Boundaries, Dynamics and Construction of Traditions in South Asia 9780857284303, 0857284304

Citation preview

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Cultural, Historical and Textual Studies of Religions The volumes featured in the Anthem Cultural, Historical and Textual Studies of Religions series are the expression of an international community of scholars committed to the reshaping of the field of textual and historical studies of religions. Titles in this series examine practice, ritual, and other textual religious products, crossing different area studies and time frames. Featuring a vast range of interpretive perspectives, this innovative series aims to enhance the way we look at religious traditions.

Series Editor Federico Squarcini, University of Firenze, Italy

Editorial Board Piero Capelli, University of Venezia, Italy Vincent Eltschinger, ICIHA, Austrian Academy of Sciences, Austria Christoph Emmrich, University of Toronto, Canada James Fitzgerald, Brown University, USA Jonardon Ganeri, University of Sussex, UK Barbara A. Holdrege, University of California, Santa Barbara, USA Sheldon Pollock, Columbia University, USA Karin Preisendanz, University of Vienna, Austria Alessandro Saggioro, Sapienza University of Rome, Italy Cristina Scherrer-Schaub, University of Lausanne and EPHE, France Romila Thapar, Jawaharlal Nehru University, India Ananya Vajpeyi, University of Massachusetts Boston, USA Marco Ventura, University of Siena, Italy Vincenzo Vergiani, University of Cambridge, UK

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA Edited by Federico Squarcini

Anthem Press An imprint of Wimbledon Publishing Company www.anthempress.com This edition first published in UK and USA 2011 by ANTHEM PRESS 75-76 Blackfriars Road, London SE1 8HA, UK or PO Box 9779, London SW19 7ZG, UK and 244 Madison Ave. #116, New York, NY 10016, USA © 2011 Federico Squarcini editorial matter and selection; individual chapters © individual contributors The moral right of the authors has been asserted. Graphics and layout © Mario Caricchio Cover photography © Clelia Pellicano All rights reserved. Without limiting the rights under copyright reserved above, no part of this publication may be reproduced, stored or introduced into a retrieval system, or transmitted, in any form or by any means (electronic, mechanical, photocopying, recording or otherwise), without the prior written permission of both the copyright owner and the above publisher of this book. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data A catalog record for this book has been requested. ISBN-13: 978 0 85728 430 3 (Hbk) ISBN-10: 0 85728 430 4 (Hbk) This title is also available as an eBook.

Contents

Preface

7

I. INTRODUCTION FEDERICO SQUARCINI Tradens, Traditum, Recipiens. Introductory Remarks on the Semiotics, Pragmatics and Politics of Tradition

11

II. DISCOURSE, CONDITIONS AND DYNAMICS OF TRADITION IN SOUTH ASIA SHELDON POLLOCK The Revelation of Tradition: †ruti, smr¢ti, and the Sanskrit Discourse of Power JOHANNES BRONKHORST The Reliability of Tradition TIMOTHY LUBIN The Transmission, Patronage, and Prestige of Brahmanical Piety from the Mauryas to the Guptas

41 62

77

III. HOW TO PRODUCE, CONSTRUCT AND LEGITIMATE A TRADITION ALF HILTEBEITEL Buddhism and the Mahåbhårata. Boundaries and Construction of Tradition LAURIE L. PATTON Trita’s Tumble and Agastya’s Ancestors: On The Narrative Construction of Dharma ROMILA THAPAR Creating Traditions through Narration. The Case of ‡akuntalå JONARDON GANERI A Dynamic Tradition of Truth-telling: Moral Innovation in the Mahåbhårata

107

133 159

175

6

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

FRANCIS X. CLOONEY, S.J. From Person to Person: A Study of Tradition in The Gurupara∫paråsåra of Vedånta De†ika’s ‡rœmat Rahasyatrayasåra CHRISTOPHER MINKOWSKI What Makes a Work ‘Traditional’? On the Success of Nœlakañ™ha’s Mahåbhårata Commentary FRANCESCO SFERRA Constructing the Wheel of Time. Strategies for Establishing a Tradition ELISA FRESCHI, ALESSANDRO GRAHELI Bhå™™amœmå∫så and Nyåya on Veda and Tradition CEZARY GALEWICZ Why should the Flower of Dharma be Invisible? Såyaña’s Vision of the Unity of the Veda

203

225 253 287

325

IV. EXPERIENCING BOUNDARIES WITHIN TRADITION: THE CASE OF THE SANSKRIT GRAMMARIANS MADHAV M. DESHPANDE Ultimate Source of Validation for the Sanskrit Grammatical Tradition: Elite Usage versus Rules of Grammar MARIA PIERA CANDOTTI Loke, vede, †åstre: Grammarians’ Partition of Tradition and Related Linguistic Domains VINCENZO VERGIANI Dealing with Conflicting Views within the Påñinian Tradition: On the Derivation of tyådr¢† etc.

361

389

411

V. VIOLATING TRADITION AND ITS BOUNDARIES FEDERICO SQUARCINI Traditions against Tradition. Criticism, Dissent and the Struggle for the Semiotic Primacy of Veridiction ANTONIO RIGOPOULOS The Nonconformity to Tradition of the Mahånubhåvs FABRIZIA BALDISSERA Tradition of Protest: the Development of Ritual Suicide from Religious Act to Political Statement

437 485

515

VI. THINKING ABOUT TRADITIONS IN SOUTH ASIA TODAY CHRISTOPH EMMRICH When Two Strong Men Stand Face to Face. The Indologist, the Pandit and the Re-Making of the Jaina Scholarly Tradition BRUNO LO TURCO Evaluation or Dialogue? A Brief Reflection on the Understanding of the Indian Tradition of Debate FERNANDO TOLA, CARMEN DRAGONETTI Unity in Diversity: Indian and Western Philosophical Traditions

571

589 607

Preface

This collection of essays is meant to explore the various forms that the theme and the notion of ‘tradition’ took within the South Asian context, during ancient and pre-colonial periods. Designed by the editor to cover a significant selection of the specialized fields of knowledge that shaped classical South Asian cultural history, the aim of this volume is to offer a stimulating anthology of papers on the different and complex processes employed during the ‘invention’, construction, preservation and renewal of a given intellectual tradition. In this regard, the contributors have expertly analysed a large variety of aspects, namely the transmission of traditional canons –both textual and practical–, the dynamisms and the strategies chosen for the renewal of a tradition, its internal and external dialectics, the procedures of its legitimation, the theoretical and pragmatic mechanisms of its survival, the criticisms of traditional knowledge systems, etc. Attention has also been paid to problems related to the primacy exercised by highly specialized traditional experts, to monopolies in the transmission of knowledge, to its means of cultural and political justification, and to the connections between a specific traditional field of knowledge and the surrounding social arena. Hence the following essays, thematically arranged according to a sixfold partion (see, supra, the table of contents), are dense and rich in scholarship and I hope they will notably contribute to the contemporary Indological understanding of the crucial institute of ‘tradition’. Such is the ambitious aim of this volume and I would like to express my deep thanks to those who duly deserve praises and tribute for such intellectual venture. First of all, I wish to thank all the authors of the essays published

8

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

in this collection, since without their positive response to my invitation to the project this volume wouldn’t be of any value. Furthermore, I am indebted with Patrizia Cotoneschi (Firenze University Press) and Pankaj D. Jain (Munshiram Manoharlal) for accepting the many challenges that this co-publication project implied. Finally, I wish to thank Maria Piera Candotti, Piero Capelli, Mario Caricchio, Alessandro Graheli, Bruno Lo Turco, Nancy Nannini Aluigi, Cristina Pecchia, Marina Rustow, Francesco Sferra, Lara Tavarnesi, Vincenzo Vergiani, for their generous help in revising and completing the ponderous amount of work that such a large collection required. In spite of my huge efforts to achieve a unified editorial profile, while respecting the specific editorial choices made by the authors, I feel that I succeeded only in part. Yet, any shortcomings or faults in this work can only be attributed to myself. I really hope that these flaws will not greatly undermine the reader’s expectations. Federico Squarcini OXFORD-DELHI-FIRENZE September 2002 / December 2005

I. Introduction

FEDERICO SQUARCINI Tradens, Traditum, Recipiens. Introductory Remarks on the Semiotics, Pragmatics and Politics of Tradition The theme of ‘tradition’ in the South Asian context, with the variety of its expressions, is the subject of this collection of essays. It is a fundamental topic on which many have reflected and much has been written. However, because of its very centrality, I believe that it can never receive enough attention. In fact, the function performed by the device of ‘tradition’ has been and is still indispensable for the great majority of the South Asian forms and systems of knowledge and meaning, since it is their main foundation of guarantee and validation. Therefore, it is crucial to understand the details and the dynamics of this function in order to effectively grasp the logic of those forms and systems of knowledge. These are, in short, the principal motives that have led me to return once again to the theme of ‘tradition’. Now, in order to counterbalance the terseness of the above statements, it would be appropriate to explain the intentions, goals and reasons that have guided the construction of this volume on ‘tradition’. However, a mere list of programmatic declarations would not do justice to the complexity of the theme and to the wide range of contexts that have been examined and discussed; it would actually generate various kinds of misunderstandings. This does not mean that the organisation and arrangement of the following collection of essays did not follow any ‘guideline’ –namely, a precise programmatic intent. Quite the contrary. Yet I believe that, instead of making a mere list of such guidelines, in these introductory remarks I should dwell upon other aspects, which I consider methodologically more relevant to the study of tradition. This attempt to reconsider a complex concept such as ‘tradition’ –with its many variants– it is therefore prompted not so much by a precise project of definition, as by the desire to radically rethink

12

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

the method and the interpretative criteria we adopt today to imagine and represent the functions of ‘tradition’. We should consider, to start with, to what extent our questions on ‘tradition’ are compelling and appropriate. Such questions, although despotic, are necessary.1 They operate like semiotic grids for the understanding and reduction of complex historical phenomena. Therefore, if we try to reflect upon our method, we should once again ask ourselves questions such as ‘What is a tradition?’, ‘What are its boundaries?’, ‘How can it be defined?’, ‘How does it define itself?’, ‘How does it tell the story of its origins?’, ‘How does it justify and legitimate its existence?’, ‘What needs are met by its coming into existence?’, ‘What are the dynamics of its reproduction?’, ‘How and why does it come to an end?’, ‘What are the means by which it maintains its distinctiveness and vitality over the course of time?’. Moving on to a different level of analysis, we may then ask ourselves to what extent such questions are relevant to the field of South Asian cultural context. By doing this, though, we run the risk of finding ourselves in a double bind. On the one hand, if we start thinking about the logic underlying such questions, we will realise that we are unable to discuss their relevance in an abstract way (in fact, these are all questions that need to be addressed through the scrutiny of specific data and materials in order to be adequately explored). On the other hand, we will find out that these questions deal with problematics that have been identified after the generalisation and universalisation of particular data. In other words, these are questions of a ‘universal’ nature but originated from the analysis of specific material conditions. Before they assumed the general abstract form that allows us to address them again to a particular circumstance, such questions were context-related. Thus, to avoid the temptation to resort to naïve ‘essentialism’,2 acute hermeneutic awareness is strongly needed. It is not so easy, in fact, to see how such way of reasoning can lead to a vicious circle that could have a paralysing effect on research. And in order to overcome this cognitive impasse, it is not enough to investigate to what extent certain questions can be effectively related to the statutes, processes and dynamics of an ‘alien’ tradition. We should rather ask ourselves how plausible they are in regard to our traditions as well as those of others. Instead of focusing only on their trans-cultural adequacy, we should also explore their intra-cultural validity. This implies that we deal with broad methodological issues, concerning both the epistemology and sociology of a tradition’s legitimation processes and of the dynamics of cultural transmission. 1 None can avoid here to still ponder Gadamer’s reflections on the role of questioning (Frage) linked to historical understanding. See Gadamer 1986: 368-384. 2 An epistemic fallacy not so infrequent in contemporary human sciences. See Fuchs 2001.

1. Introduction

13

It is true that in the past some have voiced extreme views in the attempt to free themselves from this impasse. Thus, it has sometimes been claimed that the notion of ‘tradition’ is absolutely alien to the classical South Asian civilisation or, on the contrary, that the interpretative model of ‘tradition’ is the only possible way to explain certain South Asian cultural processes. While these formulations are objectively untenable, they can still serve us as the extremes within which we can carry out new investigations. Of course, this does not mean choosing the far too obvious solution of the ‘middle way’. Instead, we should start from the awareness that all our questions concerning what ‘tradition’ in South Asia is about, are guided by some kind of interest and, therefore, through them we always build for ourselves preconceived models of understanding. After all, while being aware that it is impossible to set aside completely one’s preconceptions, we should nonetheless remember that every cognitive act always implies some kind of investigative strategy. A renewed interpretative effort to undestand the function of ‘tradition’ could start from the willingness to include the analysis of those elements that earlier strategies had underestimated or discarded.3 Otherwise, it could recourse to a polythetic interpretative model which would allow us to confront the data obtained from different fields and cultural contexts through a flexible system of trans-codification. In both cases the results achieved could lead to significant changes in the evaluation of the data itself as well as in the setup of the investigation. The interpretation of the discursive strategies through which a tradition justifies itself is a good opportunity to test this logic. It is well known that the representatives of a given tradition try to justify and legitimate their convictions –which are always exposed to judgment and criticism– through the reflexive strategy of the ‘discourse on tradition’, with its array of principles and related notions. As is widely attested, this practice is usual in the classical traditions of the Mediterranean area. Now, it is reasonable to think that the analytical model through which the developments of this ‘discourse’ are interpreted and classified, takes possession of aspects of the tradition under scrutiny, from its terminology to its idioms. However, an interpretative method which is, if not ‘universal’, at least widely applicable, cannot be elaborated on the basis of the suggestions originating from a single cultural milieu. It is quite evident here that when a model, or method, is modelled solely on the data drawn from a specific context, it has a dangerous tendency to force other ways of thinking reflexively on one’s own tradition into that same mould –even though those ways developed 3 Such was the initial approach of the historiographic project carried out by the collective of ‘Subaltern Studies’ at Delhi University, as can be seen from the first volumes they published. See, exempli gratia, Guha 1982; 1983; 1984.

14

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

independently. And yet, the limitations of such a process are only revealed when one examines the forms that the practice of reflexive discourse on tradition has taken elsewhere. In the classical South Asian intellectual world,4 for instance, this practice was not only well-established, but its peculiarities were such that they could be used to extend our way of understanding and representing the trajectories that the ‘discourse on tradition’ can take. This is precisely the reason why the materials and the reflections produced by the South Asian representatives of this ‘discourse’ must necessarily become an integral part of the dialectical processes that shape and organise the ways in which the notion of ‘tradition’ is conceived today. Therefore, in accordance with this spirit of reconsideration of the method, the criteria and the categories pertaining to the phenomenon of ‘tradition’, I believe that it is necessary to establish a preliminary framework within which contextualise and problematise the variegated picture offered by the essays collected in this volume. It is a good custom to start from the fundamentals –namely, the analysis of the meanings of words and of their semiotic, pragmatic and political implications– and then to use these outcomes in order to face the range of questions and issues that arise when examining the notion of ‘tradition’. 1. De traditione. The semiotics, pragmatics and politics of a notion Let me start from the etymology and semantics of the noun ‘tradition’,5 which derives from the Latin action noun traditio-≠nis, which in its turn means either ‘consignment’ or ‘transmission’ or ‘passage’ or ‘surrender’. The lemma traditio-≠nis is connected with the verb tradere, composed of trans (‘across’, ‘beyond’) and dare (‘to give’), of which the present tense is trado and the past participle traditus. This last term designates something that has been materially ‘handed down’. Hence the Italian term ‘tràdito’, mainly denoting what is preserved and handed down by a succession of manuscripts. This is because the verb tradere primarily designates the physical act of ‘consigning’, ‘entrusting’, ‘transmitting’, ‘transferring’, ‘handing down’, and ‘narrating’. The use of traditio in the terminology of classical Roman law is further evidence of the concreteness of tradere: in fact, it denotes a gesture that is meant 4 The centuries-old debate on the ‘valid means of knowledge’ (pramåñavåda) is precisely the symptom of a conflict both on the possibility of legitimate knowledge and on the exclusive control over the criteria that give power to the means of legitimation of knowledge. See the end of § 3 of my paper in this volume. 5 Regarding the following definitions and technical usage of the terms here mentioned, I have consulted different reference works, such as the Dictionnaire étimologique de la langue latine (Ernout, Meillet 1985), the Lexicon latinitatis Medii Aevi (Blaise 1975), the Lexicon totius latinitatis (Forcellini 1940), the Thesaurus Linguae Latinae (1900-2001).

1. Introduction

15

to ensure the correct and legitimate reception of a possession by an heir.6 Bequeathing property to one’s children is the clearest example of the physical nature of tradere, as is attested in the Roman institution of traditio ficta, namely the act of the legal consignment of a possession –subdivided into traditio brevi manu and traditio longa manu. The practical implications of this notion are also shown by some of its figurative usages, as is the case of tradere in the sense of ‘betray’. This usage was influenced by the pejorative meaning of the notion tradere already present in the text of the Bible,7 and then further reinforced by the association with the ruse by means of which Judas physically ‘handed over’ –by cheating and, therefore, ‘betraying’– Christ to the hostile alien authorities. In one and the same word, ‘tradition’, thus co-exists the meaning of the factuality, concreteness and objectivity of giving and the transitive and dynamic sense of transferring. This is further corroborated by the fact that the kinetic meaning of the verb tradere is complemented by the conservative and static meaning expressed by the Latin word traditio (corresponding to the Greek παραδοσις ), which implies both the concreteness of ‘giving’ (datio) and ‘delivering’ (trado [trans-do]). However, the meaning of traditio that prevailed is that of a particular form of uninterrupted datio, namely the continuous transmission of an original datio, considered so unique and important to be perpetually re-enacted. Such an act of tradere, regarded as a pragmatic action of giving –without a pause, or a break– from hand to hand, follows a kind of positive compulsion to repeat. Therefore, it has been seen as the ultimate guarantee of integrity since it ensures, to those who rely upon such a vehiculum, the immediate contact with the originalis –being a foundational instruction or an initial event. Thus, the word tradere covered many semiotic contexts. Yet, since it has vital importance, the act of tradere demands a more in-depth investigation into its different social and political implications. This is inevitable insofar as any instance of tradere always involves two social agents as well as an object or a content. In fact, any act of transmission requires the presence of somebody who hands over (a tradens, literally ‘someone who gives a certain thing [res]’), of the given object or content (traditum) and of a recipient (recipiens). This division of the act of tradere into its three elementary components provides a preliminary attempt to reveal the factors and inter6 In Roman law, ‘consignment’ (traditio) was acknowledged as the easiest way of transferring the ownership or possession (possessio) of an asset because it consisted precisely in the act of its material consignment. See Schiavone 2003: 307-308; Adriani 1956. Furthermore, Schiavone 2005: 5-38. 7 In the Bible, the act of ‘handing over’ is sometimes associated with leaving someone in difficult conditions or in the hands of hostile people. See, exempli gratia, Deuteronomy, 23.16; 1 Samuel, 23.11-20; 30.15; Job, 16.11-12; Psalms, 30[31].9; 62[63].10-11; 77[78].48, 50, 6162; 117[118].18; Isaiah, 19.4; 51.23; Jeremiah, 18.21; Ezekiel, 35.5; 39.23; Amos, 1.6; 1.9; 6.8.

16

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

ests that constitute, inform and influence a ‘tradition’ –namely, a specific act of transmission. This is precisely the reason why, even though the object of tradere (traditum) may well be an independent entity –a material thing, a verbal statement or a doctrine, that has significance in itself–, it is always necessary to situate it within its systemic context in order to fully understand its meaning. Consequently, the traditum has to be examined simultaneously from the semiotic, pragmatic, mediological and political standpoint.8 Seen from this perspective, any act of tradere is a crucial gesture closely connected to the social –and, hence, political– sphere, from which it cannot actually be separated. 2. From the concrete practice of giving to the abstract institution of datio and the manner of giving again In addition to the systemic and relational dimension of tradere, one must take into account also other aims of this gesture, which concern the act of tradere itself as well as its socio-political consequences. While there are always objective reasons for choosing to tradere a particular object or content to others, there are also concrete benefits connected to such practice. These concern both the individual who performs the act and the one who is affected by it. Apart from the immediate interest they may have in handing over (tradere) a particular object or content (traditum), these two may wish to present themselves, on the one hand, as the sole authors of the gesture itself and, on the other hand, as the only depositaries of the particular object or content that has been handed over. In this way they both try to assert their own exclusive claim on a part of the process of tradere. They know that, once the value and goodness of the datio originalis have been ascertained, the crucial thing is to make use of it, preserve it and give it again ideally intact. This is the concrete task of future transmitters (tradentes), who therefore become the more or less conscious actors of a reiterative institution specialising in the transmission of the given object. While this is surely valid for the transmission of material things, it acquires levels of complexity when it comes to the transmission of statements or doctrines, since in this context the symbolic dimension is more relevant and therefore those who are engaged in the act of transmitting can change their social status –from mere agents into professionals of tradere. Corporate and professional interests –both concrete and symbolic– prompt these transmitters, or mediators, to operate in such a way that their particular way of transmitting and giving again a certain object rapidly imposes itself as the only correct way of tradere. It is with this aim in mind that they come to establish their 8 See, for some illustrations of related analysis and theoretical apparatus, Assmann 1999; Debray 1997: 15-70; Boyer 1990.

1. Introduction

17

own particular way of consigning the traditum. Therefore, the emphasis placed on the correlation and mutual dependence between the original datio –that must be ‘given again’– and the necessary practices for its acceptance, preservation and restitution become an integral part of their policy. The first gesture would be nothing, they say, without the last and vice versa, to the extent that the medium becomes the content and the aim, in a logic according to which practice and content are somehow interchangeable. It is precisely such dynamics that originated the process of transition from the empirical and practical power of a concrete ‘act of transmission’ –a power derived from the sum of the intrinsic and concrete value of what is transmitted and the established symbolic status of the transmitter– to the abstract normative dimension of the modes of tradere, in which the emphasis is placed on the forms of the action rather than on its contents. This causes a crucial change of status, a sort of semantic inversion comparable to the shift from the concrete use of an object to the abstract representation of the notion of property. Here a simple individual gesture of transmission happens to be qualified beyond its merely objective value. Then, from the initial need to provide a stable foundation for one’s own conduct arose a number of devices, by means of which a certain group of tradentes tried to move certain events or doctrines from the status of a particular factum to that of a universal principium or decretum (corresponding to the Greek δο,γμα). This is especially true in the cultural and religious spheres, unlike the legal domain, which focuses on the transmission of clearly defined material goods. In these contexts, the belief according to which a good teaching, if badly transmitted, may be corrupted soon became widespread. This is a common-sense principle, but it gains full force particularly in connection with some kind of strife, for it makes it possible to discredit a teaching simply by questioning the quality of its transmission or the authority of the transmitter –by claiming, for instance, that he was not acknowledged by the community of the tradentes as a qualified transmitter.9 In religious domains, when a deity gives an object or a teaching to some chosen individual, this act has various consequences: not only is the goodness of what is transmitted guaranteed and its survival assured, but it also makes the recipient an elected and privileged individual, drawing special attention to what he has been given. Remembering and narrating the act of the original datio then becomes the means to increase the value assigned to the transmitted 9 It is particularly appropriate here to recall the customs of the Vedic poets, according to which a bard was considered able to genuinely grasp and express meanings an ‘correlations’ (bandhu) both on the basis of his compositional skills and of the consensus given to him by his colleagues (sákhyå), without which he would lose the right path (nahí pravéda sukr¢tásya pánthåm [10.71.6]) and succumb to words that are sterile, literally ‘fruitless and flowerless’ (vÌcam †u†ruvÌ∫ aphalÌm apußpÌm [10.71.5]). See R¢gveda, 10.71.1-11.

18

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

object or content. This is certainly true in many religious contexts, that are full of narrations in which teachings, truths and visions are handed down, or accessed, and that soon rise to the exceptional status of symbolic depositum or testamentum. They become special cases of divine transmissions, which need to be preserved and ‘committed to memory’ because of their unique value.10 In this way one goes from the concrete practice of giving to the symbolic institutionalisation of both the given and the givers. Moreover, now the act and the forms of giving are a ‘unique locus of truth’ since they are decisive in ascertaining the possibility of preserving or corrupting the traditum. As far as religious truths are concerned, every traditum –precisely because of its additional value– is claimed to be in need of ‘institutional guarantees’ that preserve it from the various forms of corruption, deriving both from its use and its misuse. Therefore, it becomes necessary to have recourse to that system of tutelage known as ‘tradition’. Once established, this is thought to ensure the correct preservation, use and retransmission of the original datio.11 If what I have said so far makes sense, it will be useful not only to consider the representations that traditions –as well as those who study them– have produced, but also to pay critical attention to the interests and the positions of the individuals operating within the traditions themselves. Indeed, as they are “interested producers of symbolic systems”,12 they deserve the utmost attention. It is their strategic actions that build up the legitimacy of the institution of tradition. A strategy that leads towards a ‘policy of perception’ which deals with practical and cognitive aspect. Through this policy they aim, first and foremost, at fixing the image of the unity and continuity of their activity13 and, subsequently, at establishing the devices by means of which this image is committed to memory and reproduced. Then, the various processes of selection of what is to be remembered and what is to be forgotten take place via the impersonal activ10 This is especially true in the South Asian classical context, where the precepts of extraordinary people (ancestors, seers, wise men, etc.) are received as special items and are constantly referred to. That which they hand down for posterity to remember is significantly designated with the word smr¢ti, ‘that which is remembered or memorised’. Not that everything remembered obtains the prestigious status of smr¢ti, but anything transmitted from Vedic sources can become an object of memory. See, for few examples of traditional approaches to the defintion of smr¢ti, ‡abara, ‡åbarabhåßya, vol. 2, pp. 72-74 (in Mœmå∫sådar†anam, ed. by K.V. Abhyankar, Anandashrama Press, Pune 1970-1976); Kumårila, Tantravårttika, vol. 2, pp. 94; 104 (in Mœmå∫sådar†anam, idem); Jayantabha™™a, Nyåyama∞jarœ, vol. 1, pp. 372-373 (ed. by K.S. Varadacharya, Oriental Research Institute, Mysore 1969). 11 No tradition, therefore, can renounce the synthesis of praxis and theory, that is, to the development of specific practices of symbolic incrementum, through which the mere gesture of consignment is equal to a noteworthy practice, which also renders its executors noteworthy. Because of this, the executors (now true tradentes) develop an interest in preserving and guarding that very act of consignment. 12 Bourdieu 1991: 4-5 (footnote 9). 13 See Bourdieu 1991: 5; 29. 14 See Assmann 1997: 5-58; Douglas 1990: 109-125.

1. Introduction

19

ity of the established institution of tradition.14 3. Which notion of ‘tradition’ for the intellectual history of South Asia? After suggesting this tripartite model of conceptualising the notion of tradition, I will now consider to what extent it can be adapted to the South Asian data. Since early times, the empirical experience of the appropriation of a new word –either as a lexeme or a sememe– through an act of transmission is reported in positive terms in the cultural history of South Asia. Handing down a lemma is discussed here as an example of the practice of ‘transmission’ (åmnåya) of something that was not possessed or known before. It is an unquestionable practical experience that is significantly referred to in the incipit of the ancient Sanskrit treatise on etymology attributed to Yåska.15 Talking of the lexical heritage that has been ‘transmitted’ (åmnåta), Yåska emphasises both the factuality and the guarantees offered by this institution that has made it possible to acquire formerly unknown lemmas and meanings. His discourse is structured around the same elements of the triad described above: tradens, traditum and recipiens. From this very example, the notion of ‘tradition’ to be used for the South Asian context would appear to be easily conceivable and very close to that described in the preceding paragraph. But, although the subject of ‘tradition’ has been much discussed in Indological studies –variously interpreted and dealt with from different angles–, new researches and changes in the cultural attitudes over the last decades demand us to discuss it again.16 While revisiting ‘old’ interpretative criteria –such as the opposition between ‘great’ and ‘little’ tradition–,17 various studies have chosen to talk about the ‘happening’ of a tradition,18 the ‘negation’ of tradition on the part of some Orientalists,19 the need to critically understand the historiographic role to be assigned to traditions,20 the forms of entropy that can affect a tradition,21 the theme of the 15 See Yåska, Nirukta, 1.1 (samåmnåyaΔ samåmnåtaΔ Ù sa vyåkhyåtavyaΔ Ù tam ima∫ samåmnåya∫ nighañ™ava ity åcakßate Ù nighañ™avaΔ kasmåt Ù nigamå ime bhavanti Ù chandobhyaΔ samåhr¢tya samåhr¢tya samåmnåtåΔ Ù te nigantava eva santo nigamanån nighañ™va ucyanta ity aupamanyavaΔ Ù). 16 As a matter of fact, the studies on tradition in South Asia have recently increased, and not just in number. See the detailed re-articulations of the notion of tradition presented in Kaviraj 2005: 124; 125; 126-127; 128-129; 130. Furthermore, Manring 2005; Saberwal, Varma 2005. Indeed, over the last decade the subject of ‘tradition’ has been widely reconsidered by various specialists. See, exempli gratia, Brockington, Schreiner 1999; Gopal, Champakalakshmi 1997; Champion 1996; Mohanty 1992; Halbfass 1991; Moore 1979. 17 See Agehananda Bharati 1978; Singer 1972; Singer 1959. 18 See D’Sa 1994. 19 See Sugirtharajah 2003: 75-76. 20 See Ludden 2002: 5-9. 21 See Inden 1986. 22 See Rudolph, Rudolph 1967: 269-293.

20

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

‘modernity’ of tradition.22 In spite of these propositions, some of the methodological implications of the recourse to the notion of tradition are still to be clarified. In other words, the gap between the universal category of ‘tradition’23 and the particular aspects it embraces in South Asian contexts may still turn out to be too broad and problematic. Therefore, it is necessary to think again about the theoretical guidelines and the conceptual framework, so to allow scholars to carry out a renewed analysis of the data obtained from the South Asian world. In taking the first steps towards this renewal, it might be worth concentrating our efforts, initially, on the understanding of the ways in which the institute of tradition was conceived and described in South Asian sources, retracing its history from the semiotic, pragmatic and political point of view.24 This will certainly not be sufficient to clarify all the numerous epistemic, social and symbolic implications of such terms as para∫parå (‘succession’, ‘sequence’, ‘tradition’), sa∫pradåya (‘sect’, ‘religious institution’, ‘denomination’), åmnåya (‘transmission’, ‘teaching’), va∫†a (‘lineage’), aitihya (‘historical tradition’) and ågama (‘testimony’, ‘tradition’). In fact, all these are words by means of which certain social agents defined and characterised their own tradition, often with a remarkably thorough self-reflective attitude. However, it is important to comprehend to what extent these words are not simply nouns, but rather semantic indicators that have been used over many centuries as legitimating metaphors, apt to point out the objective location –and, consequently, the actual criteria of its accessibility– of a much more abstract and intangible depositum fidei. I will now examine some of these lexical referents, considering them in the eyes of the tripartite interpretative model presented supra (see § 1, 2). 4. Starting from words: a few etymological and mediological remarks on para∫parå and sa∫pradåya Let’s consider two terms that are crucial for our reflection on the notion of tradition. These terms are para∫parå and sa∫pradåya, both of them widely used in South Asia to denote the functions that are usually assigned to ‘tradition’. An intitial survey of the available lexicographic tools shows that the feminime noun para∫parå (derived from para-m-para, with a reiteration of the stem para,25 literally both ‘distant’, ‘remote’, ‘previous’, 23 In the beginning of this introduction I have discussed to what extent its universality can be said to be partial and particular. 24 In this regard, Sheldon Pollock’s essay published in this volume is an outstanding example of this kind of approach. Furthermore, Pollock 2005. 25 Consider, on the other hand, the word pårá-, already attested in the R¢gveda, which denotes the act of ‘bringing across, from one side to another’. See Turner 1966: 457.

