.210

Example of the Samāsokti of Equivalent and

Differential Application

[ Bhinnābhinnaviśeṣaṇa Samāsoktyudāharaṇam ]

2.211

The Samāsoktis of Equivalent Application and

Equivalent and Differential Application

308

[ Tulyākāraviśeṣaṇa Bhinnābhinnaviśeṣaṇa Samāsoktī ca ]

2.212 Example of the Samāsokti of the Unusual

[ Apūrva Samāsoktyudāharaṇam ]

2.213 The Samāsokti of the Unusual

[ Apūrva Samāsoktiḥ ]

2.214 Definition of Atiśayokti [Atiśaya] Alaṃkāra

[ Atiśayoktyalaṃkāralakṣaṇam ]

2.215 Example of Atiśayokti as Such

[ Atiśayokti Svarūpodāharaṇam ]

2.216 Explanation of the Example of Atiśayokti as Such / Introduction to Further Varieties

[ Atiśayokti Svarūpodāharaṇasvarūpaprakāśanam / Atiśayokti Prabhedopakramaḥ ]

2.217 Example of the Atiśayokti of Doubt

[ Samśaya Atiśayoktyudāharanam ]

2.218 Example of the Atiśayokti of Resolution

[ Nirṇaya Atiśayoktyudāharanam ]

2.219 Example of the Atiśayokti of Inclusive Relationship

[ Ādaya/Ādhāra Atiśayoktyudāharanam ]

2.220 Conclusion to Atiśayokti Alamkāra

[ Atiśayoktyalamkāropasamhārah ]

2.221 Definition of Utprekṣā Alamkāra

[ Utprekṣālamkāralakṣaṇam ]

2.222 Example of the Utprekṣā involving

A Sentient Subject

[ Cetana Utprekṣodāharanam ]

2.223 Explication of the Example of the Utprekṣa

310

Involving a Sentient Subject

[ Cetana Utpreksodāharanasvarūpaprakāśanam ]

2.224 Example of the Utprekṣā involving

An Insentient Subject

[ Acetana Utprekṣodāharanam ]

2.225 Explication of the Example of the Utprekṣā

Involving an Insentient Subject

[ Acetana Utprekṣodāharanasvarupaprakāśanam ]

2.226 - Establishing the Distinction between

2.234 Utprekṣā and Upamā

[ Utprekṣopamābhedasādhanam ]

2.235 Introduction to Hetu / Sūkṣma / and Leśa

Alamkāras / Definition of Hetu Alamkāra

[ Hetusūkṣmaleśopakramah \ Hetvalamkāra-lakṣanam ]

2.236

Example of the Hetu of Production involving

A Directly Generated Positive Effect

[ Nirvartyābhāvakārya Kārakahetūdāharaṇam ]

2.237

The Hetu of Production involving a Directly

Generated Positive Effect

[ Nirvartyābhāvakārya Kārakahetuḥ ]

2.238

Example of the Hetu of Production involving

A Directly Generated Negative Effect

[ Nirvartyābhāvakārya Kārakahetūdāharaṇam ]

2.239

Explication of the Example of the Hetu of Production

Involving a Directly Generated Negative Effect

[ Nirvartyābhāvakārya Kārakahetūdāharaṇasva-

rūpaprakāśanam ]

2.240

Causality with Reference to the Three Categories

Of Direct Object

[ Karmatrayaviṣayakahetutā ]

312

2.241 On the Preceding Variety of Hetu and the Varieties

That are to Immediately Follow

[ Uktānuktahetuprabhedavivecanam ]

2.242 Example of the Hetu of Production involving

Transformation

[ Vikārya Kārakahetūdāharanam ]

2.243 Example of the Hetu of Production involving

Contact

[ Prāpya Kārakahetūdāharanam ]

2.244 Example of the Implicit Hetu of Indication

[ Sūcyajñāpya Jñāpakahetūdāharanam ]

2.245 Example of the Explicit Hetu of Indication

[ Vācyajñāpya Jñāpakahetūdāharanam ]