1. Introduction

21

‘ancient’, ‘subsequent’, and ‘different’, ‘other’, ‘opposed’, ‘foreign’, ‘adverse’) means ‘one next to the other’, ‘one after the other’, ‘one another’ or, much more abstractly, ‘that which connects disjointed parts’. The term para∫parå then denotes an ‘uninterrupted line’, a ‘continuous series’ or a ‘regular succession’, and is variously combined with different past participles –among others, para∫paråpråpta– meaning ‘handed down through successive transmissions’. Other derivations of the same stem are used, for instance, to designate the way to obtain ‘a life without interruptions (or perpetual)’ (tathå paramparam åyuΔ sama†nute),26 or to denote opposite extremes and convey the sense of distance, as in the section on the names of rivers in an ancient lexicon.27 Relevant is also the adjectival usage of para∫parå, referred to what is seen as ‘traditional’ or ‘hereditary’ –namely, something from the past, also used as a synonym of kulakramågata (‘obtained through the transmission of the family lineage’). Thus, semantically equivalent to the word paraspara –meaning ‘reciprocal’, ‘mutual’, and itself derived through the reduplication of para– the word para∫parå denotes the dynamical and composite nature of the act of giving and transmitting. The general idea of a ‘system of transmission’ expressed by the word para∫parå is then qualified by specific additional meanings, as in the case of the compound guru†ißyapara∫parå, that is, the ‘transmission [of knowledge] from teacher to disciple’. Furthermore, the noun para∫parå is also used as an indicator of an established version of ‘giving’, displacing the semantic load from the practical act of transmitting to the symbolic institute of transmission. Such meanings are already attested in the classical lexicons.28 The word para∫parå is in fact recorded in the Nåmaliõgånu†åsana of Amarasi∫ha (c. 5th century CE –also known as Amarako†a–), in the section that deals with terms concerning Brahmanic novitiate, forms of asceticism and sacrifice as well as of knowledge and teaching. Here para∫parå is said to denote a ‘kind of traditional instruction’.29 The word is also found in another section of the same work dealing with 26

‡atapatabråhmaña, 4.2.4.7. See Yåska, Nirukta, 2.24 (påråvataghnœ∫ påråvåraghåtinœm Ù påra∫ para∫ bhavati Ù avåramavaram Ù). 28 For further in-depth considerations on the classical uses of the word para∫parå it is worth to consult the Nåmaliõgånu†åsana of Amarasi∫ha with its commentaries and glosses (as the ancient ‘Ko†a’ of Jåtarüpa, the ~œkåsarvasva of Vandyagha™œya Sarvånanda, up to the Amaraviveka of Mahe†vara), as well as its supplements (as the Trikåñ∂a†eßa, the Håråvalœ, the Dvirüpako†a compiled by Purüßottamadeva) and other classical lexicons (such as the Abhidhånacintåmañi, Abhidhånacintåmañipari†iß™a, Liõgånu†åsanako†a of Hemacandra [edited by Pañdit Durgåprasåd, K农nath Påñdurang Parab, Pañdit ‡ivadatta and published in the 1889 within the ‘Collection of ancient lexicon’ –Abhidhånasa∫graha– by the Nirñayasågara Press of Bombay]). 29 See Nåmaliõgånu†åsana, 2.6.835 (påramparyopade†e syåd aitihyam itihåvyayam). 30 See Nåmaliõgånu†åsana, 2.6.862 (paramparåka∫ †amana∫ prokßaña∫ ca vadhårthakam). 27

22

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

the vocabulary of various ritual and sacrificial practices.30 However, the presence of the kinetic sense of the word is not lost, as is shown by a further section of Amarasi∫ha’s treatise in which para∫parå is mentioned among different synonyms for ‘indirect’ means of conveyance (such as palanquin bearers, dray horses, etc.).31 Although the meaning referring to the symbolic value of an ‘established institute’ of transmission has prevailed, there are persistent instances in the ancient sources of the use of the word para∫parå in its pragmatic and dynamic sense, in which its kinetic aspects are emphasised. Consider, for example, the practical and social implications of such compounds as va∫†apara∫parå and kulapara∫parå, in which va∫†a and kula stress the value of a transmission that takes place through the guaranteed medium of familiar or dynastic affiliations; or †rotapara∫parå and karñapara∫parå, in which †rota and karña designate the sense of hearing and its organ, the ear, respectively, through which the transmission occurs. Even more suggestive are the compounds sopånapara∫parå and sopånakapara∫parå, where –in Sanskrit as well as in Påli– sopåna means ‘step’, ‘staircase’, ‘ladder’ or ‘flight of stairs’. The same pragmatical sense is also conveyed by related compounds, such as sopånapåtha, sopånapaddhati, sopånapanti and sopånamårga –all of them of great interest because of the images they conjure up. The will to symbolically institute and the concrete fluidity of movement are, therefore, the two main features expressed by the word para∫parå as used in classical Brahmanic literature.32 It should be noted that the word para∫parå is especially widespread in the ancient Påli Buddhist literature. Here one finds the first attestations of both a specialised use of this word (exentsively used until the much more recent ‘historiographic’ Buddhist work Såsanava∫sa –in which compounds such as therapara∫parå, åcariyapara∫parå, bhikkhupara∫parå, sissapara∫parå and ganthakåraparamparå occur in the context of the distinction between lineages and forms of transmission of the teachings–) and of its being utilized to express the general sense of sequentiality and continuity (as in the term para∫paråbhojana, designating a monk’s meal that See Nåmaliõgånu†åsana, 2.7.1049 (paramparåvåhana∫ yat tad vainœtakam astriyåm). The term para∫parå frequently occurs in Brahmanic sources –less so, however, before the compilation of the Mahåbhårata. See, exempli gratia, Vasiß™hadharmasütra, 6.43 (påra∫paryagato yeßåm vedaΔ saparibå∫hañaΔ); Månavadharma†åstra, 2.18 (tasmin de†e ya åcåraΔ påra∫paryakramågataΔ); Kåtyåyañadharma†åstra, 164 (yåvan yasmin samåcåraΔ påra∫paryakramågataΔ); 891 (ya∫ para∫parayå maulåΔ såmantåΔ svåmina∫ viduΔ); Artha†åstra, 1.12.13 (para∫parå); 1.12.23 (para∫parå); 1.15.16 (para∫parå); 2.34.11 (para∫parå); 8.2.26 (påra∫paryakrameña uktam); 14.1.32 (matsyapara∫parå); Mahåbhårata, 3.195.34 (para∫parå); 6.26.2 (para∫paråpråptam); 6.115.27 (påra∫paryeña); 11.23.21 (påra∫paryeña); 12.101.26 (påra∫paryågate); 12.164.12 (påra∫parya∫); 12.326.113 (påra∫paryågata∫); 12.336.2 (påra∫paryågatå); 13.73.13 (påra∫paryågata∫); Råmåyaña, 4.55.5 (para∫parå); 5.14.30 (para∫parå); Bhartr¢hari, Våkyapadœya, 1.159 (anekatœrthabhedåyås trayyå cåcaΔ para∫ param Ù); Œ†varakr¢ßña, Så∫khyakårikå, 71. 31

32

1. Introduction

23

follows a pre-established order or sequence). Moreover, in these texts it is also possible to come across the apologetic and self-reflective recourse to the device of para∫parå.33 Considering that the word already occurs in the earliest parts of the Påli canon,34 the contexts in which it is used are far from being homogeneous. Nonetheless, it seems that from the beginning the word para∫parå is used to denote the specificity and the advantages of having recourse to the device of the transmission through a teaching lineage, as it appears from some occurrences in the Vinayapi™aka.35 Of utmost interest is the co-existence –in formulaic phrases– of the two words paramparå and sampadåya in Påli texts.36 A distinction is also made among different forms and dimensions of the paramparå, for instance, between ‘elementary’ (dhåtuparamparå) and ‘large’ (mahåparamparå) paramparå.37 Furthermore, various types and means of transmission are explicitly distinguished, as is exemplified, for instance, in the Divyåvadåna.38 33 A powerful apologetic dispositif adopted up until modern narrations of the origins of Buddhist tradition. See Såsanava∫sadœpa, 1063-1074; 1532-1542; 1635-1659 (these relevant sections are significantly entitled mahåmahindattherådyåcariyaparamparådi kathå dœpo, åcariya paramparådi kathå dœpo and åcariya paramparå kathå dœpo respectively). Further, Law 1986. 34 See, exempli gratia, Majjhimanikåya, 2, in PTS [Påli Texts Society] p. 520 (paramparåya); 3, in PTS p. 74 (upeti gabbha∞ca para∞ca loka∫ sa∫såramåpajja paramparåya); 3, in PTS p. 78; 3, in PTS p. 169 (paramparåya); 3, in PTS p. 170; 3, in PTS p. 200; Aõguttaranikåya, 3, in PTS p. 189 (paramparåya); 4, in PTS p. 191 (paramparåya); 4, in PTS p. 191 ([…] må anussavena må paramparåya, må itikiråya […]); Cullavagga, 5.6, in PTS p. 22 (åvåsa[‘home’, ‘place of residence’]paramparå); 5.37, in PTS p. 110 (ve¬u[‘bamboo’]paramparå); 6.4, in PTS p. 25 (åvåsaparamparå). 35 See, exempli gratia, Parivårapå¬i, in PTS, pp. 3; 6; 18; 39; 49; 54; 56; 81-82; 88; 128; 130 (åcariyaparamparå); 139 (åcariyaparamparå); 144; Pacittiya, 6.4.3, in PTS pp. 75-78 (section entitled paramparahojana sikkhåpada∫). 36 See, exempli gratia, Majjhimanikåya, 2, in PTS p. 520 (so anussavena itihitiha paramparåya pi™akasampadåya dhamma∫ deseti Ù […] so anussavena itihitiha paramparåya pi™akasampadåya dhamma∫ deseti); 3, in PTS p. 169 (atha kho kåpa™iko måñavo bhagavanta∫ etad avoca: ‘yam ida∫ bho gotama, bråhmañåna∫ poråñåna∫ mantapada∫ itihitiha paramparåya pi™akasampadåya, tattha ca bråhmañå eka∫sena ni™™ha∫ gacchanti’. ‘idam eva sacca∫ moghama∞∞a’nti, idha bhava∫ gotamo kimåhå’ti); Aõguttaranikåya, 3, in PTS p. 189 ([…] må paramparåya, må itikiråya, må pi™akasampadånena […]); 4, in PTS p. 191 ([…] må anussavena, må paramparåya, må itikiråya, må pi™akasampadånena […]); Suttantapi™aka, 1, in PTS p. 360 (sakkhidhamma’nti na itihitiha∫ na itikiråya na paramparåya na pi™akasampadåya); 1, in PTS p. 400 (sakkhidhamma’nti na itihœtiha∫ na itikiråya na paramparåya na pi™akasampadåya); 1, in PTS p. 482 (na itihœtiha∫ na itikiråya na paramparåya na pi™akasampadåya). 37 See, exempli gratia, Mahåva∫sa, 3.40; 35.40 (yugaparamparå tesa∫ purato påvisœ pura∫); 74.245; 91.82 (katolikaparamparå [note that the term katolika is only found in the Mahåva∫sa, where it occurs 16 times, also in connection with the word ågama]); Dhåtuva∫sa, 3 (an entire pariceddha entitled dhåtuparamparå kathå). 38 See Divyåvadåna, 190.11 (te †ravañaparamparayå cånveßamåñås tasya gr¢hapateΔ sakå†amupasa∫kråntåΔ Ù); 289.16 (eßa ca vr¢ttåntas tena bråhmañena karñaparamparayå †rutaΔ Ù); 478.10 (eva∫ karñaparamparayå sa †abdastayor duß™åmåtyayoΔ karña∫ gataΔ Ù); 499.11 (båßpasaliladhåråparamparodbhavoparudhyamånakañ™hœ anilabalåkulita galitasajalajaladapa™alå valœmalinake†apå†å satvaratvaram abhigamya maitrakanyakasya bodhisattvasya pådayoΔ parißvajyaivam åha –må må∫ putraka parityajaya yåsœti Ù).

24

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Thus, the device of para∫parå is here widely acknowledged and invoked for its legitimising force.39 This is particularly relevant for early Buddhist sources, which –being strongly dependent on the prototypical and charismatic figure of the Buddha– have a much more compelling need than Brahmanic sources to create a system of validation of their status. As a consequence, Buddhist texts sometimes present a self-reflective ‘discourse’ that is not found, instead, in the surviving earlier Brahmanic literature. The other key-word is sa∫pradåya. According to a sütra of Påñini’s grammar (Aß™ådhyåyœ, 3.1.141), this is derived from the root då-, belonging to the third gaña (present tense, dadåti) and meaning ‘to give’, through the affixation of the agentive suffix ña (=a) preceded by the increment yuk (=y).40 From this etymological explanation, then, sa∫pradåya denotes ‘that which transmits’ rather than ‘that which is transmitted’. The emphasis is placed on the symbolic value of the institute established by the ‘act of tradere’ rather than on the contents of such acts. Significantly, the term dåya, which is the basic semantic component of sa∫pradåya, is first attested in contexts that deal with the transferring or partitioning of goods and, especially, of legacies.41 This sense of material transference, intrinsic in the word sa∫pradåya, will be maintained, even though it has been accompanied by figurative and symbolic meanings since its earliest attestations.42 However, the concreteness of this ‘institute of transmission’ (sa∫pradåya) is already emphasised in some ancient sources. An interesting case is Yåska’s mention of the fact that the oral transmission of the hymns (mantra) and, later on, of their editing and collection into ‘distinct treatises’ (bilmagrahañåya imaΔ granthaΔ), served the purpose of countering the inevitable gradual disappearance of individuals –among the new generations of bråhmañas– 39 In fact in these sources it is also associated with men of power. See, exempli gratia, Sœha¬avatthü, 27.35 (vacanaparamparåya asokaråja sutvå). 40 I wish to thank Vincenzo Vergiani for his help with these points of grammar. 41 See, exempli gratia, Taittirœyasa∫hitå, 3.1.9.4-5. In the later juridical literature, the topic of inheritance, with the related guarantees, is extensively dealt with. In the Månavadharma†åstra, in particular, the presentation of the criteria for the ‘division of the inheritance’ (vibhåga, dåyabhåga) represents one of the eighteen ‘grounds for litigation’ (vyavahårapåda). See Månavadharma†åstra, 9.103-220. Other sections of the same work dealing with inheritance are Månavadharma†åstra, 1.115; 8.7; 8.27; 9.47; 9.77-79; 10.115; 11.185. The question of how to properly transmit the inheritance has been largely debated within the Sanskrit juridical tradition until the XII century CE. See, about one of the last innovative treatise on this matter, Rocher 2002. 42 See, exempli gratia, ‡atapathabråhmaña, 1.5.2.7 (sa yadå†råvayati Ù yaj∞am evaitad anumantrayata å naΔ †r¢ñüpa na åvartasvety atha yatpratyå†råvayati yaj∞a evaitad upåvartate ’stu tatheti tenopåvr¢ttena retaså bhütenartvijaΔ sampradåya∫ caranti yajamånena paro ’kß∫ yathå pürñapåtreña sampradåya∫ careyur eva manenartvijaΔ sampradåya∫ caranti tad våcaivaitat sampradåya∫ caranti vågghi yaj∞o vågu hi retas tad etenaivaitat sampradåya∫ caranti). The same, with small variations, is in the Kåñva recension. See Kåñva†atapathabråhmaña, 2.4.4.3.

1. Introduction

25

who were capable of having a ‘direct intuition of the norm’ (såkßåtkr¢tadharman).43 Evidently, these individuals were perceived as lacking the qualities of the ancient ‘seers’ (r¢ßi). However, throughout the various contexts in which it is used, the primary connotation of the word sa∫pradåya is that of the ‘repeated performance’ of an act of giving and taking. The later Nåmaliõgånu†åsana of Amarasi∫ha lists sa∫pradåya among other action nouns designating a received doctrine or teaching.44 And indeed, among the numerous later definitions of sa∫pradåya, there is one given by Uddyotakara, according to whom ‘tradition’ is the handing down of knowledge from ‘teacher’ (upadhyåya) to ‘pupil’ (†ißya).45 With regard to the tripartite interpretative model presented in preceding pages, it is necessary to draw attention to the connections between some aspects of the above semiotics and etymological derivation of the word sa∫pradåya and other related terms. Note, for example, the verb sa∫pradå- (present tense sa∫pradåti, past participle sa∫pradatta), the basic meaning of which is ‘to offer completely’, ‘to give up’, but also ‘to transmit’, ‘to impart’ and ‘to teach’. From sa∫pradå- the noun sa∫pradåna derives, which means either ‘gift’, ‘present’, or ‘transmission’, ‘teaching’. In both cases the full sense of ‘giving’ (as a datio) is dominant. This suggests a dynamic that is really very close to what I said above about the triad tradens, traditum and recipiens. In fact, in the grammatical literature –and especially in the short, but extremely interesting, sa∫pradånådhikåra (Våkyapadœya, 3.7.129-135) of Bhartr¢hari– the various complex implications of the use of the dative case are explained through a reflection on the role played by the recipient, namely, the kåraka called sa∫pradåna. Here the recipient is the final consignee and, indeed, the real motive for an action addressed to him, which he has to consent to if the action is to be properly performed and acquire its full significance. Thus, the focus is on the individual (sa∫pradåna) to whom something is entirely and effectively given.46 The concreteness that is a connotation of the word sa∫pradåna

43 See Yåska, Nirukta, 1.20 (såkßåtkr¢tadharmåña r¢ßayo babhüvuΔ Ù te ’varebhyo ’såkßåtkr¢tadharmabhya upade†ena mantrån sa∫pråduΔ Ù upade†åya glåyanto ’vare bilmagrahañåya ima∫ grantha∫ samåmnåsißuΔ Ù veda∫ ca vedåõgåni ca Ù bilma∫ bhilma∫ bhåsanam iti cå Ù). Furthermore, Wezler 2001. 44 See Nåmaliõgånu†åsana, 3.2.241-242 (vardhana∫ chedane ’tha dve ånandanasabhåjane Ù åpracchannamathåmnåyaΔ sa∫pradåyaΔ kßaye kßiyå ÙÙ). 45 See Uddyotakara, Nyåyavårttika ad Nyåyasütra, 1.1.1 (sampradåyo nåma †ißyopadhyåyasa∫bandhasyåvicchedena †åstrapråptiΔ). 46 According to the grammatical commentators, these are the implications to be drawn from Påñini’s sütra (Aß™ådhyåyœ, 1.4.32 [kármañå yám abhipraíti sá sampradÌnam]), for which the technical name sa∫pradåna designates the beneficiary or recipient that the agent intends to reach through his action (of giving).

26

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

is further shown by those ritual activities –known as ‘transmission’ (sa∫pradåna) or even ‘acquisition’ (sa∫pratti)–47 by means of which a dying father hands over his doctrinal, ethic, symbolic and practical legacy to his son. All these rituals indicate the need to ensure both an ideal and concrete continuation of the preservation of a given order. In short, insofar as they serve as guarantees, all these systems of transmission must be able to dynamically preserve what they were once given. A dual concern which requires a dual effort. From what can be seen through the semantic history of the word sa∫pradåya, in fact, only what the tradition (sa∫pradåya) actually hands down is to be understood as the true testamentum and should be preferred to any supposed original meaning. This seems to be the sense of Jaimini’s words,48 which also recurs in the remarks of some later commentators while they strive to clarify the sense of some central notions –such as that of the permanence or eternity (nityatva) of the Veda.49 According to Uddyotakara, this ‘eternity’ should even be taken in a figurative sense, namely as an emphatic reference to the length and certitude of the uninterrupted process of concrete transmission.50 Besides its employment in the ancient Buddhist literature,51 the term sa∫pradåya circulated widely in Brahmanic circles, as it became the most common word designating a specific religious tradition or denomination.52 What I have presented here, in regards of the terms para∫para and sa∫pradåya, should be considered on the one hand as the proof of an explicit awareness of the value and importance of the institute of 47 In the late Vedic period, the ceremony of sa∫pradåna was a form of justification and symbolic procedure that had the purpose of confirming the heir’s right to receive what the institute of transmission destined to him. See, for some of the earliest descriptions of this practice, Br¢hadårañyakopanißad, 1.5.17-20; Kaußœtakopanißad, 2.15. Furthermore, Olivelle 1993: 123-126. 48 See Jaimini, Mœmå∫såsütra, 1.2.8 (tulya∫ ca såmpradåyikam). 49 See Våtsyåyana, Nyåyabhåßya ad Nyåyasütra, 2.1.68. 50 See Uddyotakara, Nyåyavårttika ad Nyåyasütra, 2.1.68. 51 See, for some examples of the use of the Påli sampadåya, Majjhimanikåya, 1.3.9 (mahåsaropamasutta, 6), in PTS p. 192 (so tåya sœlayampadåya attamano hoti paripuññasaõkappo Ù so tåya sœlasampadåya attånukka∫seti para∫ vambheti […] aham asmi sœlavå kalyåñadhammo, ime pana∞∞e bhikkhü dussœlå påpadhammåti Ù so tåya sœlasampadåya majjati pamajjati pamåda∫ åpajjati Ù pamatto samåno dukkha∫ viharati Ù); 2, in PTS p. 520 (so anussavena itihitiha paramparåya pi™akasampadåya dhamma∫ deseti Ù […] so anussavena itihitiha paramparåya pi™akasampadåya dhamma∫ deseti); 3, in PTS p. 169 (atha kho kåpa™iko måñavo bhagavanta∫ etad avoca: ‘yam ida∫ bho gotama, bråhmañåna∫ poråñåna∫ mantapada∫ itihitiha paramparåya pi™akasampadåya, tattha ca bråhmañå eka∫sena ni™™ha∫ gacchanti’. ‘idam eva sacca∫ moghama∞∞a’nti, idha bhava∫ gotamo kimåhå’ti). 52 As for what I said above on para∫parå, the word sa∫pradåya is often used by Brahmanic sources, although it is not so frequent as in and after the Mahåbhårata. See, exempli gratia, Mahåbhårata, 1.82.4 (sa∫pradåya); 2.5.5 (sa∫pradåya); 3.13.18 (sa∫pradåya); 5.27.27 (sa∫pradåya); 10.11.10 (sa∫pradåya); 13.70.50 (sa∫pradåya); Råmåyaña, 2.29.18 (sa∫pradåya).

1. Introduction

27

tradition; on the other, as complementing the remarks made above on the possibility for an agent (tradens) to obtain a certain status –precisely, that of a recognised tradens– on the basis of the symbolic significance given to a practical act –the agent’s handing over of a traditum– that would normally be forgotten right away. This is to show that my initial interpretative proposal is substantiated by a variety of descriptive reports preserved in classical South Asian sources. In summary, the principle according to which a practical act accompanied by a semiotic intention will lead to a semantic increment –that, in its turn, will have practical repercussions– seems to be at work also for the South Asian context. 5. The dual role of tradition and the need for novelty From what has been said so far it appears obvious that a given tradition cannot remain still when faced with historical changes and the corrosive effects of time –as actually stated in the Bhagavadgœtå.53 In fact, this is the true reason why its process of continuation cannot be limited to mere static repetition, on pain of death. Hence every tradition has devised complex intellectual practices and strategies, thanks to which, while the elements of the originally established corpus are innovated and changed –though seeking not to formally alter the fundamental unitary picture– an attempt is made to preserve the image of integrity and unalterability.54 In such contexts, ‘new’ necessarily means unreliable while –since Vedic times–55 ‘ancient’ is presented as an undisputable sign of trustworthiness. Within the South Asian intellectual history, in fact, there are many examples of authors who, by addressing a change under the guise of ‘novelty’, sought to preserve a particular conception of the 53 See Bhågavadgœtå, 4.7-8 (yadå yadå hi dharmasya glånir bhavati bhårata Ù abhyutthånam adharmasya tadåtmåna∫ sr¢jåmy aham ÙÙ paritråñåya sådhünå∫ vinå†åya ca dußkr¢tåm Ù dharmasa∫sthåpanårthåya sa∫bhavåmi yuge yuge ÙÙ). 54 In this regard, the picture presented in Bhågavadgœtå, 4.2 is emblematic (eva∫ paramparåpråptam ima∫ råjarßayo viduΔ Ù sa kåleneha mahatå yogo naß™aΔ parantapa ÙÙ). 55 In those times, giving the shape of ‘archaic’ and ‘ancient’ was indeed a commonly applied strategy, as persuasively stated by Witzel: “In all these cases one can notice that one means to bring about continuity in spite of the great changes carried out under the Kurus, was the artificial archaization of certain parts of the new ‡rauta ritual, the use of artificial, archaic forms in the poetic and learned language of the poets, priests and ‘theologians’ of the Mantra and YV Sa∫hitå periods, and of text formation and their collection. The new ritual and its language appeared to be more elaborate and impressive but at the same time, had to give the appearance of having come down from a hallowed past”. Witzel 1995: 15. 56 Most pertinent here is the case of kalivarjya, as it is dealt with in Sanskrit juridical texts. See, exempli gratia, Br¢haspatidharma†åstra, 1.23.4 (åtatåyidvijågyåñå∫ dharmayuddhena hi∫sanam Ù imån dharmån kaliyuge varjyån åhur manœßiñaΔ ÙÙ). An extended discussion on the notion of kalivarjya can be read in Kane 1993: 926-968. Furthermore, Lingat 1999: 189-195; Kane 1997; Sannino Pellegrini 1997; Dumont 1991: 339; Smith 1987: 38-42; Heesterman 1984: 151-152; Doniger 1980: 37-43; Bhattacharya 1943.

28

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

world,56 in a way that is not so different from what characterised the recourse to the theme of the ‘classical’ in the Mediterranean area (through which, instead of slavishly repeating the past, often new visions of the past –and therefore of the present and the future– were promoted).57 Far from a descriptive distinction, the opposition between ‘new’ and ‘old’ is a judgemental dicotomy. In other words, whilst praising its coherence with the past, the tradition is thus obliged to make known also the twists it has introduced. Such is the quandary that can be found in classical South Asian sources. Each time they explain their origins, the way in which they say they were consigned to the world and the reasons why they are now well known, they display the inevitable need to highlight the specificity of their contribution. But this is the dilemma of every historically positioned traditional context: being unable to renounce the merit it would gain by publicising its particular contribution (shaped through the effort of keeping alive the tradition it has received, both by preserving and protecting or by innovating and renewing it), it must explicitly state the importance of its own action. The result is that the element of ‘novelty’ is stigmatised in ‘orthodox’ texts like the Månavadharma†åstra, which raises a severely admonishing finger against what is labeled as new and recent.58 However, ‘novelty’ cannot be avoided for two reasons: on the one hand, by not updating itself, a tradition risks to lose its persuasive force, on the other, those who, while working within a tradition, do not sufficiently emphasise the specificity of its role, risk to diminish its importance. The theme of novelty and originality becomes an essential part of traditional discourse, though the fact remains that novelty was never to be presented as an ex novo given, but if anything as a renewal, restoration, reformulation of the original.59 Considering the above, it is not advantageous to keep thinking of tradition as if it were a question of […] a corpus of norms fixed once and for all in time. Not only does it experience initial and sometimes disastrous controversies as far as the 57

On this interesting topic, see Settis 2004; Gadamer 1986: 290-295. See Månavadharma†åstra, 12.95-96: “95. The scriptures that are outside the Veda, as well as every kind of fallacious doctrine –all these bear no fruit after death, for tradition takes them to be founded on Darkness. 96. All those different from the Veda that spring up and then flounder –they are false and bear no fruit, because they belong to recent times [tånyarvåkkålikatayå]”. This is a principle of caution typical of every tradition, which is not very different in its condemnation of what is ‘new’, from what is to be found, for example, in one of Paul’s epistles (whose attribution is still debated). See 1Timothy, 6.20 (O Timothee, depositum custodi, devitans profanas vocum novitates, et oppositiones falsi nominis scientiae, quam quidam promittentes, circa fidem exciderunt). 59 See Kaviraj 2005: 129-131. 58

1. Introduction

29

unity of the community of believers is concerned, but it constantly undergoes a process of revision, and sometimes substantial creative adaptations and changes. Tradition is, from this standpoint, a social construct; a collective undertaking that has a beginning and in many respects is never finished.60

This becomes obvious by looking at the semiotic and political history of certain idiomatic expressions –where by ‘semiotic’ I mean the history of ‘shared meaning’ and by ‘political’ the history of the ‘shared legitimacy’ of precisely those meanings. If read critically, these histories show how often there is no true novelty in the new, except for the fact of the novelty of the new reading of the old –which is the element that constitutes the true innovation. Seen from this angle, the traditional approach is a ‘new way of reading and rephrasing the old’ that, while it exercises its interpretative practices (which become the ingredients of an actual ‘culture’) is also concerned with defining, organising and guaranteeing the legitimacy of its ways of looking at the past and making use of it. It is a dual crucial strategy that has to be understood simultaneously, because –as Sheldon Pollock recently stated– “[c]onceptually, it is obviously as important to understand what enables a tradition to radically transform itself as it is to understand what enables a tradition to secure continuity […]”.61 There are many examples of such strategy, starting from the süktas of the R¢gveda –from which one gathers the need to justify the production of new compositions–,62 passing through the innovation introduced by certain early grammarians and philosophers of the language,63 from that produced by the authors of mœmå∫så64 and of kåvya,65 up until the most recent cases of novelty drawn from the production of the schools of the ‘new (navya) logic (nyåya)’,66 of the ‘new grammar’ (navyavyåkåraña),67 and of the new medicine.68 These are all evident instances of renewal in certain ‘traditional’ intellectual fields, which, under the pressure of changing surroundings, had to devise new arguments and new narrative strategies. A fact already noted by Pandurang Vaman Kane: […] social ideas and practices undergo substantial changes even in the 60

Pace 1996: 9-10 (my translation). Pollock 2005: 7. 62 See exempli gratia, Elizarenkova 1995: 23-25; Fortson 1998; Galewicz 2000; Galewicz 1995. 63 See, exempli gratia, Bronkhorst 2005; Houben 2002. 64 See, exempli gratia, McCrea 2002. 65 See, exempli gratia, Bronner 2004: 54-75; Bronner 2002; Ingalls 1976; Ingalls 1965. 66 See, exempli gratia, Ganeri 2005: 44-47; 51-52; Preisendanz 2005: 66-72; 76-86; Kaviraj 2005: 131. 67 See, exempli gratia, Pollock 2001: 12. 68 See, exempli gratia, Wujastyk 2005. 61

30

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

most static societies. Many of the practices, that had the authority of the Veda (which was supposed to be self-existent and eternal) and of such ancient smrtis as those of Åp., Manu and Yåj., had either come to be given up or had become obnoxious to popular sentiment. This fiction of great men meeting together and laying down conventions for the Kali age was the method that was hit upon to admit changes in religious practices and ideas of morality. The Kalivarjya texts are also a complete answer to those who hold fast to the notion that dharma (particularly åcåradharma) is immutable and unchangeable (aparivartanœya).69

Every historical season (yuga), therefore, must have its dharma,70 because previous dharmas are not pertinent anymore.71 Similar considerations regarding the role of newness in South Asian traditions have recently led to interesting interpretations of these dynamics and to the distinction of “three kinds of newness”.72 It is now clear that in this context the relationship with one’s own tradition as a reference is always complex, and within this the contents it has handed down may be unquestioningly accepted and simply repeated, or restored, questioned, modified and even denied or abandoned. Essentially, if it is true that we can think of a tradition “[…] as the way society formulates and deals with the basic problems of human existence”, then it follows that […] since the fundamental problem of life and death is truly insoluble, it has to be attacked, formulated, and dealt with each time anew under a different aspect. Tradition therefore is and has to be bound up with the ever-shifting present. Hence the irritating flexibility and fluidity of tradition.73

So, every tradition, by definition, is an established space where constant negotiation takes place and in which avant-gardes and rearguards do battle for the last word. A fight that informs the very processes of transmission and that is carried out both at individual and collective level. This implies that tradens, traditum and recipiens are all parts and protagonists of the same agonistic dimension. 6. Novelty, negotiation and the politics of transmission It is well known that the relationship between tradition and reason 69

Kane 1993: 967. In this regard, the idea of a ‘specific dharma for any specific time’ (yugadharma) is an extraordinary dispositif, largely used by ‘orthodox’ author. A useful collection of sources on yugadharma is presented in Keshnakar 2000: 779-964. 71 It is important to read such formula from the sociological and political point of view. For example, the idea that a period of social and political crisis had produced the Brahmanic rhetoric of the ‘dark age’ (kaliyuga) is particularly relevant here. See Sharma 2002. Furthermore, on the notion of kaliyuga, Lingat 1999: 189-195; Olivelle 1993: 234-237; Stietencron 1986; Upadhyay 1979: 25; 28-31; 115-116. 72 Kaviraj 2005: 124. 73 Heesterman 1985: 10. 70

1. Introduction

31

–and hence between memory and innovation– implies a complex combination of continuity and innovation. Because of such complexity, the dialectical negotiation that constitutes any act of transmission has been the object of long-standing scholarly attention.74 But, since the transmission process that stands at the core of every tradition is dialectically structured –both around the act of preservation and renewal–, we need to add some considerations of a purely socio-political nature to what scholars like Gadamer said about the interdependence of tradition and reason.75 In actual fact, seen from a mediological perspective, the efficacy of a certain act of cultural production and transmission is always closely linked to its capacity to generate socio-political distinction and differentiation. As Régis Debray underlines: Voilà qui suffit à faire peu ou prou de toute entreprise de transmission une opération polémique, requérant une compétence stratégique (à s’allier, filtrer, exclure, hiérarchiser, coopter, démarquer, etc.), et qui peut s’appréhender comme une lutte pour la survie au sein d’un système de forces rivales tendant soit à s’éliminer entre elles par disqualification soit à s’annexer l’une l’autre par phagocytose. […] La transmission appartient à la sphère politique, comme toutes les fonctions servant à transmuer un tas indifférencié en un tout organisé. Elle immunise un organisme collectif contre le désordre et l’agression.76

In fact, the aim of the conceptual effort made by the representatives of a tradition is to succeed in transferring, without any break, pronouncements and practical convictions from one point to another of the history of a particular community. Precisely for this reason a tradition […] cannot be only flexible and situational, for its essential mission is 74 Hans-Georg Gadamer’s pages on the ‘rehabilitation of authority and tradition’ (Die Rehabilitierung von Autorität un Tradition) are still the best available for challenging reading. See Gadamer 1986: 281-295. In this pages Gadamer presents a particular modus of the coexistence of reason (Vernunft) and tradition (Tradition), a modus which deserve serious consideration: “Truly, tradition is always a moment of freedom and of history itself. Even the most authentic and solid of traditions does not develop naturally by virtue of the strong persistence of what happened once, but it needs to be accepted, adopted and cultivated. It is essentially conservation [Bewahrung], that same conservation that is at work alongside and within every historical change. But conservation is an act of reason, certainly an act characterized by the fact that it is not conspicuous”. Gadamer 1986: 286 (my translation). Thus Gadamer help us to understand in which sense the institute of tradition is an essential dispositif in order to take advantage of previously acquired knowledge and experiences –although it still remains necessary to evaluate the socio-political modus operandi of such processes of accumulation and transmission. Gadamer’s understanding has been scrutinized considering the South Asian context and materials. See Halbfass 1990: 164-17o. 75 Because of the limited attention paid to the role played by the political sphere, it is not possible to embrace Gadamer’s hermeneutic perspective in its entirety. If anything, it is necessary to make some critical corrections to it, like those Jnger Habermas has pointed out. See Habermas 1979. 76 Debray 1997: 21.

32

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

still to deal in a structured way with the insoluble life-death problem in all its situational manifestations. It must, therefore, also offer a plan or order independent of and above the actual situation. It is this transcendent order that provides man with the fixed orientation for legitimizing his actions in the middle of the situational flux. In other words, tradition has to be both immanent in the actual situation so as to keep up with shifting reality and transcendent so as to fulfil its orientating and legitimizing function. Thus, we can understand the paradoxical but traditionally common idea that transcendent law is all the time there, suspended as it were in the midair, and that it can be ‘found’ by agonistic procedures, verbal or otherwise.77

Due to such pervasive agonism, every tradition contains a whole series of simultaneous cultural processes, which demand that the highest intellectual resources are constantly tapped in order to create a synthesis between the attempt to preserve power, the constant effort to gain legitimation from history, the awareness of the constant need to adapt to reality, the task involving the complex operation of transmitting those contents understood as being ‘right’ (in fact, every tradition represents itself as ‘orthodox’ and draws one of the most important justifying factors from this self-representation) and the defence of its own truths from the ever-present threat of otherness. This allows one to draw a more articulated picture of the notion of ‘tradition’, and to place it within a diversified and dynamic social history of intellectual practice. 7. Rethinking our understanding of the intellectual history of South Asia from our usage of the notion of ‘tradition’ If, as I hope has happened at the end of this excursus, thinking of tradition –and of its representations– along semiotic, political and mediological lines can become a valuable opportunity to reconsider our interpretation of many South Asian sources and materials, likewise this is also a significant opportunity to reflect critically about the Indological tradition and on the use Indologists have made of the notion of ‘tradition’. This dual interpretative register is an element of reflection that cannot be avoided, although this has to be understood keeping in mind that “[t]o adopt the viewpoint of reflexivity is not to renounce objectivity, but to question the privilege of the knowing subject”.78 In this regard, Jan Heesterman’s appropriate words of warning, although written more than twenty years ago, are definitely still valid: […] the lingering notion of India’s persistent traditionality owes much to the observer’s feeling of having lost his own traditional moo77 78

Heesterman 1985: 11. Bourdieu 1996: 207.

1. Introduction

33

rings, which makes him cast around for the certainty of tradition. India thus becomes a screen on which to project our nostalgia for a world we have lost, even when we know that the good old times were not all that good.79

A very similar problem was raised by Romila Thapar, according to whom [t]raditions which we today believe have long pedigrees may on an historical analysis be found to be an invention of yesterday. In other words, what we regard as tradition may well turn out to be our contemporary requirements fashioned by the way we wish to interpret the past. Interpretation of the past have also come to be treated as knowledge and are handed down as tradition.80

It is certainly not superfluous here to insist that the use of an instrumental notion of ‘tradition’ has doubtless contributed to the production of static, artificial and reified images of South Asian intellectual life. Instead of conceiving the mutual relationship between tradition and novelty –but also between continuity and rupture, repetition and innovation, memory and reason– as an inherent and constitutive feature of the very modus operandi of every tradition, these reified images have proposed such a relationship, at best, as an accessory or sporadic element. On the contrary, it is an integral part of that fundamental dynamic –often forgotten in traditionalist discourse for obvious reasons of legitimation– to which a tradition can never renounce, unless it wishes to decree its own demise. In reproducing such stereotyped version of the institute of tradition, academics and traditionalists shared much responsibility. In this respect, the cautionary advice of scholars like Jan Heesterman and Romila Thapar did not serve the purpose. Although they have invited us to recognise the vices, bondages and fetters that could unite representatives of a certain tradition and its external interpreters, still we failed to take full advantage of such recommendations. Along this line, I think Romila Thapar’s considerations on the ways of representing and conceiving this bondage are so relevant that they deserve to be quoted again in their entirety: The continuity of culture is generally related to traditions which, in turn, are made up of cultural forms. Tradition is defined as the handing down of knowledge or the passing on of a doctrine or a technique. Cultural history implies looking analytically both at what goes into the making of a tradition as well as that which is interpreted by historians as tradition. We often assume that a form is handed down in an unchanging fashion and that what comes to us is its pristine form. However, the sheer act of handing on a tradition introduces change,

79 80

Heesterman 1985: 1. Thapar 1994: 8.

34

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

and not every tradition is meticulously bounded by mnemonic or other devices to prevent interpolations or change. A tradition, therefore, has to be seen in its various faces. Even the concept paramparå, which at one level appears to be frozen knowledge, reveals on investigation variations and changes.81

Therefore, as I said in the beginning, the attempt to rethink the status of tradition in South Asia cannot exclude the broader task of requalifying the categories used, nor it can overlook the self-reflexive effort that concerns any scholar who wishes to combine the ars historica with the ars hermeneutica.

81

Thapar 1994: 8.

1. Introduction

35

BIBLIOGRAPHY Adriani, M. (1956), ‘Traditio’ romana e culto della ‘fides’, in «Studi Romani», 4, n. 4, pp. 381-389. Agehananda Bharati, S. (1978), Great Tradition and Little Traditions: Indological Investigations in Cultural Anthropology, Chowkhamba Sanskrit Studies XCVI, Varanasi. Assmann, A. (1999), Zeit und Tradition. Kulturelle Strategien der Dauer, Böhlau Verlag, Köln-Wien. Assmann, J. (1997), La memoria culturale. Scrittura, ricordo, e identità politica nelle grandi civiltà antiche, Einaudi, Torino. Bhattacharya, B. (1943), The ‘Kalivarjyas’ or Prohibitions in the ‘Kali’ Age, University of Calcutta, Calcutta. Blaise, A. (1975), Lexicon latinitatis Medii Aevi praesertim ad res ecclesiasticas investigandas pertinens, Typographi Brepols, Turnholti. Bourdieu, P. (1991), Genesis and structure of the religious field, in «Comparative Social Research», 13, pp. 1-44. ________ (1996), The Rules of Art: Genesis and Structure of the Literary Field, Stanford University Press, Stanford. Boyer, P. (1990), Tradition as Truth and Communication: A Cognitive Description of Traditional Discourse, Cambridge University Press, London. Brockington, M., Schreiner, P. (ed.) (1999), Composing a Tradition: Concepts, Techniques and Relationship, Croatian Academy of Sciences and Arts, Zagreb. Bronkhorst, J. (2005), Bha™™oji Dœkßita on Spho™a, in «Journal of Indian Philosophy», 33, n. 1, pp. 3-41. Bronner, Y. (2002), What is New and what is Navya: Sanskrit Poetics on the Eve of Colonialism, in «Journal of Indian Philosophy», 30, n. 5, pp. 441-462. ________ (2004), Back to the Future. Appayya Dœkßita’s Kuvalayånanda and the Rewriting of Sanskrit Poetics, in «Wiener Zeitschrift fr die Kunde Sdasiens», XLVIII, pp. 47-79. Champion, C. (ed.) (1996), Traditions orales dans le monde indien, Ed. EHESS, Paris. D’Sa, F.X. (1994), The happening of tradition. The Mœmå∫så’s Vedapramåñam, in «HermE», pp. 75-96. Debray, R. (1997), Transmettre, Éditions Odile Jacob, Paris. Doniger, W. (1980), The Origins of Evil in Hindu Mythology, University of California Press, Berkeley. Douglas, M. (1990), Come pensano le istituzioni, Il Mulino, Bologna. Dumont, L. (1991), Homo Hierarchicus, Adelphi, Milano. Elizarenkova, T.J. (1995), Language and Style of the Vedic ¥ßis, State University of New York Press, Albany. Ernout, A., Meillet, A. (1985), Dictionnaire étimologique de la langue latine: histoire de mots, Klincksieck, Paris (4 ed.). Forcellini, E. (1940), Lexicon totius latinitatis, Typis Seminarii, Patavii

36

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

(rist.), 6 voll. Fortson, B.W. (1998), Some New Observations on an Old Topic: náva∫ vácaΔ in the Rigveda, in J. Jasanoff, H.C. Melchert, L. Oliver (ed.), Mír Curad. Studies in honor of Calvert Watkins, Innsbrucker Beiträge zur Sprachwissenschaft, Innsbruck 1998, pp. 127-138. Fuchs, S. (2001), Against Essentialism. A Theory of Culture and Society, Harvard University Press, Cambridge. Gadamer, H.-G. (1986), Wahrheit und Methode. Grundzge einer philosophischen Hermeneutik, (Gesammelte Werke, Band 1), J.C.B. Mohr, Tbingen. Galewicz, C. (1995), Tradition versus creativity in the ¥gveda, in C. Galewicz, M. Czerniak-Drozdzowicz, I. Milewska, L. Sudyka (ed.), International conference on Sanskrit and related studies, September 23-26, 1993 (proceedings), The Enigma Press, Cracow, pp. 121-126. ________ (2000), To Be Novel and Canonical: a Few Remarks on Textual Strategies in the Rigveda, in «Cracow Indological Studies», 2, pp. 77-85. Ganeri, J. (2005), Traditions of Truth: Changing Beliefs and the Nature of Inquiry, in «Journal of Indian Philosophy», 33, n. 1, pp. 43-54. Gopal, S., Champakalakshmi, R. (ed.) (1997), Tradition, Dissent and Ideology: Essays in Honour of Romila Thapar, Oxford University Press, Delhi. Guha, R. (ed.) (1982), Subaltern Studies. Writings on South Asian History and Society, No. 1, Oxford University Press, Delhi. ________ (ed.) (1983), Subaltern Studies. Writings on South Asian History and Society, No. 2, Oxford University Press, Delhi. ________ (ed.) (1984), Subaltern Studies. Writings on South Asian History and Society, No. 3, Oxford University Press, Delhi. Habermas, J. (1979), La pretesa di universalità dell’ermeneutica, in AAVV, Ermeneutica e critica dell’ideologia, Queriniana, Brescia, 1979, pp. 131-167. ________ (1979), Su ‘Verità e metodo’ di Gadamer, in AAVV, Ermeneutica e critica dell’ideologia, Queriniana, Brescia, pp. 60-70. Halbfass, W. (1990), India and Europe. An Essay in Philosophical Understanding, Motilal Banarsidass, Delhi. ________ (1991), Tradition and Reflection. Exploration in Indian Thought, State University of New York, Albany. Heesterman, J.C. (1984), ‘Orthodox’ and ‘Heterodox’ Law: Some remarks on Customary Law and the State, in S.N. Eisenstadt, R. Kahane, D. Shulman (ed.), Orthodoxy, Heterodoxy and Dissent in India, Walter de Gruyter, Leiden, pp. 149-167. ________ (1985), The Inner Conflict of Tradition: Essays in Indian Ritual, Kingship, and Society, University of Chicago Press, Chicago. Houben, J.E.M. (2002), The Brahmin Intellectual: History, Ritual and ‘Time out of Time’, in «Journal of Indian Philosophy», 30, n. 5,

1. Introduction

37

pp. 463-479. Inden, R. (1986), Tradition Against Itself, in «American Ethnologist», 13, n. 4, pp. 762-775. Ingalls, D.H.H. (1965), The K®ßñacarita of Samudragupta: A Modern Forgery, in «Journal of the American Oriental Society», 85, n. 1, pp. 60-65. ________ (1976), Kålidåsa and the Attitudes of the Golden Age, in «Journal of the American Oriental Society», 96, n. 1, pp. 15-26. Kane, P.V. (1993), History of Dharma†åstra, Bhandarkar Oriental Research Institute, Poona (repr.), vol. III. ________ (1997), Kalivarjya. Actions Forbidden in the Kali Age, in S.G. Moghe (ed.), Professor Kane’s Contribution to Dharma†åstra Literature, D.K. Printworld, Delhi, pp. 213-232. Kaviraj, S. (2005), The Sudden Death of Sanskrit Knowledge, in «Journal of Indian Philosophy», 33, n. 1, pp. 119-142. Keshnakar, G. (ed.) (2000), Dharmako†a, Pråj∞a På™ha†å¬å Mañ∂a¬a, Wai, vol. 5 (varñå†ramadharmakåñ∂a), pt. 2. Law, B.C. (1986), The History of Buddha’s Religion (Såsanava∫sa), Sri Satguru, Delhi (repr.). Lingat, R. (1999), The Classical Law of India, Oxford University Press, Delhi (repr.). Ludden, D. (2002), India and South Asia. A Short History, Oneworld, Oxford. Manring, R.J. (2005), Reconstructing Tradition. Advaita Åcårya and Gau∂œya Vaißñavism at the Cusp of the Twentieth Century, Columbia University Press, New York 2005. McCrea, L. (2002), Novelty of Form and Novelty of Substance in Seventeenth Century Mœmå∫så, in «Journal of Indian Philosophy», 30, n. 5, pp. 481-494. Mohanty, J.N. (1992), Reason and Tradition in Indian Thought. An Essay on the Nature of Indian Philosophical Thinking, Oxford University Press, New York. Moore, R.J. (ed.) (1979), Tradition and Politics in South Asia, Vikas, Delhi. Olivelle, P. (1993), The ņrama System. The History and Hermeneutics of a Religious Institution, Oxford University Press, New York. Pace, E. (1996), Presentazione, in R. Guolo (ed.), Il paradosso della tradizione. Religioni e modernità, Guerini e Associati, Milano. Pollock, S. (2001), New Intellectuals in Seventeenth-Century India, in «The Indian Economic and Social History Review», 38, n. 1, pp. 3-31. ________ (2005), The Ends of Man at the End of Premodernity, Royal Netherlands Academy of Arts and Sciences, Amsterdam. Preisendanz, K. (2005), The Production of Philosophical Literature in South Asia During The Pre-Colonial Period (15th to 18th Centuries): The Case of the Nyåyasütra Commentarial Tradition, in «Journal of Indian Philosophy», 33, n. 1, pp. 55-94. Rocher, L. (ed.) (2002), Jœmütavåhana’s Dåyabhåga. The Hindu Law of

38

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Inheritance in Bengal, Oxford University Press, New York. Rudolph, L.I., Rudolph, S.H. (1967), The Modernity of Tradition. Political Development in India, University of Chicago Press, Chicago. Saberwal, S., Varma, S. (ed.) (2005), Traditions in Motion, Oxford University Press, Delhi. Sannino Pellegrini, A. (1997), Surå e bevande inebrianti: analisi di una interdizione, in AAVV, Bandhu. Scritti in onore di Carlo della Casa, Edizioni dell’Orso, Torino, vol. I, pp. 431-448. Schiavone, A. (ed.) (2003), Diritto privato romano. Un profilo storico, Einaudi, Torino. ________ (2005), Ius. L’invenzione del diritto in Occidente, Einaudi, Torino. Settis, S. (2004), Futuro del ‘classico’, Einaudi, Torino. Sharma, R.S. (2002), The Kali Age: A Period of Social Crisis, in D.N. Jha (ed.), The Feudal Order: State, Society and Ideology in Early Medieval India, Manohar, Delhi, pp. 61-77. Singer, M. (ed.) (1959), Traditional India: Structure and Change, The American Folklore Society, Philadelphia. ________ (1972), When a Great Tradition Modernizes. An Anthropological Approach to Indian Civilization, Pall Mall Press, London. Smith, F.M. (1987), The Vedic Sacrifice in Transition. A Translation and Study of the Trikåñ∂amañ∂ana of Bhåskara Mi†ra, Bhandarkar Oriental Research Institute, Poona. Stietencron, H. von (1986), Kalkulierter Religionsverfall: Das Kaliyuga in Indien, in H. Zinser (ed.), Der Untergang von Religionen, Reimer, Berlin, pp. 135-150. Sugirtharajah, S. (2003), Imagining Hinduism. A postcolonial perspective, Routledge, London. Thapar, R. (1994), Cultural Transaction and Early India, Oxford University Press, Delhi. Thesaurus Linguae Latinae, editus auctoritate et consilio academicarum quinque germanicarum, B.G. Teubner, Stuttgart-Leipzig 19002001, 10 voll. Turner, R.L. (1966), A Comparative Dictionary of the Indo-Aryan Languages, Oxford University Press, London, vol. 1. Upadhyay, G.P. (1979), Bråhmañas in Ancient India. A Study in the Role of the Bråhmaña Class from c. 200 BC to c. AD 500, Munshiram Manoharlal, Delhi. Wezler, A. (2001), Some remarks on Nirukta 1.20 såkßåtkr¢tadharmåña r¢ßayo, in A. Michaels (ed.), The Pandit. Traditional Scholarship in India, Manohar, Delhi, pp. 215-248. Witzel, M. (1995), Early Sanskritization. Origins and Development of the Kuru State, in «Electronic Journal of Vedic Studies», 1-4, pp. 1-26. Wujastyk, D. (2005), Change and Creativity in Early Modern Indian Medical Thought, in «Journal of Indian Philosophy», 33, n. 1, pp. 95-118.

II. Discourse, Conditions and Dynamics of Tradition in South Asia

SHELDON POLLOCK The Revelation of Tradition: †ruti, smr¢ti, and the Sanskrit Discourse of Power * ‘It is the Veda –the sacred knowledge of sacrifice, ascetic acts, and holy rites– that ultimately secures the welfare of the twice-born’ (Yåj∞avalkyasmr¢ti, 1.40)

In some recent papers that consider the nature and role of †åstra viewed as a genre, the character of the rules it articulates, and the denial of history its worldview entails, I have tried to clarify some of the ways in which social-cultural practices come to be legitimated (or de-legitimated), and how ‘authoritative resources’ –that is, knowledge generating and sustaining social and cultural power– are allocated and concentrated (Pollock 1985; 1989b; 1989c).1 I would like to continue this analysis here by examining one set of higher-order categories of Sanskrit discourse, an apparently narrow topic that I nonetheless believe may contribute directly to this process of legitimation. This set of categories is in itself, moreover, basic to the formation and self-understanding of Sanskrit culture, and yet it has often been misunderstood in Western (and westernized) Indology. I want to examine here the significance of the terms †ruti and smr¢ti, and their relationship with one another, as explained in the * This is a corrected version of an essay originally published in S. Lienhard, I. Piovano 1997 (the essay was submitted to the editors in 1988 and reflects the scholarship up to that date). Had I rewritten it for the present collection I would have modified some of the interpretive framework –I have long since sought to nuance the logic of ‘legitimation’ in premodernity, for example see Pollock 1996; 2006, chap. 13– but the review of the historical semantics would have remained largely unchanged. 1 I thank Eli Franco for calling my attention to several errors in an earlier draft of this paper. For the others I have since introduced he is in no way responsible.

42

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Sanskrit tradition. At the same time, I am interested in the implications these issues have for Indian intellectual history. At stake in this discussion is not just an epistemological, let alone philological, clarification, of †ruti and smr¢ti, although I do direct attention to both matters since in my view they have never been convincingly explicated. What is really important here, I think, is that we are encountering a basic component in the construction of the legitimacy of a vast range of Sanskrit elite representations (Pollock 1989a). I would lay emphasis on the matter of ‘construction’. While the fact of ideological power in Sanskrit culture may by now be something of a banality to Western Indology, little or no systematic analysis has been directed toward this ideology in its character as discourse, toward the history of its formation, the techniques it employs, the categories it develops and presses into service. It is as a modest contribution to this analysis that I want to try to clarify the indigenous conception of the relationship of †ruti and smr¢ti –a complex question I can only outline in the brief space available to me here– for it is here we confront, I suggest, one elementary form of ideological power in Sanskrit culture. 1. ‘Tradition’ and ‘Revelation’? A review of some standard scholarly and popular reference works published over the past fifty years or so reveals a virtually unanimous consensus on the definitions of the terms †ruti and smr¢ti, which has been unquestioned despite the fact that these definitions are confusing and problematic, if not plain wrong. Here are some samples. Winternitz: “[...] the †ruti, the ‘Revelation’, i.e. that literature to which, in the course of time, divine origin has been ascribed [...] in contrast to [...] smr¢ti, ‘memory’, i.e. tradition, [which] posses[es] no divine authority” (Winternitz 1927: 161); Renou and Filliozat: “[...] ce que les Indiens désignent par smr¢ti ‘(tradition fondée sur la) mémoire’, l’opposant à †ruti ‘revélation’ […]” the latter in turn being defined by them as “[...] une ‘audition’ (†ruti), c’est-à-dire une révélation: [les textes védiques] passent pour émaner de Brahman, avoir été ‘expirés’ par le dieu sous forme de ‘paroles’, tandis que leurs auteurs humains, les r¢ßi ou ‘sages inspirés’, se sont bornés à les recevoir par une ‘vision directe’” (Renou, Filliozat 1947: 381, 270); Basham: “[...] Smr¢ti (‘remembered’), as distinct from the earlier Vedic literature, which is †ruti (‘heard’), which was believed to have been directly revealed to its authors, and therefore of greater sanctity than the later texts” (Basham 1954: 112-113); Radhakrishnan and Moore: “[...] smr¢tis, that is, traditional texts, as contrasted with the literature of the Vedic period, which is known as †ruti, revealed scriptures or ‘authoritative texts’” (Radhakrishnan, Moore 1973: xix); Raghavan and Dandekar: “[...] semi-canonical scriptures called Smriti, ‘(human) Tradition’ –as opposed to the Vedas, which are

II. Discourse, Conditions and Dynamics of Tradition in South Asia

43

Shruti ‘(divine) Revelation’” (Raghavan, Dandekar 1958: 217);2 Singer: “The cultural tradition which in India is thought of as being transmitted from what has been revealed to the seers (†ruti) and through that which is remembered (smr¢ti) by pandits and storytellers [...]” (Singer 1959: 151); Gonda: “Die ‡rautasütras beanspruchen, auf der †ruti –d.h. auf ‘dem Hören’ der ewigen Wahrheit durch inspirierte Weise in der Vorzeit– zu beruhen, die Gr¢hya- und Dharma-sütras beruhen auf der Smr¢ti –‘der Erinnerung’, d.h. dem Herkommen [...]” (Gonda 1960: 107); or again, “[...] in contradistinction to the [Vedic texts] which are regarded as ‘heard’ or ‘revealed’, and from the beginning orally transmitted (the eternal and infallible †ruti [...]), [the ßa∂aõga] were –like the epics, puråñas and especially the dharma texts– looked upon as remembered and handed down by human intermediaries (smr¢ti [i.e., ‘transmitted by human memory’])” (Gonda 1975: 34, 46); Botto: “La tradizione indigena riconosce quali fonti del dharma la ‘rivelazione’ (†ruti), ossia l’insieme dei testi vedici in quanto rivelati direttamente dalla divinità; la ‘tradizione’ (smr¢ti), cioe i testi considerati opera umana e tramandati per via umana, mnemonica” (Botto 1969: 294); van Buitenen: “†ruti (literally ‘learning by hearing’) is the primary revelation, which stands revealed at the beginning of creation. This revelation was ‘seen’ by the primeval seers [...] Smr¢ti’ (literally ‘recollection’) is the collective term for all other sacred literature [...] which is considered to be secondary to †ruti [...]” (van Buitenen 1974: 932-933);3 von Simson: “Nicht mehr zur Offenbarung (‡ruti), sonder zur autorativen Überlieferung (Smr¢ti) gerechnet wird die vedische Sütra-Literatur [...]” (von Simon 1979: 54 [in Bechert et al.]); Deutsch: “Ancient Indian religious literature was formally classified as either a ‘revelation’ (†ruti –that which has been sacramentally ‘heard’, the eternally existent Veda), or a ‘tradition’ (smr¢ti –that which has been ‘remembered’ from ancient times)” (Deutsch 1987: 125 [in Eliade: vol. 2]). Let us critically juxtapose ‘revelation’ and ‘tradition’ as formulations of these two keywords of Sanskrit culture, and consider for a moment some of the problems they cause. What, for example, warrants the easy equation ‘memory, i.e. tradition’? These two categories are no more co-extensive in India than in the West. In what sense does smr¢ti literature qua memory disqualify it for ‘divine authority’, or diminish its ‘sanctity’, as something standing in fundamental contrast to †ruti? Is it true that smr¢ti is so called because it is handed down in the ‘memory’ of ‘human intermediaries’? If it is, how is smr¢ti thereby distinguished from the Veda? For Vedic texts were not committed to 2 In de Bary et al. 1958. In the second edition this becomes: “the body of semicanonical scriptures called smr¢™i (remembered) tradition –as opposed to the Vedas, which are †ruti (revealed) tradition” (de Bary et al. 1988: 214). 3 As cited in Coburn 1984: 439.

44

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

writing until the medieval period (beginning probably no earlier than the fifth century), and even then were never thought to retain their sacral efficacy if they were not learned according to the oral tradition (Tantravårttika, vol. 2, p. 123, l. 20). And anyway, why should memory, which is operative in both cases, serve to differentiate the degree of authority in the two genres? Furthermore, isn’t the Veda as much a part of ‘tradition’ –more than a part, the actual foundation of Brahmanical tradition– and as much the object of traditional transmission –in fact, its very paradigm– as any other text of ancient India? Conversely, if the Veda is ‘heard’, and only ‘heard’, so is smr¢ti and every other form of discourse in pre-literate Sanskrit culture. What is ‘heard’, consequently, is also ‘remembered’, and what is ‘remembered’ is also ‘heard’. If, however, †ruti is taken with Renou to mean ‘audition as revelation’, how are we to make sense of the tenacious belief, however variously it has been elaborated, that the Veda was ‘seen’ by the r¢ßis, a belief which Renou adduces in the very same passage? I do not want to make too much out of this distinction between ‘hearing’ and ‘seeing’, let alone deny that ‘seeing’ may have a figurative signification. But the Indian tradition, that part which accepts revelation to begin with (contrast below), is rather clear: ‘the r¢ßis ‘saw’ dharma’ (såkßåtkr¢tadharmåña r¢ßayo babhüvuΔ [i.e. mantradraß™åraΔ, Durga]) (Nirukta, 1.6.20 [p. 52]); “[...] ‘r¢ßi’ is derived from the verbal root dr¢†; ‘the sage saw the stomans’, as Aupamanyava glosses it” (Nirukta, 2.3.11 [p. 83]); “the r¢ßis had visions of the mantras” (r¢ßœñå∫ mantradr¢ß™ayo bhavanti) (Nirukta, 7.1.3 [p. 348]). See also Påñini in Aß™ådhyåyœ, 4.2.7: “såmans ‘seen’ by particular sages are named after them, e.g., the såman ‘seen’ by the sage Kali is called the ‘kåleya’ såman” (though Kaiya™a ad loc. rationalizes ‘i.e. ‘seeing’ means ‘knowing’ the particular ritual application of the såman’). The r¢ßis are not normally said to have ‘heard’ mantras.4 Similarly, according to Gonda, van Buitenen, and many others, †ruti was something ‘heard’ in a mythic past, and this is the fact that certifies its authority. But for one thing, the idea of a unique revelation in the past contradicts a dominant –and certainly ancient– representation of the ‘beginninglessness’ of the Veda in the Pürvamœmå∫så. In this system the Vedic texts could not have been ‘heard originally’ by the r¢ßis, since there is thought never to have been an origin. This is likely to have been the position of Jaimini himself (ukta∫ tu †abdapürvatvam [Pürvamœmå∫såsütra, 1.1.29, especially as understood by Nyåyasudhå, p. 269]). This important sütra deserves special study in its own right. For most commentators, it refers to the beginninglessness of Vedic recitation, e.g., Adhvaramœmå∫såkutühalavr¢tti: “‘The ritual recitation of the Veda, which is 4 Note that ‘fifth Veda’ texts such as the Mahåbhårata are also ‘seen’. See Mahåbhårata, 18.5.33.

II. Discourse, Conditions and Dynamics of Tradition in South Asia

45

here the matter at issue, has always depended on a previous recitation, precisely because it is ritual recitation, just like present-day recitation’ –this syllogism demonstrates that there can never have been a first reciter of the Veda, whereby the Veda might have been said to have had an author”.5 For another thing, what are we to suppose to be the origin of what smr¢ti remembers? Where, that is, does Gonda’s Herkommen come from, and when, and how? If the term †ruti is supposed to connote that certain texts are ‘directly revealed’, does smr¢ti connote that other texts are somehow ‘indirectly revealed’, or not ‘revealed’ at all? And what does either position entail practically speaking, that is, with regard to the Enstehungsgeschichte of these texts as indigenously conceived? It would be easy to multiply these questions, but this should suffice to show that a number of conceptual difficulties, to which long acquaintance and acquiescence may have inured us, beset the definitions of these basic terms current in Western scholarship. And these are, I should stress, the definitions that Indology believes to be internal to the Sanskrit tradition, and intended by it, and not external and analytically constructed. Contrasts of the latter sort between the two genres of texts are possible and available (for instance, we might characterize †ruti as ‘indirect’, ‘symbolic’, as opposed to the ‘direct’, ‘rationalist’ smr¢ti etc. [Renou 1960: 27]), and with these contrasts I do not take issue, for they are not pertinent to the problem I am raising here. They tell us nothing about Indian self-understanding, about indigenous representations of culture and society, and it is there that the origins, nature, and function of ideological discourse are located. Can it be that this self-understanding, as reflected in these culturally central categories, is as confused as Indology’s representation makes it appear to be? A matter of equal importance is the implication for us of the opposition of †ruti to smr¢ti explicitly drawn in every one of the explanations quoted above and suggested by the invariable translations divine ‘revelation’ and human ‘tradition’. Difficult as both of these two Western terms may be to conceptualize satisfactorily, when paired they constitute for us nearly a bipolarity: two separate realms of knowledge/practice, distinct in origin, in the manner in which they derive their legitimacy, and in degree of 5 Adhvaramœmå∫såkutühalavr¢tti, vol. 1, pp. 16-17 (which expands on ‡lokavårttika, våkyådhikaraña v. 366, and largely reproduces ‡åstradœpikå, p. 162). See, further, ‡lokavårttika, codanåsütra vv. 143 ff. (with Kå†ikå ad loc.); sa∫bandhåkßepaparihåra, vv. 41 ff. Such is also the view, though from a slightly different perspective, of Uttaramœmå∫så. See ‡aõkara on Brahmasütra, 1.3.30 (on sa∫sårasya anåditvam). As for Jaimini’s Pürvamœmå∫såsütra itself, contrary to what the commentators claim, however, or indeed the sütra itself –if in fact this is what it means– Jaimini has not yet said any such thing; he has only established the beginningless of language, not that of the Veda. If this were not the case, why would Kumårila have to establish this in the våkyådhikaraña? The difficulty is evident in Parame†vara’s Jaiminœyasütrårthasa∫graha ad loc., and especially in the Jaiminœyanyåyamålå of Mådhava, whose analysis is quite at odds with standard Mœmå∫så theory, as represented by Kumårila above. Prabhåkara does not comment on the sütra directly.

46

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

authority. Dichotomized as ‘revelation’ and ‘tradition’, †ruti and smr¢ti almost come to represent for us the Indian equivalent of divine (or natural) law on the one hand, and common (or even positive) law on the other. I am not convinced that these terms mean, or ever meant, what their common Western translations tell us they mean. When we explore the domain in which they are likely to have originated and certainly retained a special centrality, we find something rather different, and instructive. It is in Mœmå∫så that †ruti and smr¢ti seem first to have been clearly conceptualized in their relationship to one another, an inaugural conceptualization that suggests to me the terms may have been coined in Mœmå∫så, though I do not have enough evidence to argue that here. And what the terms signify in Mœmå∫så, first of all, reveals a coherent if increasingly complex ethno-representation, and, second, helps us to recover the potential in this representation for expressing and reproducing an element of the ideology of Sanskrit culture. My argument is that the bifurcation required by such dichotomous concepts as ‘revelation and tradition’ is precisely what the categories †ruti and smr¢ti reject; that this rejection is established in the very terminology that constitutes these categories; and that, formulated first weakly and narrowly in ‘early’ Mœmå∫så,6 it was subsequently more strongly and broadly argued out by Kumårila, whereupon it was generalized throughout Sanskrit culture as one trope of the Sanskrit discourse of power. 2. The Origin of ‘Tradition’ The elaboration of the concept dharma beyond its primary field of reference –Vedic ritualism, or ‘sacrifice, recitation, and gifts’, as for instance the Chåndogyopanißad (2.23.1) defines the three components of dharma– was a development of crucial, if as yet apparently unappreciated, significance in Sanskrit social-cultural history. Far from accepting the paradox as Jan Heesterman has formulated it –that the Vedas have really nothing to do with dharma, and so have ‘ultimate authority over a world to which they are in no way related’ (Heesterman 1978)– we should rather, in keeping with actual historical sequence, reverse the paradox and so cancel it: the ‘world’ outside of ritualism had originally little to do with dharma.7 I won’t address this question any further here except to note that when dharma ultimately spilled over the conceptual confines of ‘sacrificial ritualism’ and came to encompass virtually the entire range of activities of Sanskrit society –and, by reason of its very exclusion, of non6 For ‘new, later’ (and thus ‘old, early’) Mœmå∫så, see e.g., Någe†a on Mahåbhåßya, 4.3.101. 7 I consider the expansion of the realm of dharma at somewhat greater length in Pollock 1990.