2.246 Conclusion to the Hetus of Indication /

Introduction to the Hetus involving Non-Existence

[ Jñāpakahetūpasamharah / Abhāvahetūpakramah ]

2.247 Example of the Hetu of Prior Non-Existence

[ Prāgabhāva Hetūdaharanam ]

2.248 Example of the Hetu of Non-Existence involving Destruction

[ Pradhvamsa Abhāvahetūdāharanam ]

2.249 Example of the Hetu of Reciprocal Non-Existence

[ Anyonya Abhāvahetūdāharanam ]

2.250 Example of the Hetu of Absolute Non-Existence

[ Atyanta Abhāvahetūdāharanam ]

2.251 Example of the Hetu of Non-Existence involving A Double Negative

[ Prāgabhāva Abhāvahetūdāharanam ]

2.252 Conclusion to the Hetus of Non-Existence

[ Abhāva Hetūpasaṃhārah ]

2.253 The Varieties of the Hetu of the Marvelous

[ Citra Hetuprabhedāḥ ]

2.254 The Hetu of the Marvelous

[ Citra Hetuḥ ]

2.255 Example of the Hetu of the Marvelous involving

An Effect at a Distance

[ Dūrakārya Citrahetūdāharaṇam ]

2.256 Example of the Hetu of the Marvelous involving

A Simultaneous Effect

[ Kāryasahaja Citrahetūdāharaṇam ]

2.257 Example of the Hetu of the Marvelous involving

A Preceding Effect

[ Kāryāntaraja Citrahetūdāharaṇam ]

2.258 Example of the Hetu of the Marvelous involving

An Incongruous Effect

[ Ayuktakārya Citrahetūdāharaṇam ]

2.259 Example of the Hetu of the Marvelous involving

A Congruous Effect

[ Yuktakārya Citrahetūdāharaṇam ]

2.260 Conclusion to Hetu Alamkāra / Definition of Sūkṣma Alamkāra

[ Hetu Alamkāropasam̐hāraḥ / Sūkṣmālamkāra-lakṣaṇam ]

2.261 Example of the Sūkṣma of Gesture

[ Iṅgita Sūkṣmodāharaṇam ]

2.262 Explication of the Example of the Sūkṣma of Gesture

[ Iṅgita Sūkṣmodāharaṇasvarūpaprakāśanam ]

2.263 Example of the Sūkṣma of Appearance

[ Ākāra Sūkṣmodāharaṇam ]

2.264 Explication of the Example of the Sūkṣma of Appearance

[ Ākāra Sūkṣmodāharaṇasvarūpaprakāśanam ]

2.265 Definition of Leśa [Lava] Alamkāra (I.)

[ Leśālamkāralakṣaṇam ]

2.266 Example of Leśa Alamkāra

[ Leśalamkārodāharaṇam ]

2.267 Another Example of Leśa Alamkāra

[ Aparam Leśodāharaṇam ]

2.268 Another Definition of Leśa [Lava] Alamkāra (II.)

[ Aparam Leśalakṣaṇam ]

2.269 Example of Leśa involving

Censure through Praise

[ Stutynindāyāḥ Leśodāharaṇam ]

2.270 Explication of the Example of Leśa involving

Censure through Praise

[ Stutynindāyāḥ Leśodāharaṇasvarūpaprakāśanam ]

2.271 Example of Leśa involving Praise through Censure

[ Nindāstutyāḥ Leśodāharaṇam ]

2.272 Explication of the example of Leśa involving

317

Praise through Censure

[ Nindāstutyāh Leśodāharaṇasvarūpaprakāśanam ]

2.273 Definition of Yathāsamkhya [Krama/Saṃkhyāna]

Alamkāra

[ Yathāsaṃkhyālamkāralakṣaṇam ]

2.274 Example of Yathāsaṃkhya Alamkāra

[ Yathāsaṃkhyodāharaṇam ]

2.275 Definitions of Preyas / Rasavat / Ūrjasvin

Alamkāras

[ Preyorasavadūrjasvyalamkāralakṣaṇāni ]