II. Discourse, Conditions and Dynamics of Tradition in South Asia

47

Sanskrit society– some explanation of the relationship of the two domains had as a consequence to be provided by the custodians of vaidika dharma. This is the context within which the analysis of the terms †ruti and smr¢ti take on importance for intellectual history. The signification of these categories is dependent upon the relationship in which they were held to stand to one another. In fact, this relationship determines the choice of technical terms used to refer to these entities, and their use would appear to postdate the conceptualization of their relationship. The first discussion of the topic in Indian philosophical history illustrates these points. While many of the arguments developed in this discussion have long been familiar to Indology, their significance for the meaning of smr¢ti has clearly not. The Pürvamœmå∫såsütra,8 after having shown dharma to be that which is known by means of the Veda alone, and established the Veda’s transcendent and inerrant nature (1.1), and then argued that these traits apply to the entire Vedic corpus, narrative and hymnic portions no less than commandments (1.2), has to address the problem posed by the fact that texts other than the Veda –and practices other than what is explicitly enjoined by the Veda– had come to count as dharma in daily life (this for example is the explanation of the sa∫gati in the ‡åstradœpikå, 1.3.1). What legitimacy can be claimed by such texts and practices that are not part of the Vedic canon, that is, not explicitly enunciated in Vedic sources? “The basis of dharma is sacred word, and therefore what is not sacred word has no relevance [vis-à-vis dharma]” (dharmasya †abdamülatvåd a†abdam anapekßa∫ [v.l. -kßyam] syåt [Pürvamœmå∫såsütra 1.3.1]), is the prima facie view necessitated by the postulates previously established. As ‡abara explains it, texts and practices relating to dharma that have no foundation in the Veda can have no valid foundation at all. Nor can some memory of the Veda provide the necessary foundation, because such a memory is not possible: “Something [phenomenal] that has not been experienced, or [something transcendent] that is not transmitted in Vedic texts cannot be the object of memory. These [other texts and practices in question], which relate to the transcendent and yet are not in the Veda, cannot truly be remembered since they can never have been previously cognized”. The smr¢tis cannot be based on sheer ‘memory’ (smaraña) because memory presupposes experience, and the only previous experience of something that counts as dharma is, as proven in Pürvamœmå∫såsütra (1.1.2), the Veda. Furthermore, it is not just the continuity of cultural memory that authenticates it; our ‘memory’ of the Vedas themselves is not validated merely by its unbroken tradition, but by the fact that the Vedas are actually perceptible to us. It is this actual perception of Vedic texts 8 Within this essay, Jaimini Pürvamœmå∫såsütra is cited by number, while ‡åbarabhåßya and Tantravårttika are usually cited by volume, page, and line number.

48

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

–their existing during recitation– that constitutes the ‘prior cognitive experience’ necessary to substantiate the memory of them (pratyakßeñopalabdhatvåd granthasya nånupapanna∫ pürvavij∞ånam); no such prior cognition is available to underpin ‘non-Vedic’ texts and practices. And no tradition founded on such sort of ignorance can become true simply by being beginningless (the jåtyandhapara∫parånyåya, or the principle of the ‘tradition of those blind from birth’, that is, whose knowledge is founded on ignorance and does not cease to be ignorance for being held to be immemorially transmitted) (‡åbarabhåßya, vol. 2, pp. 72-74). The siddhånta is offered in the next sütra: “On the contrary: By reason of the fact that the agents involved are the same, ‘inference’ could be a ‘source of valid knowledge” (api vå kartr¢såmånyåt pramåñam anumåna∫ syåt [Pürvamœmå∫såsütra, 1.3.2]). Insofar as the same people who perform the acts of dharma required by the Veda also perform acts of dharma ‘not based on sacred word’, we must assume that the authority for these other actions is conferred, not by directly perceptible Vedic texts, but by texts inferentially proven to exist. As ‡abara adds, it is not unreasonable to hold that the knowledge of these texts is remembered, while the texts themselves (i.e. their actual wording) have been lost.9 In brief, the authority for practices not validated by Vedic texts perceptible to us can be validated by Vedic texts inferred to have once existed. The text of Pürvamœmå∫såsütra (1.3.2) I find a little awkward to translate, not so much in itself but in view of the reading of it that is implicitly offered by ‡abara, and more explicitly elsewhere, e.g., in Mœmå∫såkaustubha and Adhvaramœmå∫såkutühalavr¢ tti.10 For anumåna in this context comes to suggest, it seems, not only the log9 ‘Therefore it stands to reason that this prior cognition exists in the case of members of the three highest social orders doing the remembering [i.e., insofar as the people who are remembering are participants in Vedic culture, they are connected with the Veda, and thus can have had a ‘prior cognition’ of Vedic texts that would substantiate their memory], and likewise it stands to reason that they could have forgotten [the actual texts]. For these two reasons we can infer the existence of texts [now lost], and thus smr¢ti is a source of valid knowledge’ (tad upapannatvåt pürvavij∞ånasya traivarñikånå∫ smaratå∫ vismarañasya [sc., granthasya] copapannatvåd granthånumånam upapadyata iti pramåñam smr¢tiΔ [‡åbarabhåßya, p. 77, lines 7-8, mispunctuated in the original]). 10 See Mœmå∫såkaustubha, vol. 1, p. 12 (which in part is also arguing that the logical operation at issue here is arthåpatti rather than anumåna; this is Kumårila’s main concern, see below at n. 22 infra): ‘The meaning of the sütra is as follows: smr¢ti [and practice], insofar as it arises ‘after’ [anu-], i.e., after perception, is referred to as the source of knowledge termed ‘anumåna,’ consisting of the fact that people in the Vedic tradition would otherwise have never so firmly accepted [the smr¢tis] [were they not derived from the Vedas, which in turn] necessarily entails the assumption that their basis is †ruti. For this reason smr¢ti would be ‘valid’’ (sütra∫ tu smr¢tyådi yato mülabhüta†rutikalpaka∫ dr¢∂havaidikaparigrahånyathånupapattirüpam anu pa†cåt pratyakßottara∫ pravr¢ttatvåd anumånapadåbhidheya∫ pramåña∫ vidyate ataΔ pramåña∫ syåt). See also Adhvaramœmå∫såkutühalavr¢tti, vol. 1, p. 60: ‘Insofar as it is based on †ruti, ‘smr¢ti would be valid,’ for there is an inferential sign prompting the inference of the †ruti text that forms the basis of smr¢ti, namely smr¢ti itself’ (†rutimülakatayå smr¢tiΔ pramåña∫ syåt yataΔ smr¢timülabhüta†rutåv anumåpaka∫ liõgam asti smr¢tir eva).

II. Discourse, Conditions and Dynamics of Tradition in South Asia

49

ical operation of inference itself, but also the Vedic text that is thereby inferred. In this latter sense anumåna can be substituted for what is elsewhere called smr¢ti, precisely as pratyakßa, ‘sensory perception’, can take on the signification ‘Vedic texts perceived’ (or even, tout court, pramåña, ‘source of valid knowledge’), and replace †ruti both in Mœmå∫så and elsewhere. The semantic weight that I think can be felt in pratyakßa and anumåna, which helps us toward a historically more accurate understanding of smr¢ti, is corroborated by other usages in the sütras, of which I shall discuss only two. The Holåkådhikaraña of the Pürvamœmå∫såsütra concerns the generalizability of regional texts and customs. A convenient example is cited by Bhå™™adœpikå: “The Gautamadharmasütras are read only by members of the Chåndogya †åkha. Are its injunctions restricted to them or not?” (Bhå™™adœpikå, p. 61). The prima facie view of the sütras is: “Insofar as the inference [sc., of a †ruti basis] can be restricted/localized, the source-of-valid-knowledge [thus inferred] would be implicated in that [i.e. would have to be considered restricted/localized in applicability]” (anumånavyavasthånåt tatsa∫yukta∫ pramåña∫ syåt [Pürvamœmå∫såsütra, 1.3.15]).11 Here, as Nyåyasudhå clearly explains, “‘inference’ refers to smr¢ti (and åcåra), while ‘source of valid knowledge’ has reference to †ruti”.12 The devatådhikaraña of the Brahmasütras addresses problems connected with the hypothesis of the corporeality of the gods. The prima facie view holds that, since their corporeality would entail mortality and this in turn would be inconsistent with the eternality of the Vedic texts (for these refer to the gods, and eternal texts can make no reference to the ‘historical’),13 the gods cannot be corporeal. The answer: “As for the [argument based on the eternality of] sacred word –it is false, because [the gods etc.] ‘are produced from’ these [words], as is proved by ‘perception’ and ‘inference’” (†abda iti cen nåtaΔ prabhavåt pratyakßånumånåbhyåm [Brahmasütra, 1.3.28]). The commentaries here unanimously and correctly identify the reference of the technical terms, ‘perception’ connoting †ruti, and ‘inference’ smr¢ti. This interpretation is corroborated further by the ‘responsion’ in Brahmasütra, 1.3.30 (dar†anåt smr¢te† ca), and by a wide range of additional variations of the formula in the Pürvamœmå∫så-, Vedånta-, and Dharma- sütras.14 11 See Adhvaramœmœ∫såkutühalavr¢tti, ad loc.: [anumånasya] vyavasthånåt de†abhedena vyavasthitatvåt tatsa∫yukta∫ de†abhedasa∫yuktam eva †rutirüpa∫ pramåñam. 12 Anumåna†abdasya smr¢tyåcåravißayatva∫ pramåña†abdasya ca †rutivißayatvam (Nyåyasudhå, p. 245, l. 29 [commenting on Tantravårttika , vol. 2, p. 173, l. 20]). See also Adhvaramœmå∫såkutühalavr¢tti, vol. 1, p. 85; Jaiminœyasütrårthasaõgraha, p. 106. 13 What is at issue is the mantrårthavådånityasa∫yogaparihåranyåya, though this is not mentioned in any of the discussions ad loc. See, also, Pollock 1989c: note 25. 14 dr¢ß™asmr¢tibhyåm; dar†ayati cårtho ’pi smaryate; pratyakßånumånåbhyåm; †abdånumånåbhyåm; dar†ayata† caiva∫ pratyakßånumåne (Brahmasütra, 3.1.8; 3.2.17; 3.2.24; 3.3.31; 4.4.20); pratyakßasa∫yogåt (i.e., på™hasya pratyakßatva in ‡åbarabhåßya on Pürvamœmå∫såsütra 5.2.21). See Pürvamœmå∫såsütra, 3.4.28 [vedasa∫yogåt]; 3.1.13 [†rutisa∫yogåt]; also Pürva-

50

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

How is it that ‘perception’ and ‘inference’ imply what at the same time is embraced by the terms †ruti and smr¢ti ? ‡aõkara on the Brahmasütra passage just cited argues from the analogy between the concepts: “‘Perception’ refers to †ruti because †ruti is independently valid [with respect to transcendent things, as perception is with respect to perceptible things]; ‘inference’ refers to smr¢ti because it is dependently valid [like inference, which originates only in dependence on perception]”. While this figurative interpretation may be doctrinally sound epistemology (sound for the Mœmå∫så system, at least), I am not so sure that, historically viewed, it is a convincing explanation. It seems to me that there is more than analogy at work in the use of the term pratyakßa to refer to †ruti, and anumåna to refer to smr¢ti. Both sets of terms appear to emerge out of the same complex of ideas represented in the Mœmå∫så reflections on the authority of texts and practices not explicitly warranted by the Veda. These texts and practices, insofar as they relate to dharma, secure validity by way of their claim to be based on Vedic texts –there exists no other source of dharma– but Vedic texts for one reason or another not accessible to us. Those that are indeed accessible are perceptible, they are something we can actually hear during instruction in recitation (when a student repeats what is pronounced in the mouth of his teacher [gurumukhoccårañånüccåraña]) and in daily repetition (svådhyåya). This is what, in the eyes of ‡abara, validated Vedic memory (pratyakßenopalabdhatvåd granthasya etc., see above). And this, finally, is what the word †ruti actually means according to the etymology still current among traditional teachers: “The Veda, insofar as it is audible to everyone, is called ‘†ruti’” (vedasya sarvaiΔ †rüyamåñatvåt †rutitvam [karmådau ktin]).15 Yet other texts and practices relating to dharma can have validity in the realm of Sanskrit thought inasmuch as they necessarily lead us to infer the existence at some other time or some other place or in some presently inaccessible mode, of Vedic texts as their basis; we no longer hear (recite) these texts word-for-word, but their sense is preserved in memory: “Smr¢ti is so called because by means of it the dharma of the Veda is remembered” (smaryate vedadharmo ’nena [karañe ktin]), again according to traditional etymology.16 In short, mœmå∫såsütra, 7.3.4 [pratyakßåt]; 1.4.14 [pratyakßa-vidhånåt]; 3.5.33 [-upade†åt]; 5.4.22 [†iß™atvåt]). Compare Baudhåyanadharmasütra, 1.1.6, where †rutipratyakßahetavaΔ is juxtaposed to anumånaj∞åΔ. 15 Panditaraja K. Balasubrahmanya Sastry, personal communication. This is the understanding of Våcaspatyam, which is what I translate in the text (s.v., p. 5155). No doubt the original signification of the verbal root †ru and of †ruti in this context is hearing/learning (sc., from one’s teacher); this connotation is pervasive in Sanskrit, and its antiquity is shown in the Buddhist fossil eva∫ mayå suta∫, part of the nidåna of a sütra, which furnishes what (in stark contrast to the Brahmanical use of †ruti) I would call the historical authentication of the text (see also Lamotte 1958: 142-43). The Pali Dictionary’s translation ‘inspired tradition, sacred lore’, raises its own set of problems. 16 Panditaraja K. Balasubrahmanya Sastry, personal communication. See again Våcaspatyam, s.v., p. 5373; ‡abdakalpadruma, s.v., vol. 5, p. 464.

II. Discourse, Conditions and Dynamics of Tradition in South Asia

51

†ruti means nothing other than ‘(Veda) actually now perceived aurally (in recitation)’, i.e. extant or available; smr¢ti, nothing other than ‘(Veda) that is remembered’, i.e. material that, having once been heard in recitation, no longer is, but remains inferentially recoverable from present reformulations (in language or practice) as having once existed as part of a Vedic corpus. Both refer in their primary connotation to one and the same thing –the Veda, whether as something actually recited or as something whose substance only can still be recalled; pa™hyamånasmaryamåñavedaΔ, as Kumårila puts it (Tantravårttika, vol. 2, p. 94, l. 2). This historically original and radical signification of †ruti/smr¢ti has considerable implications for our understanding of Sanskrit intellectual history, and I will try below to spell out some of these. I have been concerned in this section with reviewing Mœmå∫så’s epistemological analysis of †ruti and smr¢ti in order to reclaim the original signification of these terms, which is dependent on such an analysis. This reclamation stands, I think, even if the semantic distinction I draw for pratyakßa and anumåna is found to be overly fine. A thorough analysis of all the terms in the early literature is desirable, but not easily done with the research tools available. It has not been possible to conduct a sufficiently thorough lexical study; the evidence at hand only suggests that the technical use of the term smr¢ti and its being paired with †ruti belong to the very latest stratum of Vedic literature, and became current only in the post-sütra period. Since the epistemological background presupposed in the original meanings of †ruti and smr¢ti is provided by Mœmå∫så, one might hypothesize that Mœmå∫så itself was responsible for this currency.17 The controversy over how we are to explain the unavailability of the Vedic texts whose memory smr¢ti preserves is long and complex, with Pürva and Uttara Mœmå∫så, Nyåya, and Vyåkaraña all contributing to the discussion. There is no space for a detailed presentation here. Two of the prominent arguments are reasonably well-known. Early Mœmå∫så holds that the smr¢tis are derived from Vedic recensions now forgotten or geographically or otherwise inaccessible to us. Nyåya reasons that these recensions must have actually disappeared (this position is best articulated in Nyåyakusumå∞jali, but it is far earlier than Udayana). Kumårila concurrently maintains that the smr¢tis may derive commandments inferred from mantras and arthavådas that exist in extant recensions but are scattered randomly through17 smr¢ti in the relevant sense appears not to occur before Taittirœyårañyaka 1.2.1 (smr¢tiΔ pratyakßam aitihyam anumåna† catuß™ayam) where it is significantly listed with †ruti, itihåsapuråña, and (according to commentaries) åcåra, though this last equation could use additional supporting evidence; †ruti perhaps not before Månava†rautasütra, 182.4. These conclusions are based in part on as yet unpublished materials collected for the Sanskrit Dictionary on Historical Principles of the Deccan College of Pune, for which I thank Dr. Prakash Joshi. Note that the word smr¢ti (sati) never appears in Buddhist texts, Pali or Sanskrit, in the sense it develops in the Mœmå∫så tradition. For †ruta (suta), see n. 15 supra.

52

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

out them, or beyond our powers to interpret properly.18 We may assume that this position was formulated in opposition to Nyåya and thus must be relatively late. The third view, that of Pråbhåkara Mœmå∫så, may be less familiar. It argues that the Vedic texts from which smr¢ti derives were never actually extant, but are only infinitely inferable. Thus ‡ålikanåtha: In the case of the word †åstra [=Veda], likewise [as in the case of the word pramåña, see p. 192], there are two meanings: ‘knowledge’ and ‘that by which knowledge arises’, i.e. holy word. As for the latter, it is of two sorts, perceptible or inferable. What leads us to infer holy word? A statement of smr¢ti, such as ‘the aß™akås [the eighth-day ancestral rites] are to be performed’. How do we infer holy word? First of all, this smr¢ti is accepted as valid by all members of the three highest varñas unchallenged. This would be inexplicable unless the statement had some foundation. Perception and the other sources of valid knowledge cannot supply this foundation, because they do not operate on what is potential action. On the other hand, †åstra can be the source of this smr¢ti statement, since it is through this smr¢ti that we gain knowledge about a transcendent potential action [apürvakårya, knowledge that we can gain only through †åstra]. A possible objection here is that †åstra, too, cannot legitimately be posited as its source, since however zealously one examines †åstra, one cannot perceive any such statement. A scriptural statement that is not perceived cannot communicate anything, and if it cannot communicate anything, it cannot function as the source. It is true that Manu and the rest [of the compilers of the smr¢tis] did not actually perceive that scriptural statement any more than we can today. But, like us, they could make an inference. They observed that a given smr¢ti text was accepted by the mahåjanas, and so they could infer as its source a scriptural text, which had likewise been inferred by the compiler of the smr¢ti in question on the basis of some prior smr¢ti. Thus the smr¢ti-tradition that provokes the inference [of scriptural foundation] is beginningless, and given the possibility of this, the inference [of the smr¢ti’s scriptural foundation] cannot be invalidated.19

In any case, it should be clear that in Sanskrit intellectual history the dispute about smr¢ti focused largely on the precise nature of its derivation from †ruti; the fact of its being so derived was not questioned, nor consequently the primary signification and implication of its reference. 18 See, respectively, ‡åbarabhåßya, vol. 2, p. 77; Tantravårttika, vol. 2, p. 76, ll. 4-5 (with Nyåyasudhå, p. 123, ll. 19 ff.); Nyåyakusumå∞jali, chap. 2 (see, also, Åpastambadharmasütra, 4.1.10 + 1.4.8); Tantravårttika, vol. 2, p. 76, ll. 4-5; p. 105, ll. 5, 10 ff; p. 112, ll. 12-13; p. 113, ll. 14 ff; p. 145 (inferring smr¢timüla from åcåra, and thence †rutimüla; possibly also ‡abara on Pürvamœmå∫såsütra, 1.3.2, pp. 78-79 [dar†ana = †ruti]). See also, more generally, Våkyapadœya, 1.7, p. 173; Govindasvåmi on Baudhåyanadharmasütra, 1.3. 19 Prakarañapa∞cikå, pp. 249-250. On mahåjana, see n. 28 infra. Kumårila’s reasonable response (see Tantravår™™ika, vol. 2, p. 75, ll. 21-22) to such a position is to ask how a Vedic text never articulated can ever have been perceived, so as to become an object of memory.

II. Discourse, Conditions and Dynamics of Tradition in South Asia

53

This primary signification is confirmed in an important passage from the Nyåyama∞jarœ, which is noteworthy also in reminding us that, for all the differences in their analysis of Vedic ‘revelation’, Nyåya and Mœmåμså, like the entire Brahmanical tradition, agree in their understanding of the authority of smr¢ti: At all events, however we are to explain it [i.e. the loss of the texts from which smr¢ti derives], the Veda alone must be assumed to form the basis of these [smr¢ti] texts, and nothing else, since nothing else functions as a valid source of knowledge [with respect to dharma]. Moreover, only on this assumption are we doing justice to the term universally employed for these texts, namely ‘smr¢ti’. For were they based on perception [e.g., the perception of a yogin], they would be Veda-like [since according to Nyåya, the Veda is derived from the perception of God], and then what would be the point of using the word ‘memory’ to refer to them? [...] The Veda is two-fold, that which is available for us to hear and that which we must infer. ‡ruti is that which is currently audible [†rüyamåña† ca †rutir ity ucyate], smr¢ti is that which is inferable.20

Nowhere in any shastric analysis of the nature of smr¢ti, then, do we find it juxtaposed to †ruti the way Indology has always juxtaposed it, as inherently more recent, less authoritative, somehow independent and human in origin, and standing in opposition, or subordinate, to †ruti. What smr¢ t i means in classical Sanskrit culture emerges vividly in the Tantravårttika. Kumårila summarizes his view of the relationship of †ruti and smr¢ t i in the context of discussing one of the problems I raised above and left unanswered: how the memory of the Veda (smr¢ t i) and the Veda remembered (†ruti) may be ultimately distinguished, inasmuch as when reciting texts we are remembering them, and when remembering them we perforce do so in some stable and, at least potentially, recitative form. Pertinent to this problem is the question whether the actual sequence of phonemes (varñånupürvœ) of †ruti is eternal, a long and complex controversy that must await analysis elsewhere. This feature of text-invariability, along with transcendence (apaurußeyatva) and ‘autonomous authoritativeness’ (svatantrapråmåñyam), continues undoubtedly to characterize the Vedas and the Vedas uniquely in the minds of all later Mœmå∫sakas. They address all such features in their derivation of smr¢ t i; what concerns us now is the general doctrine of its nature and authority, which receives its classical and orthodox formulation from Kumårila: 20 Nyåyama∞jarœ, vol. 1, p. 372, l. 9 - p. 373, l. 6. The last two sentences are contained in a pürvapakßa (the position of the Pråbhåkara school), but its provisionality relates to the controversy recounted above (at n. 19 supra); there is no reason to suggest that Jayanta doubts the Pråbhåkara philology. His source, incidentally, is likely to have been Prakarañapa∞cikå, p. 249 (sa ca [†abdaΔ] dvividhaΔ, pratyakßo ’numeya† ca).

54

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

[A smr¢ti text] condenses rules encoded in a desultory fashion in Vedic recensions other [than those commonly met with], and which are still actually available to other men.21 Since they are not recited in the course of the non-†rauta rites to which they refer, they were made available in compilations that reproduce their sense –their literal forms were not reproduced for fear that [such a digest] might adversely affect the tradition of Vedic recitation [see vol. 2, p. 76, l. 6] [in the schools that preserve the recitation of these texts]. Although the actual Vedic texts are now hidden to our eyes, these [smr¢tis] ‘manifest’ them, in the same way that [the †ruti texts themselves are manifested] by the various articulatory sounds. [The argument that mediation of the Veda via the compilers of the smr¢ti weakens the claim of Vedic status could apply likewise to mediation via the teacher of Vedic recitation. However:] Viewing their teacher [of Vedic recitation] as trustworthy, students accept his claim that a given passage is recited in the Vedas whether or not it is recited [by the students themselves]. The statements of the authors of the [kalpa- and/or dharma-] sütras are exactly like those of such a teacher. They do nothing more or less than communicate the Vedic statements in their own particular form. They are consequently not to be devalued as mere human creations, being no more human creations than [†ruti texts themselves, which require for manifestation the human effort of] the expulsion of palatal and the rest of the articulatory sounds. For it is one and the same Veda, of equal validity, that men make known whether they do so by remembering it or by reciting it. Even the Veda, when not being recited, exists in the reciters merely in the form of latent impressions it leaves behind, or in memory traces these impressions generate [and thus the memory of the Veda is ontologically no different from the Veda as remembered in smr¢ti ]. Consequently, when the content of a Vedic passage is related by someone, this content is identical as remembered in smr¢ti to that recited in †ruti, and so cannot be invalidated by any reasoning.22

3. ‘Tradition’ Is ‘Revelation’. In the very construction of smr¢ti as a category is encoded its transcendent legitimacy. In early Mœmå∫så, however, this construction is 21 Here viprakœrña refers to the fact that rules relating to purußadharma (as opposed to kratvartha vidhis) are encoded in extant Vedas in a desultory way, and it is the purpose of smr¢ti to make these easily accessible. See, especially, Tantravårttika, vol. 2, p. 145, l. 23; Nyåyasudhå, p. 214 infra. 22 Tantravårttika, vol. 2, p. 104, ll. 15 ff. See, also, ‡ålikanåtha: ‘This [inferred holy word] is Veda, because it is transcendent speech, and that is all the word ‘Veda’ signifies’ (Prakarañapa∞cikå, p. 251). For Ajitå (the earliest commentary on the Tantravårttika), the difference between smr¢ti and †ruti is that in the former, the memory and the perception it presupposes belong to a second party; in the latter, they belong to oneself (Ajitå, pp. 32-33). In the context of discussing the question whether, when contradicting †ruti, smr¢ t i is cancelled or constitutes a legitimate option, Bhavanåtha critiques Kumårila’s vårttika ‘For the very Veda [...]’ saying: ‘Just as †ruti is manifested by articulation (på™ha) [the way smr¢ti is], so †ruti itself [like smr¢ti] is inferred, and thus [on neither account] is there any difference between the two. For even when †ruti is articulated, the fact that it is †ruti is something we must infer. Such is

II. Discourse, Conditions and Dynamics of Tradition in South Asia

55

marked by a certain tentativeness. Not all the texts and acts of the members of Vedic culture, simply because they are members (kartr¢såmånya), may be legitimated by the logic of their derivation from †ruti. A number of conditions are introduced into the equation of †ruti and smr¢ti that would work to disqualify a text or practice for canonization and scriptural authority. Such include 1) a smr¢ti’s contradicting †ruti, 2) its exhibiting evidence of self-interest or 3) an absence of transcendental content, or 4) its falling outside what in a concrete and narrow view could be included within the Vedic ‘canon’. But all of these limitations are ultimately eliminated in lateclassical Mœmå∫så. Each of these topics is large and important, and Mœmå∫så discusses them at length and with complex arguments that again it is not possible to recapitulate here. Only a few important lines of development can be schematically indicated here. 1) The Movement from Contradiction to Non-contradiction. Contradiction between smr¢ti and †ruti would inhibit the inference that what is remembered is (in any of several senses) authentically Vedic (Pürvamœmå∫såsütra, 1.3.3). It is one of the principal tasks of Kumårila in the Tantravårttika on the smr¢tipåda, however, to eliminate the theoretical possibility of such contradiction (which had been a central interpretative principle in early Mœmå∫så; see for example ‡abara on Pürvamœ∫å∫såsütra, 6.1.13-15; 6.1.20). He does this by a detailed empirical analysis of each of ‡abara’s examples (vol. 2, pp. 105, ll. 13 ff.), concluding: “Therefore, we scarcely ever find contradiction between smr¢ti and †ruti [...] [p. 111, ll. 15-16] [...] Given the possibility that the Vedic source of a smr¢ti may be located in some other Vedic school, we cannot accept the position that it can ever be totally dismissed”.23 This liberates the full potential of the legitimation power of all elite Sanskrit discourse, so long as ‘interest’ itself is never explicitly analyzed as a category. 2) ‘Interest’. Early Mœmå∫så holds that no memory can count as Vedic if some ‘interest’ or ‘motive’ (hetu, kåraña) is therein evident (Pürvamœmå∫såsütra, 1.3.4). This is so because the Veda is defined precisely as that which alone refers, and exclusively refers, to the realm beyond the realm of interests (apråpte vå †åstram arthavat what [Kumårila] intends in the vårttika, ‘For the very Veda [...]’ [...] Now it is true that †ruti and smr¢ti are equal in being manifested through articulation (på™ha). And while the one being †ruti and the other smr¢ti are equally derived from traditional usage (vr¢ddhavyavahåra), smr¢ti is unequal in having to be inferred [as deriving] from †ruti’ (Mœmå∫sånayaviveka, pp. 83-84). Varadaråja, ad loc., explains: ‘The categorization of the two genres †ruti and smr¢ti is traditional. If one argued that a discourse had to be inferred to be †ruti, the way smr¢ti has to be inferred to derive from †ruti, this would still not make the two equal [as being both inferential], since smr¢ti likewise would have to be inferred to be smr¢ti [thus adding a second stage of inference]’. 23 Tantravårttika, vol. 2, p. 112, ll. 7-8 (see Nyåyasudhå, p. 158, ll. 30-33). Compare also Nyåyama∞jarœ (p. 375) where the Vedists (svådhyåyåbhiyuktåΔ) hold that no example of †ruti-smr¢ti contradiction exists.

56

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

[Pürvamœmå∫såsütra, 6.2.18]). But interest in Mœmå∫så is never abstractly defined or even theorized.24 The failure to conceptualize interest is a condition for the following: 3) The Convergence of Non-instrumentality and Traditional Practice as Such. Non-instrumental action (the fact that some act is done for an ‘unseen’, other-worldly purpose [adr¢ß™årthatva]) is what for Mœmå∫så essentially characterizes the nature of Vedic commandments. But as Kumårila came to recognize, there is no transcendent commandment that does not have some dimension of instrumentality to it. At the same time, any instrumental act can disclose a dimension of non-instrumentality: the very fact that a practice is enacted the way it traditionally is, instead of in any other of the potentially infinite number of ways, is itself evidence that some transcendent purpose is being served.25 All of this enables the following: 4) The Enlargement of the Canon. The range of texts that can be counted as Vedic in origin was vast already in Kumårila’s day, despite his intention to limit them.26 And he supplies an argument that may have contributed to this enlargement: It is not an inference from the ‘sameness of agents’ (kartr¢såmånyåt) that leads us to postulate a Vedic source for certain smr¢tis, but an assumption based on the fact that the learned of the three varñas accept them (†iß™atraivarñikadr¢∂haparigraha).27 By the time of Jayantabha™™a, such ‘acceptance’ (now mahåjanaprasiddhyanugraha) is explicitly and exclusively a function of a text’s “conformity with the social norms known from the Vedas, such as caste”. The only texts now excluded are the scriptures of the Buddhists and the ‘Sa∫såramocakas’, or ritual murderers, insofar as they “decidedly reject social behavior that is in accordance with caste duty”.28 4. Summary and Conclusions From the moment smr¢ti was recognized as a genre, it secured legitimacy by way of its derivation from ‘transcendent speech’, a process of legitimation fossilized in the very name by which the tradition came to refer to it. Dispute among specialists centered in 24 This is true also in Nyåya. See the extended discussion of the logical necessity of assuming apürva to explain ritual behavior, and the narrow conception of this behavior, in Nyåyakusumå∞jali, 1.8; 2.3 (Kanchipuram ed., pp. 14 ff; 95 ff). 25 This is what later comes to be known as the niyamådr¢ß™a. See, for example, Tantravårttika, vol. 2, p. 78, ll. 11-15; p. 128, ll. 3-4 (with Nyåyasudhå, p. 126, ll. 20-21, 25). 26 Including strictures against sectarian ågamas. See, for example, Tantravårttika, vol. 2, p. 122, ll. 3-5 (on the ‘fourteen or eighteen’ vidyåsthånas); vol. 2, p. 112, ll. 18-19 (on the båhyagranthas, including the På∞caråtra and På†upata ågamas). 27 The argument itself is subtle. See Tantravårttika, vol. 2, p. 76, ll. 21 ff (with Nyåyasudhå, p. 124). 28 Nyåyama∞jarœ, pp. 376 ff (especially p. 377, ll. 1-3; pp. 379 ff). The term mahåjana merits more detailed historical analysis than it has so far received. Contrast for example Derrett (“[…] a synonym for Bråhmaña [...] ‘important person’” [O’Flaherty, Derrett: 56

II. Discourse, Conditions and Dynamics of Tradition in South Asia

57

general only around conceptual issues provoked by this derivation: the nature of the Vedic texts preserved by ‘memory’, and their status vis-à-vis the Vedic texts ‘actually heard’ (for example, in the matter of contradiction between the two); the hermeneutic of recovery of the ‘original’; the reasons for the inaccessibility of this original. It is likely that this conception of smr¢ti was developed by Mœmå∫så as early as the sütras; it had become a topos by the classical period.29 An initial reluctance to admit absolute equipollence of †ruti and smr¢ti was fully neutralized in the medieval period, when Kumårila claimed for all smr¢tis participation in the inerrancy of holy word. It is thus only a slight exaggeration to say that, in the elite discourse of traditional India, there exists no cultural memory –smr¢ti– separate from the memory of the eternally given. This ‘revelation of tradition’ has two faces, which in concluding I would like briefly to delineate. Mœmå∫så’s project of founding smr¢ti upon †ruti, that is, of explaining social-cultural life as deriving from revealed truth, arguably comprises some vision of the ideal. I do not mean just a longing for transcendence or utopia, for some communal existence that the agents believe to be in conformity with cosmic order. I mean more particularly that it exhibits a perceived need to give good reasons, to provide grounds for the way the lifeworld is organized, and thereby to privilege, at least in theory, justification and persuasion over imposition and subjugation. The need to justify presupposes and can nurture a sense of the need for justice. This positive dimension, the presence of an emancipatory value at the core of ideological discourse, is worth recognizing despite the fact that the reasons Mœmå∫så gives, and argues out with stunning acuity, are bad ones, that its logic of tradition is finally illogical, and that the justification it seeks is directed toward achieving an unjustifiable consensus, on purely sectional interests of the social world. The fact that these are sectional interests, and that legitimation by nature emerges from the competition and conflict over legitimacy,30 discloses for us the dark face of the ‘revelation of tradition’. Mœmå∫så’s most significant social-historical role, of course, was as the metalegal framework for dharma†åstra, the explicit program of domination of Sanskrit culture. And the validation of dharma†åstra’s code of asymmetrical power –of illegitimate hierarchy, untouchability, female heteronomy, the degradation of work– depended centrally upon the Mœmå∫så revelation of tradition. Manu’s claim –“this is all based on the Veda” (Månavadharma†åstra, 2.7-8)– n. 8]) and Chemparathy (“[…] une grande multitude de personnes […]” [Chemparathy 1983: 69]). Both scholars are referring basically to the same context, and both can hardly be correct. 29 See Kålidåsa, Raghuva∫†a, 2.2 (†ruter ivårtha∫ smr¢tir anvagacchat). 30 An obvious point, but easily overlooked. See further Bourdieu 1977: 168.

58

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

would be hollow indeed without this prior revalorization of ‘memory’ itself, which his commentators prominently reproduce.31 When tradition and revelation are forced into convergence; when ‘memory’ no longer bears the record of human achievement and ‘tradition’ no longer transmits the heritage of the historical past, the understanding of culture and society as the provisional arrangements of people making and remaking their lifeworld becomes impossible. Smr¢ti may be transmitted in the memory of men, but it has become the memory of the apaurußeya, the transcendent, whereby the structure of the human world itself –now the domain of dharma and thus incomprehensible without smr¢ti texts– is rendered apaurußeya. A culture and society that have ceased to be the products of human agency cease to be conceivable as humanly mutable, and it is this conception –the reification and naturalization of the world– that forms one essential precondition for the maintenance of social power.

BIBLIOGRAPHY Sources Åpastambadharmasütra, edited by A. Chinnaswami Sastri, Chowkhamba Sanskrit Series Office, Varanasi 1932. Baudhåyanadharmasütra,(with the commentary of Govindasvåmin), edited by A. Chinnaswami Sastri, Chaukhamba, Varanasi 1972. Bhagavad Gœtå with Eleven Commentaries, edited by Gajanana Shambhu Sadhale, Gujarati Printing Press, Bombay 1935. Brahmasütra with the Commentary of ‡aõkara, edited by Vasudeva Laxman Shastri Panshikar, Nirnaya Sagar Press, Bombay 1934. 31 See, for example, Medhåtithi on Manusmr¢ti, 2.6: ‘†ruti applies when the actual scriptural words of instruction in dharma are available to hear, smr¢ti applies when they are [only] remembered’ (†rüyate dharmånu†åsana†abdaΔ så †rutih, yatra ca smaryate så smr¢tiΔ [vol. 1, p. 80]); Kullüka on Manusmr¢ti, 2.12: ‘The Veda is the source of knowledge of dharma. In some instances, Veda is perceptible, in others it [must be] inferred from smr¢ti’ (vedo dharmapramåña∫ sa kvacit pratyakßaΔ kvacit smr¢tyanumitaΔ). See also the Gœtåmåhåtmya in Mahåbhårata, 6.43.2 (vulg. ed.): ‘the Gœtå encompasses all †åstras, Hari all gods, the Gañgå all sacred places, Manu all Vedas’. According to a recent book on RSS doctrine, “[…] the Bhagavad Gœtå was theologically considered smriti (non-canonical) by orthodox Hindus. However, the revivalists tended to blur the distinction between non-canonical texts and the canonical †rutis [...] and to treat both sets of texts as equally authoritative” (Andersen 1987: 23 n. 21). The RSS, we can see, is reclaiming or continuing an ancient mode of legitimation. For the Bhagavadgœtå (and Mahåbhårata as a whole) in particular, this had already been explicitly asserted at least as early as Jayatœrtha, who argues that their validity is a function of their being derived from the ultimate valid text, the Vedas (see Bhagavad Gœtå with Eleven Commentaries, p. 13).

II. Discourse, Conditions and Dynamics of Tradition in South Asia

59

Bhartr¢hari, Våkyapadœya, edited by Raghunath Sharma, Sampurnanand Sanskrit Vishvavidyalaya, Varanasi 1976. Bhavanåthami†ra, Mœmå∫sånayaviveka, edited by Subrahmanya Shastri, Rastriya Sanskrit Samsthan, Delhi 1977. Jayantabha™™a, Nyåyama∞jarœ, edited by Gaurinath Shastri, Sampurnanand Sanskrit Vishvavidyalaya, Varanasi 1982. Jayantabha™™a, Nyåyama∞jarœ, edited by K.S. Varadacharya, Oriental Research Institute, Mysore 1969. Kålidåsa, Raghuva∫†a, edited by H.D. Velankar, Nirnaya Sagar Press, Bombay 1948. Khañ∂adeva, Bhå™™adœpikå, edited by Ananta Krishna Sastri, Nirnaya Sagar Press, Bombay 1921. Khañ∂adeva, Mœmå∫såkaustubha, edited by A. Chinnasvami Sastri, Chowkhamba, Varanasi 1923. Kumårila, Tantravårttika. See Mœmå∫sådar†anam. Kumårilabha™™a, Mœmå∫så†lokavårttika, edited by Dwarikadas Sastri, Tara, Varanasi 1978. Kumårilabha™™a, Mœmå∫så†lokavårttika with the commentary of Sucaritami†ra, Kå†ikå, edited by Sambasiva Sastri, Government Press, Trivandrum 1927-1943. Mådhava, Jaiminœyanyåyamålå, edited by Sivadatta Sarma, Anandashrama Press, Pune 1892. Mahåbhårata, edited by V.S. Sukthankar et al., Bhandarkar Oriental Research Institute, Pune 1927-1966. Månava†rautasütra, edited by Friedrich Knauer, n.p., St. Petersburg 1900. Manusmr¢ti, (with the commentary of Kullüka), edited by Vasudevasarma Panasikar, Nirnaya Sagar Press, Bombay 1929. Manusmr¢ti, (with the commentary of Medåtithi), edited by Manasukharai Mor, Gurumandala, Calcutta 1967. Mœmå∫sådar†anam, edited by K.V. Abhyankar, Anandashrama Press, Pune 1970-1976. Mœmå∫sådar†anam, edited by Mahaprabhulal Goswami, Tara, Varanasi 1984. Parame†vara, Jaiminœyasütrårthasa∫graha, edited by Narayana Pillai, University of Travancore, Trivandrum 1951. Paritoßami†ra, Ajitå, edited by Kunio Harikai, ‘Acta Eruditiorum’, n. 4, General Education, Saga Medical School, Saga, Japan 1986. Pårthasårathimi†ra, ‡åstradœpikå , edited by Dharmadatta Jha, Nirnaya Sagar Press, Bombay 1915. Pårthasårathimi†ra, Tantraratna, edited by Ganganath Jha et al., Sampurnanand Sanskrit Vishvavidyalaya, Varanasi 19301979. Pata∞jali, Mahåbhåßya , edited by Bhargavasastri Joshi, Nirnaya Sagar Press, Bombay 1951. Prabhåkara, Br¢hatœ, edited by A. Chinnaswami Sastri, Chowkhamba, Benaras 1929-1933.

60

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Prabhåkara, Br¢hatœ (Tarkapåda), edited by S.K. Ramanatha Shastri, University of Madras, Madras 1934-1936. ‡abara, ‡åbarabhåßya. See Mœmå∫sådar†anam. ‡abdakalpadruma, edited by Radhakantadeva, Chaukambha Sanskrit Series, Varanasi 1961 [1886]. ‡ålikanåthami†ra, Prakarañapa∞cikå, edited by A. Subrahmanya Shastri, Banaras Hindu University Press, Varanasi 1961. Some†varabha™™a, Nyåyasudhå, edited by Mukunda Shastri, Chowkhamba, Varanasi 1909. Taittirœya Årañyaka, edited by Babasastri Phadke, Anandashrama Press, Poona 1981. Udayana, Nyåyakusumå∞jali, edited by Mahaprabhulal Goswami, Mithila Institute, Darbhanga 1972. Udayana, Nyåyakusumå∞jali, Kanakambujam Srinivasan Trust, Kanchipuram 1979. Våcaspatyam, edited by Taranatha Tarkavacaspati Bhattacarya, Chaukhamba Sanskrit Series Office, Varanasi 1962. Våsudeva Dœkßita, Adhvaramœmå∫såkutühalavr¢tti, edited by Pattabhirama Sastri, L.B. Shastri College, Delhi 1968-1969. Yåj∞avalkyasmr¢ti, (with the Commentary of Aparårka), Anandashrama Press, Pune 1903. Yåska, Nirukta, edited by Mukund Jha Bakshi, Nirnaya Sagar Press, Bombay 1930. Studies Andersen, W.K. et al. (1987), The Brotherhood in Saffron: The Rashtriya Svayamsevak Sangh and Hindu Revivalism, Westview, Boulder. Basham, A.L. (1954), The Wonder That Was India, Grove Press, New York. Bechert, H. et al., (eds.) (1979), Einfhrung in die Indologie, Wissenschaftliche Buchgesellschaft, Darmstadt. Botto, O. (1969), Storia delle Letterature d’Oriente, Vol. 3, Letterature antiche dell’India, Vallardi/Societa Editrice Libraria, Milano. Bourdieu, P. (1977), Outline of a Theory of Practice, Cambridge University Press, Cambridge. Chemparathy, G. (1983), L’Autorité du Veda selon les Nyåya-Vai†eßikas, Centre d’Histoire des Religions, Louvain-La-Neuve. Coburn, T. (1984), ‘Scripture’ in India: Towards a Typology of the Word in Hindu Life, in «Journal of the American Academy of Religion», vol. 52, n. 3. Dallapiccola, A.L. (ed.) (1989), The ‡åstric Tradition in the Indian Arts, Steiner, Stuttgart. Davids, T.W. Rhys, Stede, W. (eds.) (1921-1925), The Pali Text Society Pali-English Dictionary, Pali Text Society, London. de Bary, W. et al., (eds.) (1958), Sources of Indian Tradition, Columbia University Press, New York (sec. ed., edited and revised

II. Discourse, Conditions and Dynamics of Tradition in South Asia

61

by Ainslie T. Embree, Columbia University Press, New York, 1988). Eliade, M. (ed.) (1987), Encyclopedia of Religion, Macmillan, New York. Gonda, J. (1960), Die Religionen Indiens, Vol. 1: Veda und alterer Hinduismus, W. Kohlhammer Verlag, Stuttgart. ______ (1975), Vedic Literature (Sa∫hitås and Bråhmañas), Harrassowitz, Wiesbaden. (A History of Indian Literature), vol. 1, fasc. 1. Heesterman, J. (1978), Veda and Dharma, in O’Flaherty, W., Derrett, J.D.M. (eds.) The Concept of Duty, Vikas, Delhi. Lamotte, E. (1958), Histoire du bouddhisme indien, Bibliotheque du Museon, Leuven. Lingat, R. (1973), The Classical Law of India, University of California Press, Berkeley. O’Flaherty, W., Derrett, J.D.M. (eds.) (1978), The Concept of Duty in South Asia, Vikas, Delhi. Lienhard, S., Piovano, I. (eds.) (1997), Lex et Litterae. Studies in Honour of Professor Oscar Botto, Edizioni dell’Orso, Torino. Pollock, S. (1985), The Theory of Practice and the Practice of Theory in Indian Intellectual History, in «Journal of the American Oriental Society», 105, pp. 499-519. ______ (1989a), The Idea of ‡åstra, in Dallapiccola 1989. ______ (1989b), Playing by the Rules: ‡åstra and Sanskrit Literature, in Dallapiccola 1989. ______ (1989c), Mœmå∫så and the Problem of History in Traditional India, in «Journal of the American Oriental Society», 109, pp. 603-610. ______ (1990), From Discourse of Ritual to Discourse of Power in Sanskrit Culture, in «Journal of Ritual Studies», vol. 4, n. 2, pp. 291-320. ______ (1996), The Sanskrit Cosmopolis, A.D. 300-1300: Transculturation, Vernacularization and the Question of Ideology, in Houben, J.E.M. (ed.), Ideology and Status of Sanskrit: Contributions to the History of the Sanskrit language, Brill, Leiden. ______ (2006), The Language of the Gods in the World of Men: Sanskrit, Culture, and Power in Premodern India, University of California Press, Berkeley. Radhakrishnan, S., Moore, Ch. (1973) [1954], A Sourcebook in Indian Philosophy, Princeton University Press, Princeton. Renou, L. (1960), Études védiques et påñiniennes Vol. VI: Le destin du Veda dans l’Inde, E. de Boccard, Paris. Renou, L., Filliozat, J. (1947), L’Inde classique, Adrien Maisonneuve, Paris. Singer, M. (ed.) (1959), Traditional India: Structure and Change, American Folklore Society, Philadelphia. Winternitz, M. (1927), History of Indian Literature, University of Calcutta, Calcutta, vol. 1.

JOHANNES BRONKHORST The Reliability of Tradition

From among the definitions given in dictionaries of the term tradition we must retain the one that specifies that a tradition is a cultural feature (as an attitude, belief, custom, institution) preserved or evolved from the past.1 This definition reminds us that traditions are handed down from generation to generation, but also that most traditions are accompanied by the claim, often implicit, that they preserve an earlier state of affairs. A tradition is therefore something which exists in the present (any present), but which at the same time makes claims about the past. If we assist at a traditional dance performance, we are not merely entertained; we are at the same time informed about how people danced in the past. It is this claim about the past which makes it possible to speak about the reliability of a tradition. Traditions can make an implicit claim about the past which is not true. Indeed, traditions can be newly created (Hobsbawm, Ranger 1983). In that case they are strictly speaking no traditions at all, or at best unreliable traditions. Traditions, moreover, normally have a role to play in the present (each present) in which they occur: they may be linked to nationalistic movements, or to the sense of belonging that unites members of a certain group, or indeed they may be expressions of a religious identity. That is to say, traditions are rarely innocent survivals from a distant past, and far more often factors that play a role in the present. Traditions may be needed, which may tempt certain people to create new ones when the need arises. Reflections like these should remind us of the fact that the study of traditions is not at all the same as the study of history. Traditions may at times provide information about the past, but this is never 1

See s.v. in Webster’s Third New International Dictionary, 1986.

64

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

self-evident, and is always in need of verification. It should also be clear that people who like their traditions do not for that reason necessarily like their past. Indeed, historical research that brings to light that this or that tradition does not really continue a feature or habit from the past may not always be welcomed. The lover of traditional dances may not be pleased to learn that the dances he is so fond of are in fact a recent creation. This implies that traditions, once in place, may have a tendency to force the past into a straight jacket: the past has to be seen in this particular way, and dissonant opinions are not accepted. Classical Indian culture has many traditions, and does not look upon these as mere sources of amusement. Traditions constitute the heart of much that we call classical Indian culture, and no pains are spared to preserve these traditions and keep them alive. This applies to the present, but also to the past. There are plenty of reasons to believe that traditions played an important role during much of Indian history. Since in each tradition a vision of this or that aspect of the past is implied, the network of traditions that make up classical Indian culture is inseparable from a vision of India’s past, which is, to be sure, multifaceted and complex. An especially important tradition, which often serves as a sort of backbone to some of the others and which has a particularly close bearing on this vision of India’s past, is the Vedic tradition. The importance of this tradition, or more precisely of the textual corpus that is preserved by this tradition, is illustrated by the fact that certain other traditions have borrowed its name: Veda. India’s longest, oldest and most important Sanskrit epic, the Mahåbhårata, calls itself the fifth Veda. The fundamental text on Sanskrit dramaturgy and related matters, the Nå™ya†åstra of Bharata, makes a similar claim. Indian medicine is known by the name åyurveda, the Veda of long life. Other traditions claim links to the Veda without necessarily borrowing its name. Obviously these traditions felt that they could add to their prestige by imitating the Veda, or by claiming a close connection with it. The Veda occupies a very special position in the vision of India’s past that came to predominate in brahmanical circles. Briefly put, the Veda is, or is closely connected with, the origin of all there is. The most traditional representatives of Vedic orthodoxy, known by the name Mœmå∫sakas, maintained that the Veda has no beginning in time at all; it has always been there. This they often linked up with the idea that the world has no beginning either, that it too was always there, essentially in the same form in which we know it. Other currents of thought do accept that the world we live in had a beginning in time, but do not accept that the Veda was created along with all the other things that constitute this world; on the contrary, creation itself was determined by, or carried out in accordance with, the words of the Veda. In this view the Veda predates the creation of our present world. The creation of our world itself is often thought of as the most

II. Discourse, Conditions and Dynamics of Tradition in South Asia

65

recent installment of an infinitely long series of creations, which has no beginning in time. The Veda stands above or outside this infinite series, and is sometimes depicted as being pronounced anew at the beginning of each new creation, exactly in the same form as in all the preceding ones. This timelessness of the Veda also finds expression in other ways. The language of the Veda, i.e. Sanskrit, is as eternal and as unchangeable as the Veda itself. Language change does occur, but not in the language of the Veda, but in its corruptions which have led to the many languages that are spoken today. ‘Development’ is hardly the term to be used for this process, which is rather an ongoing process of corruption of the original perfect language which is Sanskrit. The essential timelessness of the Veda –or at any rate its hoary antiquity, which amounts pretty much to the same thing– has not disappeared from India with the arrival of modernity. There may not be all that many people left these days who maintain that the Veda is literally beginningless and eternal, numerous are those who assign to the Veda incredibly ancient dates. Nor has the Veda stopped, in the Indian semi-popular imagination, being the beginning and source of all that it is worth knowing. ‘Research’ discovers evidence for the presence of the most recent scientific and technological developments in the Veda, and many a Hindu may expect that further research into this ancient textual tradition may bring to light useful knowledge such as, for example, a cure for aids. Modern scholarship, one would expect, is not influenced by this traditional attitude towards the Veda. This optimistic expectation is not in total agreement with the facts. Modern indological scholarship, which was initially a european affair, brought along with it its own set of presuppositions, which were in some respects not all that different from the Indian beliefs. Note, to begin with, that the ‘discovery’ of Sanskrit by european scholarship came at a time when the idea of India as the cradle of all civilization had numerous adherents in Europe. Edwin Bryant enumerates a number of representatives of this position, among them the astronomer Bailly and Voltaire, Pierre de Sonnerat, Schelling, Friedrich von Schlegel, and Johann-Gottfried Herder (Bryant 2001: 18 ff). Sanskrit came in this way to be looked upon not just as one branch language of the Indo-European family, but as its parent-language, or at any rate very close to it. Lord A. Curzon, the governor-general of India and eventual chancellor of Oxford, maintained as late as 1855 that “the race of India branched out and multiplied into that of the great Indo-European family”. Scholarly interest for Sanskrit remained for a long time inseparable from the quest for the original Indo-European language. As in India, the study of Sanskrit remained also in Europe for quite a while closely linked to the quest for origins. These romantic ideas about India did not survive for long among serious scholars, at least not in these extreme forms. It was soon discov-

66

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

ered that Sanskrit was not the original Indo-European language. The discovery by archaeologists of the Indus valley civilisation, which in the opinion of many preceded the period in which the Veda was composed, has placed the Veda in a relatively recent historical period. However, in other respects modern scholarship has come up with results which have boosted the idea of the reliability of the Vedic tradition. The study of early phonological texts has shown that the oral preservation of at least certain Vedic texts has been more faithful than one might have considered possible. Max Mller was the first to edit and study the R¢gvedapråti†åkhya, an old text which describes the phonology of the R¢gveda in great detail. Mller discovered in this way that the R¢gveda, which is the oldest text of the Vedic corpus, had been handed down for a period of well over two thousand years without the slightest change even in a single sound.2 Some scholars nowadays go to the extent of stating that present-day recitation preserves the R¢gveda and other Vedic texts so well that one might speak of a tape-recording (Witzel 1995: 91). The classical Indian belief in the unchangeable nature of the Veda has in a way been vindicated by these and other similar findings. Modern scholarship has discarded many beliefs to which it was originally attracted, for whatever reason. No, Sanskrit is no longer the original language, it is not even the original Indo-European language. No, India no longer represents the origin of all culture, nor of all philosophy and wisdom. Yes, ancient India culture was ‘just another’ major culture, less old than some (e.g., Egypt), older than others (e.g., Islam). One might like to think that modern scholarship has been able to free itself from all unreliable presuppositions and unfounded beliefs. As so often, reality is more complex. There can be no doubt that in-depth research has dismantled numerous preconceived ideas, both those of Indian origin and those that were European. The belief in an original invasion by conquering Aryans who brought civilisation to India, a belief so convenient to Western colonisers and invented by Europeans, is one of those that have fallen by the wayside. Indeed, the reaction in scholarship against colonialism and its intellectual heritage has done much good in unmasking certain types of presuppositions. But not all presuppositions are connected with colonialism or colonialist attitudes. Presuppositions that are pleasing to those belonging to the culture studied will be less systematically subjected to critical assessment and may linger on, either because no one is 2 “Wenn man bedenkt, dass das Pråti†åkhya nicht nur Tausende von Stellen aus den beiden Texten (i.e., padapå™ha and sa∫hitåpå™ha of the R¢gveda, JB) citirt, sondern auch die anscheinend geringfgigsten Abweichungen des einen von dem andern auf das genaueste registrirt, und dass in allen wesentlichen Punkten unsere besten Handschriften der beiden Texte mit den Angaben des Pråti†åkhya bereinstimmen, so darf man wohl mit Zuversicht schliessen, dass wir wirklich den Text des Rig-Veda so besitzen, wie er vor mehr are 2000 Jahren den Verfassern des Pråti†åkhya vorlag”. (Mller 1869: 3)

II. Discourse, Conditions and Dynamics of Tradition in South Asia

67

aware of them, or because it is considered politically correct (or merely convenient) to leave them in place. It is to some of these ideas that we now turn. We have seen that the Veda, in brahmanical tradition, is at the origin of almost all there is. In this form this idea has no appeal to modern scholarship. However, in a weakened form it is still very much alive, even among serious researchers. Questions about the origin of this or that feature of classical Indian culture are routinely investigated by tracing its roots in the Veda. At first sight this may seem reasonable, given that the oldest parts of the Veda are certainly the oldest literary remains we have from that part of the world. Yet on closer inspection it will become clear that it only makes sense if one accepts another presupposition, namely, that all those features of classical Indian culture belong to traditions that have their origin in the Veda. This is not self-evidently the case. Other influences may have been at work which were altogether different from the Veda and its adherents. It goes without saying that the question here raised has to be investigated separately for each case that may attract our attention, and that general and unsupported assertions are of little use. Some classical traditions may derive directly from Vedic roots, others may not. Unfortunately modern scholarship often avoids the question altogether, and has a tendency to dive straight into the Vedic texts. An example is the research into the origins of the Sanskrit drama. In this case it is particularly simple to think of a non-Vedic source. The classical Sanskrit drama being a court drama, it is hard not to think of the rulers who, on the Indian sub-continent itself, cultivated a courtly drama not long before the Sanskrit drama manifested itself. These rulers were, of course, the Greeks, whose historical presence in northwestern India (and whose love for drama) is not contested. In spite of this, indological research discards the presence of the Greeks as a possible factor in the development of the Sanskrit drama, and prefers to concentrate on possible Vedic roots, knowing all the while that Vedic culture had no courtly drama and late-Vedic and early post-Vedic culture no sympathy for this kind of entertainment. By way of justification for this omission indologists tend to refer back to arguments which were originally presented by Sylvain Lévi at the end of the nineteenth century, but which are outdated in the present state of our knowledge and stopped being supported by their originator himself later on in his life. In spite of this, scholars refrain from carrying out a renewed reflection on this issue and obviously feel more comfortable with their old habit of searching for Vedic antecedents (Bronkhorst 2004). There are serious reasons for exercising restraint while looking for the origin of everything Indian in the Veda. It is becoming ever more clear that it is not justified to identify the Aryans –i.e. those who called themselves årya, the authors and early users of the Vedic

68

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

texts– with the Indo-Aryans, the speakers of Indo-Aryan languages.3 Those who adhered to the årya ideology (the ‘Aryans’) were no doubt a sub-group of the Indo-Aryans, but it is by no means evident that they were in the early centuries more than a minority. And it is not at all certain that this minority was in any way representative of the other speakers of Indo-Aryan. Indeed, “the emergence of an årya ideology can be traced [...] to the geographical milieu of the R¢gvedic hymns, bounded by the Indus and Sarasvatœ rivers, and need not be linked to the spread of Indo-Aryan languages” (Erdosy 1995: 3). Few scholars nowadays would doubt that Indian civilisation has other sources than only the Veda. The very presence in South-Asia of speakers of languages belonging to other families, such as Dravidian and Munda, supports this. Scholars like to speculate what elements in Indian civilisation might have ‘pre-Aryan’ roots. However, even the early speakers of Indo-Aryan languages themselves were most probably divided in groups many of which did not adhere to, or even know about, the årya ideology that finds expression in the Vedic corpus. Unfortunately only the Vedic Indians have left us a literary corpus whose oldest parts date back to a period from which we have no other literary remains. A close inspection of the other literary remains that we do possess (all of them admittedly younger than the oldest parts of the Veda) indicates that, among the speakers of Indo-Aryan, there existed at least one other important ideology, utterly different from the årya ideology, which left its traces not only in non-Vedic movements and religions, but deeply influenced the tradition which saw itself as the continuation of the Vedic tradition: brahmanism or, if you like, hinduism. I am not the first to draw attention to the ideology of those who often appear in the texts under the name ‡ramañas. In order to do justice to my predecessors, but also to introduce some important qualifications, I cite a passage from the third edition of G.C. Pande’s Studies in the Origins of Buddhism:4 We find, thus, that in the Vedic period there existed two distinct religious and cultural traditions –the strictly orthodox and Aryan tradition of the Bråhmañas, and, on the fringe of their society, the straggling culture of the Munis and ‡ramañas, most probably going 3 Parpola writes: “we must distinguish between the modern use of the name ‘Aryan’ to denote a branch of the Indo-European language family, and the ancient tribal name used of themselves by many, but not necessarily all, peoples who have spoken those languages” (Parpola 1988: 219). Similarly Erdosy: “Until recently, archaeologists, and to a lesser extent linguists, had persistently confused ‘Aryans’ with ‘Indo-Aryans’” (Erdosy 1995: 3). Many scholars distinguish, often on linguistic grounds, two or more waves of immigration of ‘Aryans’, only one of which is responsible for the production of the Vedas. See Deshpande 1995: 70 ff; Witzel 1995a: 322 ff). 4 Other authors who have drawn attention to the separate tradition of the ‡ramañas include A. K. Warder and Padmanabh S. Jaini.

II. Discourse, Conditions and Dynamics of Tradition in South Asia

69

back to pre-Vedic and pre-Aryan origins. Towards the close of the Vedic period, the two streams tended to mingle and the result was that great religious ferment from which Buddhism originated. (Pande 1983: 261)

The part of this citation which I fully support concerns the “[…] two distinct religious and cultural traditions” that existed in the Vedic period. Besides the årya ideology incorporated in the Veda there was the ideology of the ‡ramañas. This ideology belonged to certain ascetics commonly referred to as ‡ramañas, but obviously not only to them. Ascetics come from social milieus, and are never more than a tiny minority in their particular milieu. The ideology of the ‡ramañas (to be discussed below) was not the exclusive property of those who left the world to become ascetics, but characterized the community in which they grew up. It is significant that Pande, in spite of drawing this important distinction between two altogether different cultures that coexisted in the Vedic period, feels obliged to speculate as to the origins of the culture of the ‡ramañas. He calls it a ‘straggling culture’, which suggests that it had wandered off from the earlier Vedic culture. He also speculates that the culture of the ‡ramañas most probably had pre-Vedic and pre-Aryan origins. All this is speculation which is not based on any reliable evidence. It merely distracts attention from the important observation that already several centuries before the beginning of the Common Era (i.e. at the time when Buddhism and Jainism made their appearance) there existed in northern India an identifiable culture, the culture of the ‡ramañas, which had no visible links with Vedic culture. There is a further element in Pande’s passage which has to be considered with much caution. It is the mention of Munis besides ‡ramañas. This mention suggests that there is a historical connection between the ‡ramañas here talked about and the Munis and other marginal figures referred to in early Vedic texts from the R¢gveda onward. The assumption of such a connection could be misleading, as will become clear below. In the terminology here adopted, the ‡ramaña tradition is the one which has given rise to religious movements such as Buddhism, Jainism and Åjœvikism; all of these can in a way be said to belong to this tradition. This ‡ramaña tradition is distinct from the Vedic tradition and cannot be derived from it. A variety of arguments support this position. They are unfortunately rarely taken into consideration by the majority of scholars, who go on repeating the by now classical opposite position according to which certain developments recorded in Vedic literature are the basis from which all those other religious movements arose. I am primarily referring to the ideas about karma and rebirth, and the possibility of liberation from these, which we find in the Vedic Upanißads. These ideas –so the argument runs– arose at the time of the Upanißads; all developments in which they

70

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

play a role must therefore be more recent. This way of reasoning is at the basis of all subsequent reflection, whether it concerns the background of early Buddhism and Jainism, or questions of chronology. A structure of mutually coherent ideas has thus been erected, from which it is not easy for the modern scholar to escape. And yet there are clear indications that it is wrong. This is not the occasion to repeat all the arguments against it which I have presented elsewhere, but some cannot be mentioned often enough. The conviction, for example, that the ideas of karma, rebirth and liberation made their first appearance in the Upanißads is contradicted by those Upanißads themselves. They often ascribe those ideas to outsiders, and on one occasion the Kßatriya who supposedly revealed this knowledge to the Brahmins pointed out to them that, because Brahmins had not been aware of this important knowledge, worldly power had so far belonged to Kßatriyas. I am not, of course, trying to revive the old theory according to which these new ideas had been thought out by Kßatriyas.5 My emphasis is quite different: these upanißadic passages may well be the only ones in the whole of sacred brahmanical literature –Vedic and postVedic– where it is publicly admitted that a new idea was introduced into the Vedic tradition by outsiders. We are well advised to take this admission seriously. It is also clear that these new ideas were ignored for a long time by many within the brahmanical tradition. The Mœmå∫sakas –representatives of the most orthodox Brahmins if there are any– still ignored them a thousand years after these upanißadic passages had been composed. Other supposedly brahmanical texts, such as the Mahåbhårata, appear to be unaware of them in many of their narrative portions; these ideas become more prominent in the didactic parts (Brockington 1998: 244 ff). There can be no doubt that the ideas of karma, rebirth, and liberation did gradually find their way into the brahmanical traditions, but the nature of this process of infiltration has been obscured in modern research by the belief that these ideas were part and parcel of those traditions since upanißadic times. A more thorough study of this process of infiltration brings to light fascinating details. It shows, for example, the way in which the so-called å†rama system unites originally different forms of asceticism (Bronkhorst 1998). It also shows how most of what we call brahmanical philosophy is a response to challenges that originated in the ‡ramaña tradition. In other words, what is here called the ‡ramaña tradition did not only give rise to non-Vedic religions such as Buddhism, Jainism and Åjœvikism. It also exerted a lasting and often determining influence on many features that came in due time to be associated with the orthodox brahmanical tradition. In 5 Nor am I denying that there may have been some association with Kßatriyas. See Salomon 1995.