2.276 Example of Preyas Alamkāra

[ Preyas Udāharaṇam ]

2.277 Explication of an Example of Preyas Alamkāra

[ Preyas Udāharaṇasvarūpaprakāśanam ]

2.278

Another Example of Preyas Alamkāra

[ Āparam Preyas Udāharanam ]

2.279

Explication of an Example of Preyas Alamkāra

[ Preyas Udāharanasvarūpaprakāśanam ]

2.280

Example of the Rasavat involving Śṛṅgāra Rasa

[ Śṛṅgārarasa Rasavadudāharanam ]

2.281

Explication of the Example of Rasavat involving

Śṛṅgāra Rasa

[ Śṛṅgārarasa Rasavadudāharanasvarūpaprakāśanam ]

2.282

Example of the Rasavat involving Raudra Rasa

[ Raudrarasa Rasavadudāharanam ]

2.283

Explication of the Example of Rasavat involving

Raudra Rasa

[ Raudrarasa Rasavadudāharanasvarūpaprakāśanam ]

2.284

Example of Rasavat involving Vīra Rasa

[ Vīrarasa Rasavadudāharanam ]

2.285 Explication of the Example of Rasavat involving

Vīra Rasa

[ Vīrarasa Rasavudāharaṇasvarūpaprakāśanam ]

2.286 Example of the Rasavat involving Karuṇa Rasa

[ Karuṇarasa Rasavudāharaṇam ]

2.287 Explication of the Example of the Rasavat involving

Karuna Rasa / Indicating the Form of Rasavat

Alamkāra involving the Remaining Rasas

[ Karuṇarasa Rasavudāharaṇasvarūpaprakāśanam /

Aparararasavadalaṃkārasvarūpasūcanam ]

2.288 Example of the Rasavat involving Bībhatsa Rasa

[ Bībhatsarasa Rasavudāharaṇam ]

2.289 Example of the Rasavat involving Hāsya Rasa

[ Hāsyarasa Rasavudāharaṇam ]

2.290 Example of the Rasavat involving Adbhuta Rasa

[ Adbhutarasa Rasavudāharaṇam ]

2.291 Example of the Rasavat involving Bhayānaka Rasa

[ Bhayānakarasa Rasavudāharaṇam ]

2.292 The Distinction between Rasa in Mādhurya Guṇa and

Rasa in Rasavat Alamkāra

[ Mādhuryaguṇe Rasasya Rasavadalamkāre

Rasasya ca Bhedaḥ ]

2.293 Example of Ūrjasvin Alamkāra

[ Ūrjasvin Alamkārodāharaṇam ]

2.294 Ūrjasvin Alamkāra

[ Ūrjasvin Alamkāra ]

2.295 Definition of Paryāyokta Alamkāra

[ Paryāyoktalamkāralakṣaṇam ]

2.296 Example of Paryāyokta Alamkāra

[ Paryāyoktalamkārodāharaṇam ]

2.297 Explication of the Example of Paryāyokta Alamkāra

[ Paryāyoktodāharanasvarūpaprakāśanam ]

2.298 Definition of Samāhita Alamkāra

[ Samāhitālamkāralakṣanam ]

2.299 Example of Samāhita Alamkāra

[ Samāhitālamkārodāharanam ]

2.300 Definition of Udātta Alamkāra

[ Udāttālamkāralakṣanam ]

2.301 Example of the Udātta of Character

[ Āśaya Udāttodāharanam ]

2.302 Example of the Udātta of Wealth

[ Vibhūti Udāttodāharanam ]

2.303 Explication of the Examples of Udātta Alamkāra

[ Udāttodāharanadvayasvarūpaprakāśanam ]

2.304 Definition of Apahnuti Alamkāra / Example of

Apahnuti as Such

[ Apahnutyalaṁkāralakṣaṇam / Apahnuti

Svarūpodāharaṇam ]

2.305 Example of the Apahnuti of Restricted Scope

[ Viṣaya Apahnutyudāharaṇam ]