II. Discourse, Conditions and Dynamics of Tradition in South Asia

71

what follows I will concentrate on one such feature, viz. yoga and related issues, against the background of the observations just made. Let us begin with the Yogasütras. They are often described as Pata∞jali’s Yogasütras. This attribution of the Yogasütras to someone called Pata∞jali is common among modern scholars, yet it is not based on reliable evidence. Those who attribute de Yogasütras to Pata∞jali usually ascribe the Yogabhåßya to someone called Vyåsa. This attribution is late, and is contradicted by the earliest extant testimonies. Several authors –among them Devapåla the author of a commentary on the Laugåkßigr¢hyasütra, Våcaspatimi†ra the author of the Nyåyavårttikatåtparya™œkå, and ‡rœdhara the author of the Nyåyakandalœ– cite sentences from the Yogabhåßya and attribute them to Pata∞jali. What is more, the colophons of the combined Yogasütras plus Yogabhåßya do not distinguish between sütras and bhåßya, but call the two together Yoga†åstra; this Yoga†åstra they call påta∞jala, which means ‘the Yoga†åstra of Pata∞jali’. No ancient tradition has preserved the Yogasütras independently of the Yogabhåßya, and a detailed analysis of the text provides us with reasons to believe that the author of the Yogabhåßya brought the Yogasütras together, at least in some cases from different sources, and composed a commentary, the bhåßya, which sometimes demonstrably deviated from the original intention of the sütras. Since I have dealt with these issues in an article that has come out long ago (Bronkhorst 1985), I will not enter into details. However, the same colophons that ascribe the Yoga†åstra –i.e. the Yogasütras plus Yogabhåßya– to Pata∞jali, also describe that Yoga†åstra as ‘expository of Så∫khya’ (så∫khyapravacana). That is to say, these colophons do not describe the Yoga†åstra as presenting a separate philosophy, namely the Yoga philosophy, but as presenting the Så∫khya philosophy. This is not surprising, because an analysis of the theoretical positions taken by the Yogabhåßya shows that they coincide in all essentials with the positions ascribed to the Så∫khya teacher Vindhyavåsin in the Yuktidœpikå, called ‘the most significant commentary on the Så∫khyakårikå’ by its most recent editors (Wezler, Motegi 1998). The idea of a separate yoga philosophy did not yet exist at that time (Bronkhorst 1981). There is therefore no need to search for the early history of the yoga philosophy, for there was none. We can concentrate on the early history of yoga practice. The identification of the theoretical positions taken in the Yogabhåßya as being those of Vindhyavåsin allows us to date this text at least approximately. Vindhyavåsin is known to have lived around the year 400 CE. The Yogabhåßya may date from that time, or from slightly later. We have already seen that the Yogasütras cannot be dated earlier, at least not in the collected form in which we know them. Modern scholars have noted the indebtedness to Buddhism of the yoga practice presented in the Yogasütras since 1900. Émile Senart drew attention to it in an article that was published in that very year. Louis de La Vallée Poussin returned to the topic and explored it fur-

72

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

ther in an article that came out in 1937. I myself have been able to draw attention to some further elements borrowed from Buddhist practice (Bronkhorst 1993: 71 ff). The influence of Buddhism on the Yogasütras is not therefore in doubt. It does however raise a number of serious questions, such as: has yoga practice always been influenced by Buddhism? Is yoga practice nothing but a borrowing from Buddhism, dressed in a slightly adjusted theoretical garb? Do we have to look for the origin of yoga in Buddhism? The answer to all these questions, in my opinion, is: no. The Yogasütras present us with a mixture, part of which is of Buddhist origin, and part of which is not. As a matter of fact, we can study the earlier history of yoga by leaving the Yogasütras for the time being on one side and concentrating on earlier sources. There are plenty of those, among them a number of Upanißads (Ka™ha, ‡vetå†vatara, Maitråyañœya, and others) and of course the Mahåbhårata. The yoga we encounter in these texts is as a rule quite different from that in the Yogasütras. The emphasis is here on motionlessness of body and mind. One passage from the Mahåbhårata should serve as an illustration: Having made his senses firm with his mind, [...] and having made his mind (manas) firm with his intellect (buddhi), he is motionless like a stone (14). He should be without trembling like a pillar, and motionless like a mountain; the wise who know to follow the precepts then call him ‘one engaged in Yoga’ (yukta) (15).6

Suppression of bodily and mental activity, which often includes the suppression of breathing, are a frequent theme in those early texts. This theme can be followed further back in time. Its earliest manifestation in the surviving literature is linked to the Jainas and Åjœvikism; this we know from the Jaina canonical texts of the ‡vetåmbaras (the Digambaras have not left us any texts from the earliest period, and the Åjœvikism no texts at all), but also from Buddhist texts that criticise the Jainas. There is therefore no doubt that motionlessness of body and mind was an ideal which many early ascetics aspired to, and also that this ideal was not confined to just one religious current. The popularity of this ideal should not surprise us. Motionlessness of body and mind is linked to the belief that activity –i.e. motion of body and mind– leads to rebirth and continued suffering. Escape from the cycle of rebirths was hence believed to be possible through the discontinuation of all bodily and mental activity. This conviction could take extreme forms, such as that of seeking death through inactivity at the end of a long process during which one would remain standing, refusing to eat and in the end suppressing the breath, meanwhile keeping one’s mind completely motionless. But extreme or not, it is clear that 6 Mahåbhårata, 12.294.14-15 (sthirœkr¢tyendriyagråma∫ manaså mithile†vara Ù mano buddhyå sthira∫ kr¢två påßåña iva ni†calaΔ ÙÙ14ÙÙ sthåñuvac cåpy akampaΔ syåd girivac cåpi ni†calaΔ Ù budhå vidhividhånaj∞ås tadå yukta∫ pracakßate ÙÙ15ÙÙ).

II. Discourse, Conditions and Dynamics of Tradition in South Asia

73

this conviction is necessarily connected with the belief in karma and rebirth. And this belief, as I pointed out earlier, belonged originally to the ‡ramaña tradition. It is not therefore surprising that our earliest evidence for this kind of practices comes from Jainism, which promulgated these practices not only at the time of Mahåvœra (a contemporary of the Buddha) but already at the time of Pår†va, who according to tradition lived 250 years before him. Having discussed one of the two main historical roots of the practices of the Yogasütras, we now turn to the other one. This, as already indicated, is Buddhism. The question we have to address is: how is Buddhism to be situated with regard to the ‡ramaña tradition on one hand, and the Vedic tradition on the other. The easy answer to this question is that Buddhism arose from the ‡ramaña tradition. The full answer is more complicated, and I will now try to disentangle some of the complications that are relevant in the present context. Buddhism, like Jainism and other currents belonging to the ‡ramaña tradition, is based on the belief in karma and rebirth. However, Buddhism gave a different twist to this belief. Recall that the early Jainas and those others who practised motionlessness of body and mind were convinced that all and any activity would carry undesired consequences. The only effective response was therefore to stop all activities, voluntary or involuntary, conscious or unconscious. Only total suppression of all bodily and mental activities, including in the end even breathing and thinking, could in this way liberate a person from repeated existence. Buddhism, on the other hand, did not share the belief that every single movement carries karmic consequences. No, only activities that were the result of desire had this effect. More precisely, desire is the force that carries a person from one existence to the next. Given this different point of departure, the practices of the early Buddhists could not but be different from those of the Jainas, Åjœvikas and others. It would of course be totally pointless for a Buddhist to practice complete motionlessness of body and mind. He might not physically or mentally act in that case, but his desires would remain unaffected. No, the way the early Buddhists conceived of karma and rebirth entailed that they needed a different practice altogether. Their aim was to eradicate desire and therefore to effect a psychological transformation. Asceticism based on immobility would not bring that about. What they needed was a psychological method. This is what the Buddhist texts contain in the form of succeeding levels of meditation. These are supposed to allow the practitioner to reach ever deeper levels of interiorisation. At the deepest level of interiorisation he is supposedly able to bring about the psychological changes required. He then emerges from his meditation a different person, free from desire and liberated from rebirth and suffering. This is what the Buddha claimed had happened to him, and to all others entitled to be called arhats.

74

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

For our present purposes it is important to see that the mental exercises of Buddhists and most others who continued the ‡ramaña tradition were profoundly different from each other. Both might use the same terms –dhyåna is used both by Buddhists and Jainas, for example– but this does not prove that they did the same thing. Most ascetics belonging to the ‡ramaña tradition tried to suppress all mental activity, which includes consciousness itself. In Buddhism suppression of consciousness was no aim, and could be no aim, for the meditator was supposed to consciously bring about the required psychological changes at his deepest level of interiorisation. The Buddhist meditator, even if he did not succeed in eradicating all desires, cultivated modified states of consciousness; the other meditators of the ‡ramaña tradition tried to suppress consciousness altogether. Historical processes are always messy and irregular. They never present themselves in the kind of pristine purity that would delight the historian. The history of yoga and meditation in India is no different. The distinction which I have just pointed out between Buddhist meditation and the mental practices of the early yogins soon got blurred. The textual sources we have to work with show signs of contamination, if you allow me the expression. It is practically impossible to determine whether these sources originated in circles where yoga and meditation were actually practised. Let us not forget that literary traditions are not normally preserved by practising ascetics. As a result our sources may very well be the products of lineages of teachers and pupils who practised minimally or not at all. Even the Yogabhåßya, as I argued long ago (1985), shows signs that its author may not have had any direct experience of yogic states. The modern study of yogic practice and meditational states in ancient India necessarily passes through a prolonged stage of intense philological study of texts which are on the one hand our only source and which may on the other be far removed from the object of our study. Leaving philological detail aside for the time being, the preceding reflections allow us to conclude that the yoga of the Yogasütras continues a line of practices that were current in the ‡ramaña tradition. These practices originally concentrated on the immobilisation of body and mind, and were intimately and essentially linked to the belief in karma and rebirth. This lineage continued and finds expression in a number of early brahmanical texts, and is still recognisably present in the Yogasütras, as for example in its very first sütra: yoga† cittavr¢ttinirodhaΔ. But another lineage of practices was introduced by Buddhism, based on a different understanding of karma and rebirth. These alternative practices emphasised mental interiorisation, and consequently the search for modified states of consciousness. The Buddhists tried to strictly distinguish their practices from those of the others, but with mitigated success. Mutual influence between the two is discernible from an early date onward, and culminates in the Yogasütras.

II. Discourse, Conditions and Dynamics of Tradition in South Asia

75

It is important to keep in mind that the early tradition of yoga (using the term is risky, because not all early texts employ it) has, in most of its manifestations, no connection with mysticism in the sense of search for modified states of consciousness. This element was introduced by Buddhism, for the reasons indicated earlier. Before the rise of Buddhism, and to a considerable extent also after it, yoga had nothing to do with anything that might be called mystical. That is not to say that there were no people who had mystical experiences; there may always have been such people, in all cultures on earth, including South-Asia. It only means that, in researching the earliest history of yoga, we should not fall in the trap of collecting early indications of what might look like ecstatic states. Yet this is that has often happened. We all know how often early yoga is linked to the sages with long hair (ke†in) mentioned in the R¢gveda, or with the vråtyas. Yet their inclusion in early yoga is based on a fundamental misunderstanding, for early yoga has nothing to do with ecstatic states, not even (until the arrival of Buddhism) with ‘enstatic’ states. An equally serious misunderstanding, which still comes up from time to time in careless publications, finds expression in the point of view that Buddhist practice owed much, if not all, to yoga. This misunderstanding dates from the time when the chronological relationships between various texts was a lot less clear than it is now. There are indeed elements in the Yogasütras which we also find back in the early Buddhist texts. The Yogastüras stand however at the end of the long tradition during which Buddhist elements entered into the yogic tradition, not vice-versa. The kind of yoga that existed at the time of early Buddhism (it is not clear whether the term yoga was already used at that time) was firmly rejected by the latter, and replaced by something altogether different. A further source of confusion has been the fact that the Vedic tradition, too, knew ascetic practices in connection with its rituals. These are however to be understood in their sacrificial context, and have nothing whatsoever to do with the belief in karma and rebirth. Here, too, the later tradition made ever fewer distinctions between the Vedic form of asceticism and the ‡ramañic one, and ended up confounding them completely. Once again the historian is confronted with texts and traditions that are contaminated to different degrees. Yet the early texts distinguish clearly between the two forms of asceticism, and it is clear that the Vedic sacrifice offers no help in tracing the origins of yoga (Bronkhorst 1998). The preceding reflections have illustrated that it is a mistake to look for the origin of everything Indian in the Veda. The Vedic tradition is extraordinarily reliable in the way it has preserved the Vedic texts. The accompanying claim that the Veda is the origin of everything, on the other hand, is not reliable at all, and is in many cases demonstrably wrong. Indologists should take heed.

76

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

BIBLIOGRAPHY Brockington, J. (1998), The Sanskrit Epics, (Handbuch der Orientalistik, Abt. Indien, Bd. 12), E.J. Brill, Leiden. Bronkhorst, J. (1981), Yoga and se†vara Så∫khya, in «Journal of Indian Philosophy», vol. 9, pp. 309-320. ______ (1985), Pata∞jali and the Yoga sütras, in «Studien zur Indologie und Iranistik», vol. 10, (1984 [1985]), pp. 191-212. ______ (19932), The Two Traditions of Meditation in Ancient India, Motilal Banarsidass, Delhi. ______ (19982), The Two Sources of Indian Asceticism, Motilal Banarsidass, Dehli. ______ (2004), Sylvain Lévi et les origines du théâtre indien, in «Asiatische Studien / Études Asiatiques», vol. 57, (2003 [2004]), n. 4, pp. 55-73. Bryant, E. (2001), The Quest for the Origins of Vedic Culture. The IndoAryan migration debate, Oxford University Press, Oxford. Erdosy, G. (ed.)(1995), The Indo-Aryans of Ancient South Asia. Language, Material Culture and Ethnicity, (Indian Philology and South Asian Studies, 1), Walter de Gruyter, Berlin-New York. Hobsbawm, E., Ranger, T. (ed.)(1983), The Invention of Tradition, Cambridge University Press, Cambridge. La Vallée Poussin, L. de (1937), Le bouddhisme et le yoga de Pata∞jali, in «Mélanges chinois et bouddhiques», vol. 5, (1936-1937), pp. 223-242. Mller, M. (1869), Rig-Veda-Pratisakhya, das älteste Lehrbuch der vedischen Phonetik, F. A. Brockhaus, Leipzig. Pande, G.C. (19833), Studies in the Origins of Buddhism, Motilal Banarsidass, Delhi. Parpola, A. (1988), The coming of the Aryans to Iran and India and the cultural and ethnic identity of the Dåsas, in «Studia Orientalia», vol. 64, pp. 195-302. Salomon, R. (1995), On drawing socio-linguistic distinctions in Old IndoAryan: The question of Kßatriya Sanskrit and related problems, in Erdosy, G. (ed.), pp. 293-306. Senart, É. (1900), Bouddhisme et yoga, in «Revue de l’Histoire des Religions», vol. 42, pp. 345-364. Wezler, A., Motegi, S. (ed.)(1998), Yuktidœpikå, the most significant commentary on the Så∫khyakårikå, (Alt- und Neu-Indische Studien, 44), Franz Steiner, Stuttgart. Witzel, M. (1995), Early Indian history: Linguistic and textual parameters, in Erdosy, G. (ed.), pp. 85-125. Witzel, M. (1995a), R¢gvedic history: poets, chieftains and politics, in Erdosy, G. (ed.), pp. 307-352.

TIMOTHY LUBIN The Transmission, Patronage, and Prestige of Brahmanical Piety from the Mauryas to the Guptas*

1. Introduction A†oka’s reign ushered in an era of major transformations in India’s religious life. The 6th to 4th centuries BCE had witnessed, in the plains around the Ganges and Jumna rivers, the emergence of the first urban centers in South Asia since the Indus Valley Civilization. While the precise reasons and mechanisms for this development are not fully understood, it has long been taken for granted that it is not mere coincidence that it was accompanied by equally dramatic changes in the sphere of religion –especially the rise to prominence of ascetical movements. Although the precise origins of such movements are unknown, they appear on the scene in two species: one emerging from within the Brahmanical priestly religion that was already more than half a millennium old, and another comprising self-consciously anti-Brahmanical doctrines traced back to legendary teachers whose authority resided in their personal accomplishments (Gotama the Buddha, the ‘Enlightened’; Mahåvœra the Jina, the ‘Victor’). As this story is generally told, it is the Buddhists and the Jains who are in the limelight, bringing new ideas to an old world, challenging the cant and rigid orthodoxy of the brahmin priesthood, just as the early Christians were said to have exposed the hypocrisy * This research was begun with an American Institute of Indian Studies fellowship funded by the National Endowment for the Humanities; further work was done during tenure of a Fulbright-Hays fellowship and an N.E.H. fellowship. The author also acknowledges a small grants from the American Academy of Religion and Washington and Lee University. Earlier versions of part of this material were presented in a workshop at the Institute of Social and Cultural Anthropology, Oxford University in 2001, and at the 31st Conference on South Asia in Madison, Wisconsin in 2002. The comments of Himanshu Ray on the former occasion, and of Tom Trautmann and David Lorenzen on the latter, were of great help.

78

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

and superficiality of Pharisaic piety. And rather as the new Christian vision was embraced and established in the Roman Empire by Constantine, the emperor A†oka Maurya proclaimed the Buddha’s dharma to be the religion of this new world. Peripatetic mendicants (Buddhist, Jain, or other), having renounced the settled life of the village, rubbed shoulders with the traders and soldiers with whom they shared the highways. Their doctrines, dismissive of Brahmanical caste strictures and home- and caste-centered ceremonial obligations that would have been impractical for people who traveled frequently and who had to mix with strangers and transients in the towns, offered an alternative set of ideals vividly exemplified in the person of the mendicant himself. Veneration of mendicants, and by extension of their legendary models, Buddhas and Jinas, seemed to be a mode of piety congenial to the new cosmopolitan world. Accordingly, Buddhicized traders have been credited with carrying Indian culture beyond the Subcontinent, into Central and Southeast Asia. Yet the Brahmanical tradition did not wither away. Sanskrit learning, the province of the traditional brahmin, entered an innovative phase. Works composed in this period and the immediately succeeding centuries –Påñini’s grammar, along with the commentary of Pata∞jali; the core of the Mahåbhårata epic– became the classics of later generations. By the early centuries of the Common Era, Sanskrit ceased simply to be a liturgical language of the Brahmin priest; it came to be widely accepted as the most impressive vehicle of refined discourse in the court and in the academy, the language of belles-lettres as well as of formal state pronouncements and legal records (at least those important enough to merit engraving on copper or stone).1 It begins to appear in royal inscriptions (especially in their ornamental portions), and is adopted even by the initially antiBrahmanical Buddhists and Jains. This adoption of what was essentially the cultural property of the brahmins, indeed, the liturgical language of a hereditary priesthood,2 is powerful testimony to the effectiveness of the brahmins’ methods of transmitting their texts and practices, and of making them appealing and authoritative to others. This essay examines the mechanisms by which Brahmanical tradition reproduced itself,3 especially the regimens of discipline 1 The broad outlines of this process have been sketched by Aklujkar 1996; Pollock 1996; 1995. 2 The corpus of old Sanskrit texts is made up overwhelmingly of works belonging to the Vedic cult, or explicitly ancillary to it. Even if the epics contain a ‘core’ with roots prior to the Maurya period, the forms in which we possess them are surely post-Maurya, and in any case exhibit the clear traits of brahmin redaction (any Brahmanical doctrinal views), if not original authorship. 3 By ‘Brahmanical tradition’ I mean those forms of religion and social doctrine that were defined by brahmin authorities, generally in Sanskrit and generally represented as derived from and in accordance with the Veda or the pronouncements of the r¢ßis. The early ‡aiva and Vaißñava religions (e.g., the På†upata and the På∞caråtra), insofar as they were shaped by brahmin authority, belong to this tradition.

II. Discourse, Conditions and Dynamics of Tradition in South Asia

79

(vratas) undertaken in tandem with text-study, and their role in establishing the knowledge of Sanskrit religious texts (and the use of Sanskrit more broadly) as an important criterion of piety and high social status. I argue that such regimens functioned as markers of belonging to the Brahmanical religion and ‘pure’ Årya society, while also offering the ordinary householder a form of personal piety that promised all the rewards of the old priestly ‘high cult’. At the same time, disciplinary regimens provided a traditionally recognized framework for mendicant movements and new deity cults, which helped carry Brahmanical texts, ideals, and practices, via royal patronage, into new regions in India beyond the Ganges Valley and on into Southeast Asia. The second part of this essay will consider what early inscriptions can show us about how Brahmanical doctrine and practice were projected in the public sphere, noting instances in which particular subjects, texts, and especially disciplinary practices are cited, and observing that the grants and foundations recorded in these inscriptions helped spread the tradition and enhance its prestige. My remarks, intended only as a point of departure, will focus mainly on early grants from Orissa. 2. Modes of Transmission The Brahmanical tradition, although it produced a mendicant ideology and practice, remained rooted in the rural setting. Whereas the anecdotes and parables of the Buddha in older accounts like those collected in the Suttanipåta are commonly set in the new cities (nagara),4 the Vedic literature up to the time of the dharmasütras (3rd—1st c. BCE) hardly mentions them;5 the opposition between civilized sphere and its opposite is expressed in terms of gråma versus arañya –the rural village settlement versus the wild regions. When Brahmanical texts deign 4 Biographical information extractable from the songs of the Theragåthå and Therœgåthå has been used to argue that early monks and nuns came mainly from the cities, and from the brahmin and other high castes (Gokhale 1965 [quoted in Ray 1994: 124]). As with all social and political data extracted from Påli literature, it cannot simply be accepted as representative of Mauryan-era realities (this is especially true for the much later commentary, on which Gokhale also relied). Moreover, we might expect the words attributed (correctly or not) to mendicants from socially prominent families to be better represented in the literature than others, so the social position of these figures can hardly be taken as indicative of broader demographic patterns in the early saõgha. Moreover, there may have been a conscious impulse to appropriate the prestige of the brahmins by representing them as converts to Buddhism (a theme found throughout the literature). Even so, the prominence of the cities as the backdrop to Buddhism’s earliest surviving selfdepictions suggests that the tradition saw them as its natural field of action. 5 The word nagara is not even found until the Jaiminœyabråhmaña (1.11[.1], 1.247[.9], 1.257[.12], 2.397[.8], 2.410[.6], 2.424[.1]) and Jaiminœyopanißadbråhmaña (3.40.1), where it is the basis of the personal name Nagarin. In the Baudhåyanakarmåntasütra (Baudhåyana†rautasütra 24.14) nagara is mentioned parallel to gråma. The sole occurrence of the term in the gr¢hyasütras is at Månavagr¢hyasütra, 2.14.28 where the crossroads of a village, town (nagara), or market (nigama) is prescribed as the site for an expiatory offering to exorcise Vinåyaka spirits that have possessed someone. As in the related

80

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

to take account of the urban environment (only in the last couple of centuries BCE), they regard the towns as posing severe difficulties to pious observance. For instance, the Åpastambadharmasütra deems cities and trade centers as inappropriate for the study of the Veda: ‘He [who is a Veda-reciter] should avoid entering towns’ (Åpastambadharmasütra, 1.11.32.21); ‘He should refrain from reciting the Veda in market-towns (nigameßu)’ (Åpastambadharmasütra, 1.3.9.4; similarly, Vasiß™hadharmasütra, 13.11 [‘in towns’, nagareßu]; Gautamadharmasütra, 2.7.45 [16.45] mentions this as the view of ‘some’ authorities). The dharmasütras class towns with casinos, fairs, cemeteries, and neighborhoods of cañ∂ålas.6 Ritualist brahmins do not appear to have established monastic or scholarly centers comparable to those of the Buddhists.7 What institutions did brahmin priests and scholars develop that allowed them to carry on and eventually to attain equal success in many of the domains where Buddhism was successful? The Vedic literature, a vast canon of liturgical and exegetical texts scrupulously preserved through rote memorization, is the earliest extant product of Brahmanical culture. Although these texts hardly ever contain references to datable historical events, a fairly persuasive relative chronology (at least for certain classes of texts, and within such classes, for some exemplars) has been proposed.8 The distribution in the texts of such data as place names has helped yield a reconstruction of the phases of settlement of Sanskrit-users (calling themselves Åryas) in South Asia, from the R¢gveda (centered mainly on the Panjab) to the Upanißads and ritual texts, which increasingly speak of the Ganges valley and other regions to the south.9 Overall, this literature concerns a complex sacrificial system that depended on reciprocal bonds between tribal chieftains (and other Traiyambaka homa for Rudra and Ambikå (who also receive offerings in the Vinåyakakalpa), the crossroads is the place where inauspicious forces are addressed (Baudhåyana†rautasütra, 5.16-17). One difference in Månava is the specific mention of urban crossroads as a potential offering-place. 6 On the other hand, in discussions of royal policy, the råjan and the †üdras are both enjoined to protect towns as well as villages (Åpastambadharmasütra, 2.10.26.4-6); similarly, Vasiß™hadharmasütra, 16.15 treats both locales on a par. 7 A†oka speaks of ‘giving to brahmins and monks’ (båhmañasamañåna∫ [...] dåne in the Girnar version of the 8th Rock Inscription) as his new royal policy, and he frequently praises generosity and respect toward these groups as a virtue. They are always mentioned in the same breath, and in either order indiscriminately, so that one has the impression that he did not make much distinction between them. We cannot even know for sure to whom these labels applied: did bråhmaña signify a caste member, a professional priest, or a mendicant in the Upanißadic mould? Did †ramaña cover the latter, as well as the Buddhist and the Jain monk? But when particular acts of patronage are mentioned, it is almost always the Buddhists who are the beneficiaries. 8 For the earlier strata of the Vedic literature, see Witzel 1997a; for the dharmasütras see Olivelle 2000: 4-10. 9 Witzel (1987; 1989) speaks of the ‘widening of the geographical horizon’ in the later Vedic texts (1987: 203); in spite of the deliberate insistance that the lands beyond the Madhyade†a were unfit for brahmins, the prescription of rites of atonement for those who went abroad is in effect an admission that many did.

II. Discourse, Conditions and Dynamics of Tradition in South Asia

81

patrons of the Vedic cult) and the brahmin priests. The priests conferred power and prestige on their patrons, through their ritual deployment of mantras believed to invoke divine aid and other benefits. According to their own accounts, the brahmins’ status was predicated upon their hereditary connection with the mystical source of divine knowledge, bráhman. Yet even Brahmanical texts acknowledge that, although it was conventionally so recognized, the hereditary connection alone was not really sufficient: scorn was heaped by brahmin and Buddhist authors alike on the ‘one [merely] related to Brahmins’ (brahmabandhu),10 or ‘brahmin by birth [alone]’ (jåtibråhmaña).11 Bráhman –and true Brahmin status– had to be transmitted through a ‘second birth’ in the form of an initiation rite, known in the literature as upanayana, induction or introduction into the rule of brahmacarya, under the guidance of a master. This rule of conduct, to be adhered to scrupulously until the end of the period of study, is both a transforming rite de passage and an apprenticeship in priestcraft or scholarship. If we contrast the institutional structures of the brahmins and the Buddhists, we must begin with the fundamental differences between the professional representative of each tradition. The highest Buddhist ideal was embodied in the monk, the imitator of the Buddha. Initiation into the sa∫gha meant a break with the norms of secular society and a radical, permanent separation, with only ritualized and formal relations with the laity. In place of the old social linkages, the mendicant (bhikßu) entered into a trans-local community, a network of collectives of mobile individuals. Considerations of family and place of origin were devalued (in principle at least). As far as the lay community was concerned, one monk was basically like any other (except perhaps in the case of charismatic eremites) (Ray 1994). Buddhist ideals were spread far and wide in the exemplary person of the monks themselves, while a canon of texts was compiled –according to traditional accounts, it was established in great synods– and in perhaps the last couple of centuries BCE began to circulate physically in the form of manuscripts. Although divergent schools of thought emerged, their exponents mingled in great, cosmopolitan monastic ‘universities’. By contrast, Brahmanical institutions were diffuse and intensely localized, at least prior to the early dharma-texts. Canonical texts and practices belonged to and defined individual lineages, each consisting of a particular division (caraña) of a particular branch (†åkhå) of a particular priestly office that was a matter of heredity. Initiation into study meant a virtual adoption by an individual teacher. The diffusion of textual knowledge was dependent upon teacher-to-pupil lineages and texts themselves were treated as belonging to individual descent groups until Mauryan times at least. Thereafter, a pan-Årya scholastic and literary tradition began to take shape (although the core Vedic 10 11

Aitareyabråhmaña, 7.27; Chåndogyopanißad, 6.1.1. See, also, Suttanipåta, 2.7, v. 312. For the same sentiment in other terms, see, e.g., Baudhåyanadharmasütra, 1.1.10.

82

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

texts continued to be treated as proprietary knowledge of individual lineages). The brahmin graduate could become a professional priest or scholar. As such he would depend for his livelihood on patronage that (like the jajmånœ relations of India today) might become a longterm, even multigenerational, relationship. Some few were able to rise to positions of prestige and authority in the royal court (as purohita or in a wide variety of other offices); the most famous such figure was Kau™ilya, the legendary minister and advisor to Candragupta Maurya. Brahmins did gather to meet in assemblies called parißad, sa∫sad, or sabhå to decide questions of ritual or social dharma, and to serve as a local court of law. But the development of durable, large-scale Brahmanical institutions lagged behind that of Buddhist monasteries. When it came, it took the form of brahmin settlements on endowed, tax-free lands (agrahåras) and royally sponsored temples. 3. The Priestly Codes While the high Vedic cult was being systematized, the simpler offerings that had long been made by a householder in a single fire –hence, the ‘domestic rites’ (gr¢hyåni karmåñi), or ‘simple worship’ or ‘worship with cooked food’ (påkayaj∞aΔ)– continued to be governed only by custom. However, sometime during this period, the various Vedic schools began to compile codes for the gr¢hya rites, by analogy with the †rautasütras. Beginning with the gr¢hyasütras, and continuing in the dharmasütras, the earliest formulations of a Brahmanical ‘dharma’ ostensibly based on the Veda, we find a progressive effort to establish Vedic knowledge and a simplified Vedic practice as the basis of a unified trans-regional Årya culture. At the same time, the ritual modes for transmitting Vedic knowledge were adapted for use by sectarian movements that helped spread Brahmanical religious culture throughout South Asia and deep into Southeast Asia. The sütra-texts are extremely difficult to contextualize. The sequence of genres in the ritual code literature is clear enough (†rautasütras → gr¢hyasütras → dharmasütras); although the spans of time in which works in each of these genres were composed and redacted to take their current forms overlap considerably, there is no denying that there were no gr¢hyasütras until the †rautasütra genre was already well known, since they explicitly presuppose the existence of the †rautasütras. Olivelle has proposed some concrete (if not completely decisive) criteria for dating the four old dharmasütras to the period 3rd-1st centuries BCE.12 The oldest strata of the gr¢hyasütra literature likely took 12 Olivelle (2000), leaving out of account the Vaikhånasadharmasütra, a much later work. More recently, he has tentatively suggested a slightly later, post-Maurya daterange based on the possibility that the Brahmanical adoption of ‘dharma’ as the overarching principle was influenced by A†oka’s use of the concept in his royal policy (see Olivelle forthcoming).

II. Discourse, Conditions and Dynamics of Tradition in South Asia

83

shape during the preceding two or more centuries, that is, to the 5th (or 6th?) centuries through the 4th –in other words, to the period precisely during which the trade centers in the Ganges Valley were emerging and Buddhism was taking shape. In this period, Sanskrit texts were apparently being produced mainly in the north, and therefore in the urbanizing sphere. The gr¢hyasütras themselves rarely mention specific place names, but do note the diversity of domestic practices from place to place. One respect in which the intention of the authors of the gr¢hyasütras is clear is in their concern to bring to household practice some semblance of the consistency (within each Vedic school) of the †rauta cult.13 The comment near the beginning of the ņvalåyanagr¢hyasütra is well known: ‘Now various indeed are the dharmas of the (different) countries, and the dharmas of the (different) villages; one should observe them in the wedding. But we shall state what is common (to all)’ (ņvalåyanagr¢hyasütra, 1.7.1-2 [atha khalüccåvacå janapadadharmå gråmadharmå† ca tån vivåhe pratœyåt yat tu samåna∫ tad vakßyåmaΔ]).14 4. The Marketing of Vedic Knowledge Domestic (gr¢hya) ritual had always maintained an independent character vis-à-vis the high, vaitånika cult of the †rauta priests, shaped by custom rather than a canonical rubric. Leaving aside the important question of what circumstances led to the promulgation of †rauta codes that prescribe the proper forms of the complex multi-fire rites, why might the priestly authors have considered it necessary, in a secondary stage, to codify the ceremonies outside the †rauta system as well? Let us first consider what little we know about the historical context.15 In the 6th to 4th centuries BCE north India witnessed the dramatic growth of trading towns along rivers and, perhaps, near sources of iron ore. These towns generated wealth that made it possible for local rulers to consolidate larger realms; at the same time, the socially plural and occupationally more specialized cities required new socio-economic institutions (such as market-centers [nigama] and 13 Gonda shows how the Yajurveda †rautasütras broadly agree on the order in which the rites are treated (Gonda 1977: 494). 14 Particular variant practices are cited in ss.15-16; see Påraskaragr¢hyasütra, 1.8.11 (authorizing local wedding traditions); Baudhåyanadharmasütra, 1.2.1-8 for more extensive observations about regional differences and their controversial status. Vasiß™hadharmasütra, 19.7-8 specifies that it is the duty of a king to ascertain ‘all the dharmas specific to various regions, castes, and families’, and to enforce them. 15 The following sketch sums up the picture provided by Erdosy (1988; 1995a; 1995b), although the dates he accepts for the Sanskrit texts he mentions are probably a bit too early. In particular, it has been shown (Trautmann 1971; Fussman 1987-1988) that the Artha†åstra cannot be treated as a record of the Maurya era, as Erdosy and Allchin (1995) are inclined to do. They also are too confident in regarding Pali texts as accurate depictions of the situation in A†oka’s time.

84

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

guilds [†reñi]) and political structures. Meanwhile, the Brahmanical system was dependent upon the stable caste society of the village, and may have had difficulty adapting to the urban setting, where people of different regions mixed. If the village patronage networks were disrupted by the growth of the urban polities, and if, as may more confidently be asserted, the Brahmanical priestly order was slow to adjust to the culture of the towns, the gr¢hyasütras might contain some hints of a response to these developments, even though the developments themselves are barely alluded to at all. It may well be that the composition of rules for domestic rites began before the socio-economic changes had made a great impact on village life. Still, there are signs that the genre came to embody the Zeitgeist of the period. A striking indication of this doctrinal shift is the often-expressed view that gr¢hya rites were equivalent or superior to †rauta rites, and not simply pale shadows of them. Such arguments pick up themes heard also in the mystical reflections of the Årañyakas and Upanißads: the idea that all sins could be expunged by reciting a litany in the wilderness to the accompaniment of a series of ghee offerings in a single fire (Taittirœyårañyaka, 2), or that all the rewards of a pious †rautin life could be secured through the regular performance of a few simple ‘super-sacrifices’ (mahåyaj∞as). Claims for the sufficiency of mantra-recitation as a form of worship in itself paved the way for the gr¢hyasütras’ codification of a variety of regimens consisting of ascetic discipline, recitation, and perhaps simple homas to expiate sins and to fulfill wishes. In any case, we find in the gr¢hyasütras several apparent novelties that are best understood in relation to one another (Lubin forthcoming). The first of these is the explicit statement that Veda study is not merely permissible for kßatriyas and vai†yas as well as brahmins, but that it is duty of all three upper estates of society: that recourse to the Veda marks them as Årya. 5. Brahmacarya and Årya Status The study of the Veda was a sacramental activity governed by a regimen (vrata) of brahmacarya. A student must be accepted by a master whom he will serve obediently and with whom he will reside for (it is said) up to twelve years or more. During this period, he must remain sexually chaste, eat only food gathered as alms (which he in turn offers first to his master), tend the master’s fire unfailingly, and serve him obediently. This regimen and the quite parallel consecration (dœkßå) undertaken by the sponsor of a Soma sacrifice, are of great antiquity. The earliest references to the Veda student (at R¢gveda, 10.109.5, and in the Atharvan hymns, ‡aunakasa∫hitå, 11.5 and Paippaladasa∫hitå, 16.5) allude to elements of the rule of brahmacarya, and certain features of the initiation rite are discussed in the ‡atapathabråhmaña (11.5.3.13-11.5.4.18). But the formal codification

II. Discourse, Conditions and Dynamics of Tradition in South Asia

85

in the gr¢hyasütras of the rules for initiation (upanayana) into this vrata exhibits a number of peculiarities. First there is the fact that although provisions are made for the initiation of kßatriyas and vai†yas, at many points (and in some gr¢hyasütras, throughout the discussion) the authors otherwise seem to assume that the initiate will be a brahmin.16 Of course, the class-name ‘bråhmaña’ indicates a special connection with brahman (the Vedic logos), and may once have been applied properly to those learned in brahman, or who had completed the period of brahmacarya (apprenticeship in recitation and priestcraft), only later being viewed as denoting a hereditary status. For example, the ‡atapathabråhmaña declares that, by means of the upanayana, ‘after three nights, he is born a brahmin’. On the other hand, the (brahmin) authors of the gr¢hyasütras are at pains to show that the fact that initiation is allowed and enjoined for the two middle social classes does not efface the distinctions between them and the brahmins. Hence a series of regular distinctions is introduced, often in a virtually parenthetical way. In most of these cases, the brahmin option is the one that in other contexts is the single standard. For instance, when class-distinctions are made in the gr¢hyasütras, brahmin initiates are assigned the hide of a black antelope (aineya) to wear; in other sources, where no distinctions are mentioned, this is the standard hide. Variations in the rules on the appropriate age for initiation indicate one of the developments in this process. From the dharmasütras onward, the standard formula applied is that brahmins should be initiated at eight, but not later than sixteen, years of age, kßatriyas between eleven and twenty-two, and vai†yas between twelve and twenty-four.17 As it happens, we can see that this pattern was introduced, hesitantly, in the gr¢hyasütras, where it is invoked inconsistently, and only in some schools.18 On the other hand, the code-makers had to contend with customary practice in prescribing a new univer16 Older references to brahmacarya presume a brahmin student, and there are only scattered allusions to non-brahmin Veda scholars (usually kßatriyas). 17 The lower limits correspond to the three most common meters in the R¢gveda hymns (the gåyatrœ, triß™ubh, and jagatœ); the upper limits are the lower ones doubled. Brahmins are thus expected to begin at the earliest age, which for a class of future priests and teachers may seem sensible enough. 18 Among these, Månava does not make distinctions of class by age at all. Four gr¢hyasütras (Kå™haka, Månava, Hirañyake†in, Jaimini) prescribe age seven for the brahmin, rather than eight. Though most of the gr¢hyasütras recognize some sort of class distinctions, but Kå™haka gives 7, 9, 11 as the ages; Hirañyake†in and Jaimini give 7, 11, 12 (Jaimini offering other options for the brahmin); Vaikhånasa gives a variety of options (for a brahmin student only), and age eight is not presented as standard. Even some of those who prescribe age eight for the brahmin, namely ‡åõkhåyana, Kaußœtaki, Våråha, Bhåradvåja, allow options for the brahmin; Baudhåyana, Påraskara allow options for any of the ages. In short, the 8, 11, 12 pattern is only partially adopted, with more consistency in regard to kßatriyas and vai†yas, doubtless because for initiates of these estates there was no separate custom already established that had to be accommodated in the new system. See Lubin (forthcoming) for more details, including a table. A monograph-length treatment of the question is in preparation.

86

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

sal standard for initiation of the children of brahmins, particularly what appears to be a preference in some schools for the age seven. For this purpose a peculiar accounting device was invoked to help smooth over the disruption caused by the new age-eight standard. In this context alone, several of the gr¢hyasütras that assign age eight to the brahmin choose to calculate the age beginning with conception (garbhåß™ame) rather than with birth. In this manner, one could technically fulfill the age-eight requirement while continuing to initiate brahmin boys seven years after birth. The metrically inspired age differentiation, like all the other differentiations by caste, seems to have been introduced in tandem with the notion that initiation should be standard for all three of the higher classes, an idea nowhere asserted in earlier strata of the literature. The implications of this rule are profound: even if (as is likely) longterm Vedic study remained primarily a brahmin endeavor, an ideal was projected of Årya society as unified by a common devotion to Vedic learning and an increasingly homogenized domestic ceremonial system. Marginal groups who wished to be recognized as Årya could ‘prove’ their belonging by being accepted as Vedic students and by Sanskritizing their religious practice. Meanwhile, a radical simplification of Vedic ritual duties, which began in the årañyaka-stratum of the exegetical literature, is rubricated in the gr¢hyasütras as well: the doctrine of the ‘five great sacrifices’ (including recitation as an offering to the sages). 19 The importance of recitation of the Veda as the marker of Årya status is implied in the rule that brahmins ‘should not officiate in a sacrifice offered by one who does not recite the Veda’ (ayåjyo ’nadhœyånaΔ [Åpastambadharmasütra, 2.10.9]). One other piece of evidence that the vrata was being promoted as the preeminent marker of religious identity is the gr¢hyasütras’ inconsistency in ordering the sequence of rites in the life-cycle of an observant Årya. These rites, the sa∫skåras in the narrow sense of the word,20 are always presented in chronological order according to the natural criterion of age. However, a chicken-or-egg dilemma arises here: Does the cycle begin with the marriage, the sa∫skåra that institutes a Vedic household (gr¢ha) and obligates the head of household to begin a lifelong series of rites? Or does it begin with the start of Veda-study?21 19 ‡atapathabråhmaña, 11.5.6.1-10; Taittirœyårañyaka, 2.10, 14; ņvalayanagr¢hyasütra, 3.2.2; Påraskaragr¢hyasütra, 2.9.2-16; Baudhåyanagr¢hyasütra, 2.9.6, 14ff.; Vaikhånasagr¢hyasütra, 3.17, 4.17; Gautamadharmasütra, 1.5.3, 8; 1.8.7; Åpastambadharmasütra, 1.12.14-1.13.1; Baudhåyanadharmasütra, 2.6.1; Månavadharma†åstra, 3.81, 87-89; 4.21; Vißñusmr¢ti, 59.20 ff.; Kürmapuråña, 2.18. 20 Later works, beginning perhaps with the Vaikhånasagr¢hyasütra, expand this category to cover all the gr¢hya rites. 21 This difference was noted by Kane 1974: 195; Olivelle (1993: 126-127) spelled out some of the implications of this shift by noting that the doctrine of the sequence of å†ramas necessitated moving the upanayana to first position in the sequence of sa∫skåras. However, the change of sequence, even if it is presupposed by the classical å†rama system,

II. Discourse, Conditions and Dynamics of Tradition in South Asia

87

Most gr¢hyasütras begin, often after providing general rules for gr¢hya offerings, with marriage as the sa∫skåra that creates an autonomous ritual agent (the gr¢hapati, with his wife).22 The initiation and brahmacarya come in relation to the upbringing of sons. Yet in a few sütras (viz., the Bhåradvåja, Hirañyake†in, Ågnive†ya, Jaiminœya), the initiation is placed first, in accordance with the novel notion that it is the indispensable prerequisite for belonging to Årya society and, by extension, the prerequisite for a Vedic marriage. Other gr¢hyasütras, like the Månava and Kå™haka, seem to represent an intermediate situation, in which the rules of brahmacarya are presented first, while the rite of initiation (upanayana) remains lodged in the sequence of childhood rites! The impression that the initiation-first format is the later one is reinforced by the fact that all the dharmasütras, and all the later dharma-literature, adopt this pattern. Thus, according to the older scheme, the ‘childhood sa∫skåras’ were introduced as the responsibility of the householder to perform upon his offspring, and they thus followed logically and temporally from the marriage. In the new standard scheme, initiation into Veda study was set apart from childhood rites and marked the start of the process, becoming the prerequisite for marriage; the subject of the initiation shifted from being the child of the ritual agent to being the nascent ritual agent himself. Since initiation with a teacher had even in the earliest references to the subject been described as a rebirth in a more perfect and holy state, later texts refer to these classes as the ‘twice-born’ (dvija). However, this term (or its variants dvijåti and dvijanman) does not begin to be used with the meaning ‘a member of one of the three higher varñas’ until after the doctrine of obligatory study for all three upper varñas had been advanced in the gr¢hyasütras. Prior to this time, it was used (when it was used at all) to refer to a brahmin.23 However, in the Gautamadharmasütra (2.1 [10]) we find the doctrine fully fledged in the discussion of the legitimate occupations of the four appears to have been made even before the å†rama debate (which is completely absent in the gr¢hyasütras) had begun. Rather it was the rise in importance of the observance of vratas (which become the distinctive feature of three of the four å†ramas) that was reflected in the change; the å†rama system was a further development. 22 Exceptional are the Våråha, which begins with the birth rite (jåtakarma), leaving the prebirth rites for treatment after the marriage, and the Jaiminœya, which begins with the rite for begetting a male child (pu∫savana). 23 I have not encountered the word dvija in this meaning in any ‡rautasütra, nor does it occur in any of the gr¢hyasütras except in the following few cases: Baudhåyanagr¢hyasütra, 2.9.14 = Bhåradvåjagr¢hyasütra, 3.15.3 (v. 7) = Ågnive†yagr¢hyasütra, 2.6.5 (v. 7), one of a group of stanzas found almost identically in these three works, perhaps a late addition to them; other occurrences of dvija/dvijåti –in Baudhåyanagr¢hyaparibhåßåsütra 1.1.24, 27; Baudhåyanagr¢hya†eßasütra, 1.12.12; 1.21.2, 4; 1.22.5; 3.21.1; 4.4.3; 4.17.19; 5.2.10; Påraskaragr¢hyasütra, pari†iß™a†aucasütra 2; Ågnive†yagr¢hyasütra, 1.3.3 (six times); 1.3.5; Vaikhånasagr¢hyasütra, 4.12; 5.8 (twice); 6.7– are probably no earlier than the dharmasütras, given that they occur either in obviously late works (such as the Ågnive†ya and the Vaikhånasa) or in later sections of the sütras (viz., pari†iß™as of Baudhåyana and Påraskara). Moreover, even in many of these cases, the word must be understood as a synonym for

88

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

estates, where dvija is a virtual synonym for årya, and †üdra for anårya.24 On the other hand, we are justified in wondering whether, in the context of a group of stanzas describing the capacity of learned brahmins to establish dharma in an assembly, the word dvija in Baudhåyanadharmasütra, 1.1.13 could possibly be meant to include kßatriyas and vai†yas. To sum up, the trends discernable in this ritual literature are the following. In the gr¢hyasütras, the claim is first made that study of the Veda is not merely available to but incumbent on kßatriyas and vai†yas as well as brahmins, with the corollary that initiation and the daily use of Vedic mantras become the defining mark of elite, Årya status in a religiously and ethnically diverse society. The trend toward identifying initiation and brahmacarya (rather than marriage) as the starting point for constructing a framework for an orthoprax life of piety, and the multiplication of similar vratas as a framework for personal piety, were developments parallel with the rise of ascetical (†ramaña) movements such as Buddhism. The priestly canonization of household (gr¢hya) ritual, with an accompanying emphasis on trans-regional standardization and the promotion of simplified forms of observance, made the prestige and alleged power of the Vedic cult accessible to a wider range of social and economic statuses. The most prominent instance of a simplified ritual format was the pa∞ca mahåyaj∞åΔ. The legitimacy of this model was provided by expositions of it in the årañyakas; the gr¢hyasütras meanwhile inserted it into the ritual cycle. The prominence of this model is demonstrated by several copper-plate inscriptions recording the endowments or land-purchases intended to support the performance of the five mahåyaj∞as. For instance, in a fifth-century record from Bengal, we find the following petition (vij∞åpita): ‘Please give, according to the acknowledge (anuvr¢tta) law ([dharma?-]maryådå) of the permanent endowments (akßayanœvœ) of unproductive [land] (aprada, scil. kßetra), in order to promote the performance of my five Great Offerings’.25 This and similar records show that land grants ‘brahmin’ rather than as a general label for the three upper varñas. (The word dvija in Baudhåyanagr¢hya†eßasütra, 4.8.1 means ‘bird’) As regards the dharmasütras, Åpastamba does not use the word at all; dvijåti occurs five times in Baudhåyanadharmasütra, four times in Gautamadharmasütra; Vasiß™hadharmasütra has dvijåti once and dvija 12 times. Vaikhånasadharmasütra (=Vißñusmr¢ti, 8-10) has dvijåti twice, dvija thrice, and dvijanman twice. The term becomes even more common in the verse smr¢tis. It is also found in R¢gvedakhila, 4.2.6, 8; R¢gvedakhila, 7.7. 24 See Gautamadharmasütra, 1.6.11: ‘An Årya, even if younger, [should be shown honor] by a †üdra’ (avaro ’py åryaΔ †üdreña); 2.1.50: ‘The †üdra, the fourth estate, has one birth (only)’ (†üdra† caturtho varña ekajåtiΔ). Likewise, Baudhåyanadharmasütra, 1.5.10.20 prescribes Årya personal habits for ‘†üdras under the authority of Åryas’ (†üdråñåm åryådhiß™hitånåm). 25 Dåmodarpur (West Bengal) copper-plate inscription of Kumåragupta I, 128 GE (=446-447 CE): arhatha mama pa∞camahåyaj∞a-pravarttanåyånuvr¢ttåpradåkßaya-nœ[vœdharma?]-maryyådayå dåtum iti (ll. 6-7); with uncertain letters underlined, and missing characters supplied in braces (e.g., see the grant cited in the next note) (CII 32: 288-291).

II. Discourse, Conditions and Dynamics of Tradition in South Asia

89

were at least sometimes elicited by direct requests made by brahmins, who justified their petition by citing the Vedic rites that the grant would support.26 Moreover, ‡åstric stanzas are cited in many of these documents to attest to the inviolability of such endowments.27 The shift toward universalization is virtually completed with the Månavadharma†åstra, which was no longer understood as specially relevant to the Månava division of the Black Yajur Vedins but on the contrary lay claim to recording divine knowledge applicable to all Åryas, and essential to the successful governance of a royal state. 6. Further Roles for the Vrata With Veda study no longer simply the badge of brahmin status but that of Årya status, and mantra recitation widely recognized as a leading act of piety in itself, the set of rules to be followed when reciting also took on a broader significance. In the gr¢hyasütras and later works, a large array of specialized vratas are defined, some for the study of special texts within a given †åkhå, and others for the expiation of moral and ritual faults, or for the fulfillment of special desires. These are referred to, depending on the individual text, as vratas or dœkßås, or they are referred to by names of their own. Most of these are based to some degree on the general vrata of Veda-study (brahmacarya), and include some sort of initial rite (upåyana) parallel to the upanayana (or upåyana) of brahmacarya. Study or textrecitation is a part of some of these, but in others, the rules of ascetical discipline become the chief feature. A particularly rich source for early treatments of such practices was the Yajurveda tradition, which was in the vanguard in shaping late-Vedic piety. Space does not permit anything approaching a complete overview of these practices; a few examples will suffice. Some of the practices that found a place in the gr¢hya-codes were first introduced in the Taittirœyårañyaka. For example, the practice of private Veda recitation (svådhyåya) as a form of expiation is taught in Taittirœyårañyaka, 2.16-18. It is argued there that by acting as a priest in someone else’s sacrifice and receiving fees (dakßiñås), one ‘is drained, as it were, emptied out’ (rícyata iva vÌ eßá prévá ricyate yó yåjáyati práti vå gr¢hñÌti). The text suggests a remedy for this sorry condition: ‘without eating, he should thrice perform private recitation of the [entire] Veda’. As an alternative, one might simply recite the Såvitrœ verse continuously for three days and nights (Taittirœyårañyaka, 2.16.2). To perform such recitation, one should 26 In the Dåmodarpur copper-plate inscription of Kumåragupta I, 124 GE (=442-443 CE), the request of a brahmin named Karpa™ika for a piece of idle land (aprada-aprahatakhila-kßetra), according to the ‘law of endowment’ (nœvœdharma), to be used for the performance of agnihotra rites (CII 32: 282-287). 27 See note 46 below.

90

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

go out into the wild, into the forest (árañya) (Taittirœyårañyaka, 2.17.1). We also find in this work the fullest form of the küßmåñ∂a-verses, which are prescribed for one who is impure (ápüta) especially on account of sexual transgressions like the releasing of semen elsewhere than in a woman (yó ’yonau rétaΔ si∞cáti) (Taittirœyårañyaka, 2.8), a crime on a par with theft and even abortion (bhrüñahatyå),28 one of the most serious offences (Taittirœyårañyaka, 2.8.2).29 The basic action of the regimen is the offering of libations while reciting the verses. ‘He pursues a dœkßå in proportion to his sin (énas); the dœkßita makes continual offerings with these (küßmån∂a-verses)’ (yÌvad éno dœkßÌ ¶paiti dœkßitá etaíΔ satatí juhoti` [Taittirœyårañyaka, 2.8.4]). The dœkßå lasts for a year, at the end of which he is pure, but provision is made for shorter terms, which are all made equivalent to a year (Taittirœyårañyaka, 2.8.5-6). During this period he also must not eat meat, sleep with a woman, or sit on an elevated seat, and he should take care to avoid untruth. Just as in the Soma dœkßå, a special diet (likewise called the vrata) is prescribed for this küßmåñ∂adœkßå: hot milk for a brahmin, barley porridge for a råjanya, or a milk and grain mixture for a vai†ya. The text makes the parallel explicit: ‘One should declare this (to be) the vrata-food also in the soma ceremony’ (átho saumyé ’py adhvará etád vratá∫ brüyåt [Taittirœyårañyaka, 2.8.9]).30 The Kå™hakagr¢hyasütra is one of the few works of that class to include rules for the kr¢cchra (the ‘difficult’ penance) and its variants, the atikr¢cchra, taptakr¢cchra, and såntapana.31 Although not involving Veda-study, these require a mode of discipline otherwise very similar to that of the aß™åcatvåri∫†atsa∫mita, including chastity, standing by day and sitting by night (or reclining only at night), wearing a hemp or goat’s-hair garment, avoiding of honey, meat, salt, and †råddhafoods, and conversation with impure persons (Kå™hakagr¢hyasütra, 5). In each of these regimens, a simple vrata-upåyana rite is performed as the start of each segment: the Vasus are invoked as vratapatis and offered a sthålœpåka on the first three days, the Rudras on the second, the Ådityas on the third, and the Maruts and Aõgirases on the fourth (Kå™hakagr¢hyasütra, 8.1-2). In classical dharma†åstra, a profusion of ‘boutique’ regimens such as the famous cåndråyaña (‘moon-course’, in which the number of balls of food permitted waxed and waned with the moon) were taught, for the occasional use of householders 28 Many commentators understand this term to mean brahmahatyå, murder of a (learned) brahmin; see Malamoud 1977: 74-77. 29 This is a rare instance in an accented text of ritualized mantra-recitation by members of the second and third varñas, with distinctions by class (as in the gr¢hyasütras). 30 Malamoud (1977: 106; 171-172) translates this wrongly (“On doit prononcer aussi ce voeu dans la liturgie sômique”) precisely because of the ambiguity inherent in the conventional gloss, ‘vow’ (‘voeu’). See Lubin 2001. The vrata clearly is the milk-food mentioned in the preceding sentence (Taittirœyårañyaka, 2.8.8), as it is in the soma dœkßå. 31 Kå™hakagr¢hyasütra, 5-6; briefly mentioned at Baudhåyanagr¢hya†eßasütra, 5.6.26. The dharmasütras routinely prescribe these rules.

II. Discourse, Conditions and Dynamics of Tradition in South Asia

91

or the routine use of professional ascetics, who would become, especially after the Guptas, some of the most prominent exponents of Brahmanical knowledge. The Vedic vrata also was adapted for use by the new deity cults that sprang up from within or on the periphery of Vedic Brahmanism. The På†upata sect furnishes a striking example of a vrata adapted to serve as a sectarian initiation: the på†upatavrata as prescribed in a pari†iß™a (supplement) to the Atharvaveda (Atharvavedapari†iß™a, 40).32 The initiation into this vrata is a variant of the upanayana, and the vrata itself is taken to be an extension of brahmacarya, following the study of other parts of the Veda. Just as in the upanayana, the preceptor presents to the tonsured disciple a belt of mu∞ja grass and a wooden staff.33 The initiate is taught a gåyatrœ verse addressed to Rudra, and an Atharvan verse, ‡aunakasa∫hitå, 7.87.1, is recited over him.34 Fire offerings to Vratapati (Agni) and Rudra are made, followed by the initiate’s bathing in the ashes, the most distinctive practice of the På†upatas (Atharvavedapari†iß™a, 40.3.3-9). A range of possible durations is given, as is common with Vedic study vratas, but it may be also adopted on a permanent basis (naiß™ikam). One bound by the vrata is encouraged to reside in a shrine (åyatana) to Mahådeva. The reward is freedom from spiritual bondage, control over the senses, and communion (såyujya) with Pa†upati (Atharvavedapari†iß™a, 40.6.14-16). 7. Summary: Rules of Discipline, Textual Study, and Cultural SelfDefinition The decision to expand the scope of Veda study by making it a common ideal for all male Åryas may have secured the social value of Sanskrit learning in a world being wooed by the austere charms of the heterodox movements. The immediate effect may have been to divide the territory. Buddhism initially won out in the urban zones, where traditional social and cultural structures were fragmentary and diluted. Meanwhile, Brahmanism reinvented itself in a form that 32 This pari†iß™a has been translated and commented upon by Bisschop and Griffiths (2003). 33 The fact that the staff may by substituted by sword, club, or kha™våõga, as well as the reference to the skull-bowl (kapåla, 40.6.5), indicates that this pari†iß™a was redacted by someone familiar with (or by an adherent of) the Låkula subsect of the På†upatas, a group commonly called Kåpålikas, and thus cannot be much earlier than the 4th c. CE. The oldest epigraphical references to the movement are from 355-356 (copper-plates from Bagh [Ramesh, Tewari 1990]) and 380-381 (Mathura pillar inscription [Bhandarkar 19311932]), which allow us to deduce that the Lakulœ†a belonged perhaps to the early 2nd c. CE. 34 tat purußåya vidmahe mahådevåya dhœmahi Ù tan no rudraΔ pracodayåt: Atharvavedapari†iß™a, 40.2.5 = Maitråyañœsa∫hitå, 2.9.1:119.7-8; Kå™hakasa∫hitå, 17.11: 253.20-21; Taittirœyårañyaka, 10.1.5, 46.1; similar is Mahånåråyañopanißad, 3.2 (3.1, 3, and 4 have variants), 17.4. Another version of this stanza is also given (40.2.6). That we owe this description of the rite to På†upatas belonging to the ‡aunaka branch of the Atharvaveda can be seen in the use of ‡aunakasa∫hitå, 7.87.1, which appears also in the Atharva†irasopanißad (7.1), another Atharvan På†upata text.

92

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

simultaneously provided a model of domestic piety and personal sanctification (in a range of degrees of ascetic rigor) that had, especially in rural society, much of the appeal that Buddhist piety had in the cities and along trade routes: namely, a code of self-discipline and direct personal access to the presumed power of sacred mantras. The new Brahmanism, called ‘dharma’ (perhaps in imitation of Buddhist usage of the term),35 also aimed at establishing an ecumenical set of standards that could serve to coordinate the separate traditions of the individual Vedic schools. This encouraged a degree of standardization across the ever wider and more ethnically diverse territory inhabited by brahmins. The reproductive mechanism of this tradition was the regimen of brahmacarya, which sanctified the teacher-pupil relation as a spiritual filiation, and ensured the preservation and expansion of the texts and practices of the various subtraditions. Although it was apparently not administered or regulated by any central seat of authority, this system created a strong, trans-regional web of individual teacher-to-student bonds that worked well in low-population-density areas with a relatively stable caste society. Once the idea of a universal dharma was accepted — that is, a sense of dharma as a unified vision of human action— Smårta Brahmanism was positioned to compete with the universal Buddhist vision in the new urban courts. At the same time, certain new religious movements adapted some of the structural features of the old Brahmanism to their own purposes, and in so doing, helped broaden the scope of what Brahmanical culture might include. 8. Patronage We can gauge the eventual success of this dharma synthesis by tracking the increasing patronage of Brahmanism by post-Mauryan kings,36 and the adoption, beginning at least by the middle of the 2nd 35 Olivelle has recently proposed that early Buddhist teachings, and A†oka’s edicts in particular, were a catalyst in endowing the old Brahmanical term dharma (in the narrower sense of ‘model or rule of ritual practice’) with the broader significance, ‘righteousness, piety’ and by further extension, the doctrine embodying such piety (Olivelle forthcoming; see also the ‘Concluding Postscript’ of Olivelle 2004). Similarly, the use of the term deyadharma (the duty of giving, charity), which is common in the Buddhist literature, and appears in epigraphs recording Buddhist grants (e.g., Mathurå Buddhist image inscription of Huvißka, year 51 [=128/129 CE?], recorded in Sanskritized Prakrit), reappears later in Brahmanical grants. 36 Early inscriptions describing gifts to brahmins include a cave inscription (ca. 70-60 BCE) at Nånåghå™ by the Såtavåhana king ‡åtakarñi’s widow, a patron of the Vedic cult (Ray 1986: 36-37; Dehejia 1972: 19). Later, royally sponsored Vedic sacrifices were commemorated with inscribed stone yüpas (posts to which animal victims were tethered); an early one from Kau†åmbœ (early 2nd c. CE?) mentions the seven basic soma sacrifices by name, apparently sponsored by a royal minister (mantrin) ‡ivadatta who had received the grant of a village from the king. The same inscription records the endowment of a ‡aiva institution for the support of mendicants (carakair bhoktavyam), invoking the grace of Mahe†vara (Altekar in EI 24: 245-253).

II. Discourse, Conditions and Dynamics of Tradition in South Asia

93

century CE, of the brahmins’ liturgical language as the preferred language of expression in inscriptions.37 Sanskrit begins to appear in inscriptions only in the 1st century BCE (dating paleographically). Surviving examples include a brief record found at Ayodhyå, written by one Dhana[*deva?] who, perhaps in a Brahmanical echo of A†oka, styles himself dharmaråjan (sic), and appears to trace his lineage to Pußyamitra, founder of the ‡uõga dynasty (Sircar 1942: 96-97; Salomon 1998: 86-87). Another inscription of this period, found in multiple versions at Håthœbå∂å and Ghosuñ∂œ, provides for the worship of the Vaißñava deities Sa∫karßaña and Våsudeva. During the 1st century CE numerous short inscriptions in highly Sanskritized Prakrit were made around Mathura (Lders 1961). These were mostly Brahmanical in orientation until the reign of the Kußåña king Kanißka (from 78 CE). Thereafter, grants to Buddhists become common in a slightly less Sanskritized Prakrit than is generally found in Brahmanical grants. The first example of this is the famous Junåga∂h inscription of the mahåkßatrapa Rudradåman (EI 8: 36-49). Inscribed on a rock alongside edicts of A†oka, it commemorates Rudradåman’s renovation of an artificial lake described as having been first constructed by Candragupta Maurya and improved by A†oka and the Yavana king Tußåspha (line 8). This self-conscious claim to sustaining the legacy of the Mauryas is obvious, but there is a crucial difference: Rudradåman records this not in Prakrit, as A†oka did, but in a Sanskrit prose typical of the classical style (many compounds, few finite verbs).38 Moreover, he too presents himself as the upholder of dharma, in his case in the Brahmanical sense.39 An (already stereotypical) allusion 37 Sheldon Pollock emphasizes the fact that Sanskrit, which up to this point has been regarded primarily as the immutable language of religious expression, is being used in place of or side by side with vernaculars in a political context. Inscriptions constitute our first clear evidence of Sanskrit being recorded in writing, and it comes at virtually the same moment as the birth of kåvya (written, ‘aestheticized’ literary poetry, that is, literature per se), another feature of the courtly context (Pollock 1996; 1998: 10-19). In Pollock’s view, Prakrits “[…] disappeared from the epigraphical record throughout India in the space of a century, [...] and retained only a residual status in the literary-cultural order”, where “[…] under the influence of Sanskrit [they were] turned into cosmopolitan idioms [usable] anywhere within the Sanskrit cosmopolis” (Pollock 1998: 11). All this Pollock calls the creation of a ‘Sanskrit Ecumene’. Such linguistic developments were part of the broader trends in state formation and royal policy that are in full swing by the 4th century, with the rise of the Gupta dynasty. Royal affiliations with the cults of ‡iva and Vißñu are demonstrated in the minting of coins linking the king and the god; the erection and endowment of massive stone temples; and patronage of brahmins and ascetic groups. In this period the Puråñas are compiled, comprising mythologies, Dharma†åstric material, ritual prescriptions, and, significantly, royal genealogies that derive known royal lines from one or the other of the legendary lunar and solar dynasties. 38 Kielhorn in EI 8: 39-40. The Sanskrit is correct by Påñinian standards, overall, despite inconsistent application of sandhi rules. 39 Rudradåman is he ‘whose strong attachment to dharma is given impulse by his correctly raising his hand [in pronouncing judgement?]’ (yathårtha-hastocchrayårjitorjitadharmånurågena [sic] [ll. 12-13]). See Månavadharma†åstra, 8.2; this is Kielhorn’s conjecture in EI 8: 48.

94

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

to giving support to ‘cows and brahmins’ (an idea found, for example, in Råmåyaña, 1.24.13, etc.) is there;40 Kielhorn even sees, in a damaged portion, an allusion to the first three ‘aims of man’ as taught by Manu: dharma, artha, and kåma (line 11). Perhaps the key detail that might throw light on Rudradåman’s motive in having this inscription composed in Sanskrit is the description of him as ‘having attained wide fame for mastering, remembering, fathoming, and practicing the great sciences of word-and-meaning, music, logic, and so forth’ (†abdårthagåndharvvanyåyådyånå∫ vidyånå∫ mahatœnå∫ pårañadhårañavij∞ånaprayogåvåptavipulakœrttinå [l. 13]). The notion that expertise in the various branches of vidyå was the dharma of a kßatriya directly reflects the influence of the Brahmanical doctrine of Sanskrit learning as a criterion of high varña.41 The fact that this IndoScythian ruler was one of the first to employ Sanskrit in a political forum suggests that this innovation was a calculated effort to demonstrate publicly the legitimacy of his rule by embracing the sacral authority of the brahmins. Likewise, Sanskrit inscriptions from Nagarjunakonda also seem to be related to the influence of the Western Kßatrapas.42 Thus, as was observed already by Sylvain Lévy in 1902, and more recently by Damsteegt and Salomon, the shift to using Sanskrit, the brahmins’ liturgical language, for the business of state was primarily the initiative of foreign rulers –Scythians and Kußåñas– anxious to align themselves with a priestly class firmly rooted in Åryåvarta, the ‘Land of the Åryas’ (Lévy 1902: 117-119; Damsteegt 1978; Salomon 1998: 86-98). Once introduced by arrivistes, this policy was fully established as the royal standard by the imperial Guptas. The spirit of this policy is nicely crystalized in a verse from the Lakkhå Mañ∂al pra†asti of the 7th or 8th c. (Dehra Dun district) in which one King Åryavarman is praised, in the Åryå meter, for his åryavratatå, his ‘Årya piety’ (Salomon 1998: 276-280). From this time onward, the granting of villages to individual brahmins and to groups of brahmins to support their holy practices and to earn merit (puñya) for the donors becomes more common, both in the old Brahmanical heartland of the Madhyade†a, but also in regions officially outside the bounds of Åryåvarta (e.g., Kaliõga). These landgrants were usually reinforced by the quotation of stanzas that praised the giving of land as the best of gifts and threatened punishment in hell for those who interfered with such an endowment or took away the land from those endowed. These stanzas were identified in the 40 This notion appears elsewhere in Gupta epigraphs, even where the donation is to Buddhists, as in the Sanchi stone inscription of Candragupta II, of the year 93 (CII 32: l. 10 [tad etatpravr¢tta∫ ya ucchindyåt sa gobrahmahatyayå sa∫yukto bhavet pa∞cabhi† cånantaryair iti]), where it is conjoined with a properly Buddhist category of sins, the anantaryas. 41 The term varña itself is used (l. 9). 42 Memorial pillar inscription from the time of King Rudra-Purußadatta, recording a Kßatrapa alliance with the Ikßvåkus (EI 34: 20-22).