2.306 Explication of the Apahnuti of Restricted Scope

[ Viṣaya Apahnutyudāharaṇasvarūpaprakāśanam ]

2.307 Example of the Apahnuti of Specific Nature

[ Svarūpa Apahnutyudāharaṇam ]

2.308 Explication of the Example of the Apahnuti of

Specific Nature

[ Svarūpa Apahnutyudāharaṇasvarūpaprakāśanam ]

2.309

Conclusion to Apahnuti Alamkāra

[ Apahnutyalamkāropasamhārah ]

2.310

Definition of śleṣa [śliṣṭa] Alamkāra and the

Illumination of Its Fundamental Categories

[ Śleṣālamkāralakṣaṇam Tadbhedaprakāśanamca ]

2.311

Example of the śleṣa of Integral Words

[ Abhinnapada śleṣodāharaṇam ]

2.312

Example of the śleṣa of Divisible Words

[ Bhinnapada śleṣodāharaṇam ]

2.313

Indicating the Varieties of śleṣa Previously

Mentioned

[ Uktaśleṣabhedasūcanam ]

2.314

Indicating the Varieties of śleṣa yet Unmentioned

[ Anuktaśleṣabhedasūcanam ]

324

2.315 Indicating the Varieties of śleṣa yet Unmentioned

[ Anuktaślesabhedasūcanam ]

2.316 Example of the śleṣa involving Integral Action

[ Abhinnakriyā śleṣodāharaṇam ]

2.317 Example of the śleṣa involving Congruous Action

[ Aviruddhakriyā śleṣodāharaṇam ]

2.318 Example of the śleṣa involving Incongruous Action

[ Viruddhakarman śleṣodāharaṇam ]

2.319 Example of the śleṣa involving Restriction

[ Niyamavat śleṣodāharaṇam ]

2.320 Example of the śleṣa involving the Negation of Restriction

[ Niyamākṣepa śleṣodāharaṇam ]

2.321 Example of the śleṣa involving Congruous Meanings

[ Avirodhin śleṣodāharaṇam ]

2.322 Example of the śleṣa involving Incongruous Meanings

[ Virodhin śleṣodāharanam ]

2.323 Definition of Viśeṣokti [Viśeṣa] Alamkāra

[ Viśeṣoktyalamkāralakṣaṇam ]

2.324 Example of the Viśeṣokti involving

A Deficiency of Attribute

[ Guṇavaikalya Viśeṣoktyudāharaṇam ]

2.325 Example of the Viśeṣokti involving

A Deficiency of Genus

[ Jātivaikalya Viśeṣoktyudāharaṇam ]

2.326 Example of the Viśeṣokti involving

A Deficiency of Action

[ Kriyāvaikalya Viśeṣoktyudāharaṇam ]

2.327 Example of the Viśeṣokti involving

A Deficiency of Objects

[ Dravyavaikalya Viśeṣoktyudāharaṇam ]

2.328 Example of the Viśeṣokti of Cause

[ Hetu Viśeṣoktyudāharaṇam ]

2.329 The Viśeṣokti of Cause / Conclusion to

Viśeṣokti Alaṃkāra

[ Hetu Viśeṣokti / Viśeṣoktyalaṃkāropasaṃhāraḥ ]

2.330 Definition of Tulyayogitā Alaṃkāra

[ Tulyayogitālaṃkāralakṣaṇam ]

2.331 Example of the Tulyayogitā of Praise

[ Stuti Tulyayogitodāharaṇam ]

2.332 Example of the Tulyayogitā of Censure

[ Nindā Tulyayogitodāharaṇam ]

2.333 Definition of Virodha Alaṃkāra

[ Virodhālaṃkāralakṣaṇam ]

2.334

Example of the Virodha of Actions

[ Kriyā Virodhodāharaṇam ]

2.335

Example of the Virodha of Attributes

[ Guṇa Virodhodāharaṇam ]

2.336

Example of the Virodha of Objects

[ Dravya Virodhodāharaṇam ]

2.337

Example of the Virodha involving

Attributes and Action