II. Discourse, Conditions and Dynamics of Tradition in South Asia

95

inscriptions as belonging to a text of ‘dharma’ or ‘dharma†åstra(s)’; often the source was identified more specifically as Vyåsa (e.g., the ‘Vyåsagœtå’ or Mahåbhårata) or Manu (e.g., Månavadharma†åstra), or another of the sages.43 The recipient’s qualification for such patronage, wherever it was mentioned, was his training in textual recitation and the application of mantras in ritual performances, or expertise in a learned discipline such as grammar, logic, law, astrology, or poetics. The authority of the brahmin was thus explicitly justified, in principle anyway, by his mastery of sacred knowledge. Thus for example a copper plate of Samudragupta records his gift of two villages to a learned brahmin as an agrahåra: ‘So you should listen to this Traividya [one who has studied three Vedas] and obey his commands’ (tad yußmåbhir asya ca ttraividyasya †rotavyam åj∞å ca kartavyå [ll. 7-8]).44 The plate styles Samudragupta ‘sponsor of an a†vamedha rite’ (a†vamedhåharttuΔ). The regions corresponding to modern Orissa provide an interesting case of the phenomenon described here. The early inscriptions from this region, which lay beyond the pale of Åryåvarta, date from the beginning of the Gupta era. The local rulers, under Gupta influence (and suzerainty), began to record a long series of grants of villages to brahmins, using the Sanskrit language and a very consistent format.45 The brahmins are generally identified by name, but also by gotra (brahmin clan; in practice, by clan-subdivision, gotragaña), by †åkhå (‘branch’ of the Veda, i.e. individual mantra sa∫hitå) and/or caraña.46 When the ‡ailodbhavas ascend the throne in central Orissa (seventh century), their records are so meticulous that they note a recipient’s pravara (list of ancestors recited at the invocation to Agni in a Vedic sacrifice) as well as their gotra. In many cases, the donees are described as (sa)brahmacårins47 or chåtrabråhmañas (‘brahmin stu43 In fact, although the Mahåbhårata contains a long section on bhümidåna (13.61), and although the Månavadharma†åstra also contains similar material, most of these stanzas do not appear in either source. The names Vyåsa and Manu rather stand for dharma†åstra in general, and the stanzas would seem to constitute a loose body of orally preserved individual maxims, most of which did not appear in any generally recognized compilation. This is evident from the fact that from one inscription to the next, even over a short period of time and in inscriptions of a single donor, the stanzas were recorded in varying forms and sequences. For an alphabetical list of most of these stanzas, see Sircar 1965, app. II: 170-201. 44 CII 32: 224-228 (n. 3); this is dated to the fifth year of his reign, although some doubts have been raised on the authenticity of the record. 45 For convenience I refer here to Rajaguru’s (1958: vol. 1, pt. 2) thorough compilation of early inscriptions from Orissa, although the pattern continues in later inscriptions (1958: vol. 2). Inscriptions are cited in the present article by their number in Rajaguru 1958. 46 The term caraña is conventionally used to denote, within a †åkhå, a subschool identified with the ritual sütra which it follows. In the Orissa inscriptions, by contrast, it is a larger category than †åkhå, and thus is virtually identical with ‘Veda’. For instance, donees may be identified as belonging to the Chandoga caraña (i.e. the Såmaveda) and the Kauthuma †åkhå (nn. 36, 40); Våjasaneyin caraña (i.e. the ‡uklayajurveda) and Kåñva †åkhå (n. 44). 47 In seven out of fifteen records of the Må™haras and ‡rœråma-Kå†yapas: nn. 3, 6, 8, 9, 11, 14, 15. It is sometimes a reference to prior studies rather that a description of the donee’s current status.

96

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

dents’)48 in the course of stating their affiliations, or they are presented as teacher and student (†ißya) (n. 43). In some cases, the scholarly credentials are spelled out at length. An inscription of Pr¢thivœ Mahåråja (n. 15) records a grant ‘to Padma†arman, a (former) Taittirœya student, belonging to the Bhr¢gu gotra, who has complete knowledge of the Veda and Vedåõgas, who delights in the six rituals, who observes the rules of moral restraint (yama), moral principle (niyama), and full-recitation, and who has discerned the farthest reaches of several sciences including puråña, Råmåyaña, and dharma†åstra’.49 A grant of the same king from three years earlier (n. 14) makes an even longer eulogy of the donee, a fellow-student of the Chandogas (chandogasabrahmacåriñe), learned in 3000 subjects, author of twenty commentaries, the son of one who studied, taught, and commented on a thousand subjects, and grandson of one who was master of †ruti and smr¢ti and whose mind had been purified by performance of the agniß™oma and other rites. Another record creates an agrahåra for a group of brahmins ‘well-behaved and engaged in study’ (v[r¢]ttådhyayanavatå∫), headed by a brahmin teacher (bråhmañaupådhyåya), from another agrahåra.50 The stated motive for endowments to brahmins is usually to increase the donor’s store of merit (puñyavr¢ddhi), with attendant rewards for self and family in this life and the next. Behind this formula, however, is the assumption that patronage of Brahmanical ritual and Sanskrit erudition confirmed the donor as a legitimate receptacle of kßatra (ruling power) in the eyes of rivals and of his own courtiers alike. But it was not necessarily enough simply to be a brahmin to merit such patronage. At least until the Guptas, grants to Buddhist and Jain monks and institutions were more common. Hence, in Brahmanical grants, the listing of Vedic affiliation, the mention of the recipient’s qualifications in terms not merely of descent but of learning, scholarship, and Vedic ritual observance were essential to projecting brahmins as a ‘good investment’. Thus far, we have seen no evidence of members of other high castes engaging in study of the Veda or of other subjects, despite the injunctions in the codes that they could and should do so. This may in part be due to a lack of occasion for recording such practice in inscriptions: only Brahmins could teach or officiate as priests and were thus entitled to support through endowments. But this silence surely also reflects the fact that actual study by non-brahmins 48

In two records from north Orissa: nn. 23, 27. veda-vedaõga-paragåya ßa™karmmaniratåya yama-niyama-paråyañåya puråñaråmåyaña-dharma†åstrady-aneka-vidyå-påradar†ane bhr¢gusagotråya taittirœyasabrahmacåriñe padma†armmañe [retaining orthographical irregularities, such as the omission of some long vowels; uncertain letters underlined]. 50 This is mistranslated by Rajaguru, who wishes to emend to vr¢ttådhyayanavratå∫; this is impossible to reconcile with the plural genitives on either side of the compound (nånågotracarañånå∫ […] bråhmañånå∫), which which it must agree. 49

II. Discourse, Conditions and Dynamics of Tradition in South Asia

97

remained a rare exception. The persuasiveness of the Årya model was manifested mainly in the growing recognition of the value of supporting brahmins in these pursuits, and this recognition correlates with the proliferation of endowed brahmin settlements. 9. Political Factors in the Spread of Brahmin Communities The migration and resettlement of brahmins all over South Asia and into much of Southeast Asia from the early historic period onward is a subject that is far from being well documented or understood.51 In broad terms however, it appears that those laying claim to social dominance or kingship in regions outside the Brahmanically defined heartland of Åryåvarta or Madhyade†a made a policy of importing brahmins from areas within or near this central, ‘pure’ region and settling them on tax-free land. Settling brahmins in agriculturally underdeveloped lands might have acted as a mechanism for bringing them under the plow while simultaneously drawing them into the sphere of the ‘civilized’ world as conceived by Brahmanical tradition.52 Many inscriptions emphasize the brahmin recipients’ extraordinary rights over the land and their own authority over the surrounding peasantry.53 Kulke has pointed out that the brahmin officials of the court and the (more numerous) brahmins of the rural agrahåras formed a nexus that allowed an ‘inner colonization’ of the countryside by the king (Kulke 1978b). Meanwhile, brahmins were installed as priests in the new royal temples, where by composing måhåtmyas and sthala puråñas they helped assimilate the local (often tribal) deity to pan-Indian Sanskrit mythological and cultic norms. In this way, the king (often a scion of the dominant local tribe) could maintain the allegience of his ‘core constituency’ while appropriating an Årya cultural framework to unify and lend prestige to the state.54 In such cults, non-brahmin tribal priests often 51 For use of inscriptions to track the spread of brahmin communities, see Witzel 1981; 1985; 1993; for Southeast Asia, see, esp., Bhattacharya 1961; Christie 1964; Coedès 1983; Kulke 1986 (a critical review of earlier scholarship). 52 The fact, frequently mentioned in land-grant records, that the land being bestowed is not agriculturally productive might imply the expectation that the recipient will make it productive; e.g., see notes 26 and 27 above. However it has also been argued that such settlements would only be possible in regions already agriculturally productive enough to support a village of non-agriculturalists (Stein 1967-1968). 53 Injunctions to all parties to obey the brahmin donees and to respect the terms of the grant are routine in north India in this period. An early (ca. 5th c.) example from Pülåõku„ici in the South likewise ordains the superior rights (mœyå™ci) of the brahmin titleholders over tenants (Veluthat 1993: 198). 54 Kulke (1978b: 125-127) provides three excellent examples from 5th-6th c. Orissa of “[…] this early type of royal patronage of autochthonous deities”: the hill temple of Ambikå Mañinåge†varœ (for the inscriptions, see Rajaguru 1958: 120-123; 133-135 [nn. 23, 27]); the goddess Stambe†varœ patronized by Råja Tuß™ikara (ca. 500) (Kalahandi copperplate; see Sircar in EI 30: 274 ff); and the Gaõga kings’ patronage of the ‡abara (Saora) tribal god under the name Gokarñasvåmin on Mahendragiri. See, also, Eschmann 1978.

98

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

retain special functions connected directly with the shrine image, while brahmin priests carry out the standard Brahmanical temple rites (a pattern that continues in the worship of Jagannåtha at Puri). The temple brahmins included both those trained in Vedic recitation, and those who had taken ‡aiva or Vaißñava dœkßå and were thus qualified to perform temple püjå.55 The village agrahåras served as the training ground and source of manpower to staff the large urban temples and the court bureaucracy, as they continue to do today, especially in south India.56 Likewise patronage from the royal courts, especially from the Gupta era forward, led to the construction and staffing of brahmincontrolled temple institutions, which especially in the South also became vehicles for investment (Stein 1960). Such temple institutions allowed the ruler to present himself publicly as the earthly counterpart and representative of the Lord of the Universe, a role dramatized by the cult of the devaråja (‘divine king’) in Southeast Asia.57 When a new state was founded, brahmin priests could be called upon to ‘confirm’ the new ruler’s kßatriya lineage and to secure the blessings of a heavenly king. 10. The Learned Brahmin Ruler A less common trope in the inscriptions is that of the brahmin dynasty, such as the Kadambas. An inscription on a pillar in front of the temple to ‡iva Prañave†vara at Tå¬aguñ∂a in the Shimoga district of Karnataka, dating probably to the 6th century CE, depicts the royal family as brahmins turned conquerers.58 After invoking ‡iva, the second verse praises brahmins as ‘gods on earth, best of the twice-born, speakers of the Såma-, R¢g, and Yajur- Vedas’.59 The Kadamba lineage itself is described as ‘a high twice-born family, [...] sons of a three-sage line in the Håritœ pravara, born in the gotra of Månavya, foremost of 55 The dœkßå of the Tantric sects differed fundamentally from any Vedic vrata or dœkßå, but it was parallel in the symbolism of divinization, in providing access to knowledge and qualification to practice, and in serving as a status-marker. 56 ‘Tantra’ (i.e. temple ritual) specialists in the manas (brahmin family compounds) of Kerala today are often away from home for up to half the year serving in distant temples. 57 One point of controversy has been the significance of the ‘devaråja cult’. Kulke (1978a) has made it clear that the devaråja per se was the deity as divine king, while the human king was understood as the god’s human representative. Rather than being a god in human form, the king was at most said to embody a ‘portion’ (a∫†a) of the deity’s nature. On the other hand, the title appears to be applied to the Gupta king Candragupta II in the Sanchi stone inscription of the year 93 (mahåråjådhiråja†rœ-candraguptasya devaråja iti priyanåma […] yetasya [CII 32: ll. 7-8]); the ambiguity of the referent of the title is caused by the loss of a few characters, which the editor proposes to restore as priyanåmadheya∫ bhavaty etasya. If the word devaråja indeed applies to the king, we must wonder what the cultic significance of it might be. 58 Edited and translated by F. Kielhorn in EI 8: 24-36; re-edited by Sircar 1942: 450-455. 59 bhüsurå dvijapravarås såmargyajurvvedavådinaΔ (l. 1).

II. Discourse, Conditions and Dynamics of Tradition in South Asia

99

sages’.60 This precise brahmin pedigree is then augmented with a list of the ritual observances and textual studies (vv. 5-8). The inscription portrays the family’s rise to royal stature as the restoration of brahmins to their primordial superiority to kßatriyas –a superiority lost only in the current decadent epoch: ‘In this Kali Yuga, alas, brahminhood is feebler than the kßatriyahood. Hence, if, even after fully serving the teacher’s household and diligently studying one’s branch of the Veda, perfection in brahman is subordinate to the king, what greater sorrow is there than this?’61 Thinking thus, Mayüra†arman took up arms against the Pallava lords and won their recognition as a regional ruler. In this case, Vedic learning directly propels the family to worldly power, and justifies their possession of it. In a less elaborate earlier record, the mahåråja Måtr¢vißñu, a feudatory of the Gupta emperor Budhagupta, in sponsoring the erection of a flagpole for Lord Janårdana, advertises his status as ‘the great-grandson of Indravißñu, the brahmin sage, bull of the Maitråyañœyas, who took delight in his ritual practice, worshipped with sacrifices, and had learned the recitation [of his Veda]’.62 In these and similar inscriptions, the claim to brahmin status is documented by referring to ritual observance and Vedic study, in the latter case, specifying the branch of the Yajurveda to which the king’s progenitor belonged, and in which he trained (adhœta) in recitation (svådhyåya). 11.Conclusion Between the Ganges urbanization and the end of the first millennium CE, the Brahmanical tradition adapted to dramatically changing social and economic situations, while expanding into new territories far beyond the old bounds of Åryåvarta. It was able to do this while maintaining a strong sense of continuity through its ability to project its idealized vision of a varña society under the religious 60 dvijakula∫ prå∫†u […] tryår†avartmahåritœputram r¢ßimukhyamånavyagotrajam (l. 2). This should probably be understood to mean that they belonged to the Hårita (or Håriti) gaña of the Kevala Aõgiras gotra, in whose pravara the r¢ßis Aõgiras (or Måndhåtr¢), Ambarœßa, and Yuvanå†va are named (Brough 1953: 121-135). r¢ßimukhyamånavyagotrajam, rather than alluding to an actual gotra in the technical sense, may simply be meant to place the family in the brahmin class generally by referring to Manu, the prototypical Årya. It is worth noting that the Gotrapravarama∞jarœ specifies, as a sort of afterthought, that råjanyas may adopt the pravara of their purohita or of their teacher, or ‘if (a råjanya) makes the pravara-recitation according to his rank [sårß™i∫]’, the sages Manu, I∂å, and Purüravas should be mentioned (212-215). Thus, Manu might have been seen as a sage to be named for a king who lacked proper pravara. Is this, then, a suggestion that the Kadambas were not in fact true brahmins, and had adopted the Hårita gotra and three-sage pravara from his purohita? 61 kaliyuge ’sminn aho bata kßatråt paripelavå vipratå yataΔ ÙÙ gurukulåni samyag åråddhy †åkhåm adhœtyåpi yatnataΔ brahmasiddhir yyadi nr¢pådhœnå kim ataΔ para∫ duΔkham ity ataΔ ÙÙ (l. 4). 62 Erañ pillar inscription, 165 GE (=483-484 CE?): svakarmmåbhiratasya kratuyåjinaΔ adhœtasvådhyåyasya viprar†er mmaittråyañœyavr¢ßabhasyendravißñoΔ prapauttreña (CII 32: 339341 [n. 39], ll. 4-5).

100

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

leadership of dvijas, men sanctified by sacred knowledge, and empowered to mediate that knowledge to others. Those who assent to the authority of that knowledge (the Veda) and of it mediators (learned, observant brahmins), and who in turn are qualified to be served ritually by such mediators –are thereby deemed Åryas. The crucial ritual mechanism of this tradition, the vrata of brahmacarya (and the initiation that set it in motion), was doubly necessary in this process. First, it was the gateway to the knowledge sought by both the mumukßu, the seeker after liberation, and the bubhukßu, the seeker after earthly rewards. Second, because in itself it provided a general model of personal piety and stamped the observant with outward marks: the sacred thread of the orthodox householder and his ritualized use of mantras. Brahmins’ reputation for learning –for which the feats of memory and erudition of even just a few served as proof– lent them the authority to assert themselves persuasively as arbiters of dharma in a world where they had many plausible rivals. What is most distinctive of the tradition, perhaps, is that however far the brahmin priest was able to insinuate himself into the life of the urban courts, the tradition maintained a rural, parochial character, rooted in the village or even, in the case of the Brahmanical ascetic, in the å†rama, the monastic retreat, which never grew to the massive proportions and complexity of many of the Buddhist monasteries. This decentralized structure, with the scope it gave to the local parißad, might well be judged the reason for its persistence through a history that has seen many religions rise, and more than a few fall.

BIBLIOGRAPHY CII EI

Corpus Inscriptionum Indicarum Epigraphia Indica

Aklujkar, A. (1996), The Early History of Sanskrit as Supreme Language, in Houben, J.E.M. (ed.), Ideology and Status of Sanskrit, E.J. Brill, Leiden, pp. 59-85. Allchin, F. R. (1995), The Archaeology of Early Historic South Asia: The Emergence of Cities and States, Cambridge University Press, Cambridge. Altekar, A.S. (1947-1948), Nandsa Yupa Inscriptions, in EI 27: 252-267 (+ 2 plates). Bhandarkar, D. R. (1931-1932), Mathurå Pillar Inscription of Chandragupta II, Gupta Era 61, in EI 21: 1-9 (+ plate). Bhattacharya, K. (1961), Les religions brahmaniques dans l’ancien Cambodge d’après l’épigraphie et l’iconographie, EFEO, Paris.

II. Discourse, Conditions and Dynamics of Tradition in South Asia

101

Bisschop, P., Griffiths, A. (2003), The På†upata Observance (Atharvavedapari†iß™a 40), in «Indo-Iranian Journal», vol. 46, n. 4, pp. 315-348. Brough, J. (1953), The Early Brahmanical System of Gotra and Pravara: A Translation of the Gotra-Pravara-Ma∞jarœ of PurußottamaPañ∂ita with an Introduction, Cambridge University Press, Cambridge. Christie, A. (1964), The Political Use of Imported Religion: An Historical Example from Java, in «Archives de sociologie des religions», vol. 9, n. 17, pp. 53-62. Coedès, G. (1968), The Indianized States of Southeast Asia, edited by W.F. Vella, translated by S.B. Cowing, University of Hawaii Press, Honolulu. Damsteegt, Th. (1978), Epigraphical Hybrid Sanskrit: Its Rise, Spread, Characteristics and Relationship to Buddhist Hybrid Sanskrit, E.J. Brill, Leiden. Erdosy, G. (1988), Urbanisation in Early Historic India, BAR, Oxford. ______ (1995a), Language, Material Culture and Ethnicity: Theoretical Perspectives, in Id. (ed.) The Indo-Aryans of Ancient South Asia: Language, Material Culture and Ethnicity, de Gruyter, Berlin, pp. 1-31. ______ (1995b), The Prelude to Urbanization: Ethnicity and the Rise of Late Vedic Chiefdoms, in Allchin (1995), pp. 75-98. Eschmann, A. (1978), Hinduization of Tribal Deities in Orissa: The ‡åkta and the ‡aiva Typology, in Eschmann, A., Kulke, H., Tripathi, G.C. (eds.), The Cult of Jagannath and the Regional Tradition of Orissa, Manohar, New Delhi, pp. 79-98. Fussman, G. (1987-1988), Central and Provincial Administration in Ancient India: The Problem of the Mauryan Empire, in «The Indian Historical Review», 14 (July-January), pp. 43-72. Gokhale, B.R. (1965), The Early Buddhist Elite, in «Journal of Indian History», vol. 42, n. 2, pp. 391-402. Gonda, J. (1977), The Ritual Sütras. A History of Indian Literature, Otto Harrassowitz, Wiesbaden, vol. 1, fasc. 2. Hultzsch, E. (1925), The Inscriptions of A†oka, in CII I, Oxford(new ed.). Kane, P.V., (1974), History of Dharma†åstra, vol. 2, pt. 1, Bhandarkar Oriental Research Institute, Poona. Kulke, H. (1978a), The Devaråja Cult, Paper n. 108, Southeast Asia Program, Cornell University, Ithaca N.Y. (orig. pub. in 1974, H. Kulke, Der Devaråja-Kult, in «Saeculum», vol. 25, n. 1, pp. 24-55 [rep. in Kulke 1993: 327-381]). ______ (1978b), Royal Temple Policy and the Structure of Medieval Hindu Kingdoms, in Eschmann, A., Kulke, H., Tripathi, G.C. (eds.), The Cult of Jagannath and the Regional Tradition of Orissa, Manohar, New Delhi, pp. 125-128 (rep. in Kulke 1993: 1-16). ______ (1986), Max Weber’s Contribution to the Study of ‘Hinduization’ in India and ‘Indianization’ in Southeast Asia, in Kantowsky, D.

102

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

(ed.), Recent Research on Max Weber’s Studies on Hinduism, Weltforum-Verlag, Mnchen, pp. 97-116 (rep. in Kulke 1993: 240-261). ______ (1993), Kings and Cults: State Formation and Legitimation in India and Southeast Asia. Manohar, New Delhi. Lévy, S. (1902), Sur quelques termes employés dans les inscriptions des Kßatrapas, in «Journal Asiatique», ser. 9, vol. 19, pp. 95-125. Lubin, T. (2001), Vratá Divine and Human in the Early Veda, in «Journal of the American Oriental Society», vol. 121, n. 4, pp. 565-579. ______ (forthcoming), The Householder Ascetic and the Uses of Self-discipline, in Flgel, P., Houtman, G. (eds.), Asceticism and Power in South and Southeast Asia, Curzon Press, London. Lders, H. (1961), Mathurå Inscriptions (Unpublished Papers Edited by Klaus L. Janert), Abhandlungen der Akademie der Wissenschaften in Göttingen, Philologisch-Historische Klasse, Dritte Folge, n. 47, Vandenhoeck & Ruprecht, Göttingen. Malamoud, C. (1977), Le Svådhyåya: Récitation personnelle du Veda. Taittirœya-Årañyaka, livre II, ICI, Paris. Olivelle, P. (1993), The ņrama System, Oxford University Press, New York. ______ (2000), Dharmasütras: The Law Codes of Åpastamba, Gautama, Baudhåyana, and Vasiß™ha, Oxford University Press, Oxford. ______ (2004), The Semantic History of Dharma: The Middle and Late Vedic Periods, in «Journal of Indian Philosophy», vol. , n. 5-6. ______ (forthcoming), Power of Words: The Ascetic Appropriation and Semantic Evolution of ‘Dharma’, in Flgel, P., Houtman, G. (eds.), Asceticism and Power in South and Southeast Asia, Curzon Press, London. Pollock, S. (1995), Literary History, Indian History, World History, in «Social Scientist», vol. 23, n. 10-12, pp. 112-142. ______ (1996), The Sanskrit Cosmopolis, 300-1300: Transculturation, Vernacularization, and the Question of Ideology, in Houben, J.E.M. (ed.), Ideology and Status of Sanskrit, E.J. Brill, Leiden, pp. 197-247. ______ (1998), India in the Vernacular Millennium: Literary Culture and Polity 1000-1500, in «Daedalus», vol. 127, n. 3, pp. 1-34. Rajaguru, Sri S. (1958), Inscriptions of Orissa (300-700 A.D.), Shiromani Press, Berhampur (Orissa), vol. 1, pt. 2. Ramesh, K.V., Tewari, S.P. (eds.) (1990), A Copper-Plate Hoard of the Gupta Period from Bagh, Madhya Pradesh, Archaeological Survey of India, New Delhi. Ray, H.P. (1986), Monastery and Guild, Oxford University Press, Delhi. Ray, R.A. (1994), Buddhist Saints in India, Oxford University Press, New York.

II. Discourse, Conditions and Dynamics of Tradition in South Asia

103

Salomon, R. (1998), Indian Epigraphy: A Guide to the Study of Inscriptions in Sanskrit, Prakrit, and the Other Indo-Aryan Languages, Oxford University Press, New York. Sircar, D.C. (1965), Indian Epigraphy, Motilal Banarsidass, Delhi. ______ (1942), Select Inscriptions Bearing on Indian History and Civilization, From the Sixth Century B.C. to the Sixth Century A.D., Calcutta University, Calcutta, vol. 1. Stein, B. (1960), The Economic Function of a Medieval South Indian Temple, in «Journal of Asian Studies», vol. 19, n. 2, pp. 163176. Trautmann, T.R. (1971), Kau™ilya and the Artha†åstra, E.J. Brill, Leiden. Veluthat, K. (1993), The Political Structure of Early Medieval South India, Orient Longman, New Delhi. Witzel, M. (1981), Materialien zu den vedischen Schulen I. Ueber die Caraka-‡åkhå, in «Studien zur Indologie und Iranistik», vol. 7, pp. 109-132. ______ (1985), Regionale und Überregionale Faktoren in der Entwicklung vedischer Brahmanengruppen im Mittelalter (Materialien zu den vedischen Schulen, 5), in Kulke, H., Rothermund, D. (ed.), Regionale Tradition in Sdasien, Beiträge zur Sdasienforschungen 104, F. Steiner, Stuttgart, pp. 37-75. ______ (1987), On the Localisation of Vedic Texts and Schools (Materials on Vedic ‡åkhås, 7), in Pollet, G. (ed.), India and the Ancient World: History, Trade and Culture Before A.D. 650, Leuven, pp. 173-213. ______ (1989), Tracing the Vedic Dialects, in Caillat, C. (ed.), Dialectes dans les littératures indo-aryennes, Paris, pp. 97-264. ______ (1993), Toward a History of the Brahmins, in «Journal of the American Oriental Society», vol. 113, n. 2, pp. 264-268. ______ (ed.) (1997), Inside the Texts, Beyond the Texts: New Approaches to the Study of the Vedas, HOS Opera Minora 2, Harvard University, Cambridge.

III. How to Produce, Construct and Legitimate a Tradition

ALF HILTEBEITEL Buddhism and the Mahåbhårata. Boundary Dynamics in Textual Practice

It has long been felt that in the formation of boundaries between the religious traditions of South Asia, the composers of the Mahåbhårata would have played a considerable role in generating the dynamics of what was to become Hinduism. But since the Mahåbhårata is quiet if not exactly silent on the non-Brahmanical traditions, and particularly about Buddhism, scholars have not found it easy to discern how it might have constructed such borders and, still more durably, how it might have generated a new textual praxis that could be used by later epic and puråñic authors to patrol them –if indeed such borders existed. One strain of scholarship approached this question from the standpoint that the Mahåbhårata would have grown from oral origins into a massive ‘encyclopedia’, one that could eventually claim, ‘whatever is here may be found elsewhere; what is not here does not exist anywhere’ (Mahåbhårata, 1.56.33; 18.5.38).1 From this vantage point, a text of such self-sufficiency and self-importance could, at the most, have absorbed some minor references to the heterodoxies only haphazardly as a reflex of its snowball descent through the centuries (Hopkins 1969: 363-402; 475). This view concurs with an assimilationist model of Hinduism’s relation to other traditions. Another approach has been to suspect that the Mahåbhårata has more to say about Buddhism than it makes immediately obvious, and that what it has to say would have to have been said at some significant time in history. This view requires a more dialogical or interactive model such as is favored in this essay. But it is important to 1 For an argument that the term ‘encyclopedia’ has been misleadingly applied to the Mahåbhårata, particularly with reference to this verse, which, rather than defining the exhaustiveness of the text, is pitched toward an ‘ontological debate’, see Hiltebeitel 2001: 162-163.

108

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

emphasize that the question is posed not with regard to the relation between traditions themselves as ‘wholes’, but with regard to the position of texts, and mainly one text, in defining one aspect of the relation between these traditions. The question of the relation between Buddhism and the Mahåbhårata is an old one, going back most famously to the younger Adolf Holtzmann’s ‘inversion theory’ of 1892-95, which views the original Mahåbhårata as a Buddhist epic celebrating Duryodhana in the image of A†oka, and in memory of a national resistance against the Greeks, only to be subject to later Brahmanical inversions, marked by the rise of Kr¢ßña and Vißñu, that turned the plot upside down (Hiltebeitel 1979: 69). As we shall see, certain authors are still playing with some of the same gamepieces. But they come to quite different conclusions and have, I think it fair to say, a better understanding of the text and the historical possibilities for contextualizing it. What is perhaps surprising is that the issue took on sudden new steam, with three authors –Madeleine Biardeau, James Fitzgerald, and Nicholas Sutton– taking up the topic between 1997 and 2002.2 Fitzgerald also offered a preview of his argument in a footnote in his 1980 dissertation (Fitzgerald 1980: 151, n. 1 [see then Fitzgerald 2001: 64, n. 5]), the very year that I developed an argument of my own for a different Buddhist backdrop,3 one that I would now like to rethink in relation to these more recent offerings. At the beginning of the summation of her views under the heading Conclusion: Épopée et Bouddhisme in her Le Mahåbhårata: Un récit fondateur du brahmanisme et son interprétation (Biardeau 2002: vol. 2, 745-783), Madeleine Biardeau asserts that, even if everything [said about the enormous poem] remains hypothetical, it remains nonetheless that we have thereby pushed its signification, or rather that we have given it greater depth, opened new questions but also endowed it with a certain coherence: it is a matter of an apocalyptic crisis of the world that leads a temporal cycle (not identical with the puråñic temporal cycle, but figuring it) from its genesis to its end,4 while passing through a long war-sacrifice, or, better still, a long sacrificial session of apocalyptic magnitude –since it brings about the birth and disappearance of all the sovereigns of the 2 One should also be alerted to work in progress by Greg Bailey, for the moment exemplified in his paper presented at the Conference on Religions in the Indic Civilization and kindly supplied by the author (Bailey 2003). I comment on this paper only lightly. 3 The article was written in late 1979 and delivered in January 1980 at the ‘Seminar on Ancient Mathurå’, sponsored by the American Institute of Indian Studies at New Delhi and Mathura, but was not published until 1988 (Hiltebeitel 1988: 93-102). See Biardeau’s kind encouragement to revisit the topic in Biardeau 2002: vol. 1, 21 n. 7. Fitzgerald and I developed our early views on these matters entirely independently. 4 That is, from the burning of Khåñ∂ava Forest toward the end of the first Book of the Mahåbhårata to the disappearance of the Yådavas in the sixteenth, which is ‘fully accepted by Kr¢ßña’ and marks also the end of the magic power of Arjuna’s and Kr¢ßña’s weapons (747). Here as throughout, my translation.

III. How to Produce, Construct and Legitimate a Tradition

109

Earth, all the while assuring the continuity of the future existence of the Earth purged for a time of the wicked. (Biardeau 2002: vol. 2, 747)

From this opening, Biardeau turns immediately to the matter of writing among ‘other questions’ one might ‘pose’:5 For example, could such a vast poem belong to oral literature, or must one see in it the first written work of what post-Vedic culture produced up to then, despite all that this leaves in shadow and confusion. One sets apart the rock and pillar edicts of A†oka themselves, which no doubt inaugurate the practice of epigraphy, promising a great development and representing one of the sources of Indian history. Still, one must note that these edicts make use of two scripts [Bråhmœ and Kharoß™hœ]. Today, certain specialists think that it is materially impossible to regard it as an oral composition because of its dimensions. But we know what libraries the Brahmans educated according to traditional methods still transport with them in their memories. It is true that these methods tend to disappear, but we have enough witnesses of them to believe that it would be possible to conceive a work of this scale.6 (Biardeau 2002: vol. 2, 747-48)

Biardeau notes that there is no evidence of a scribe other than Gañe†a, who appears nowhere else in the [Critical Edition] text, but, on the contrary, evidence of well identified and diverse narrators, and continues: We know that a brahman poet of the time of the epics does not compose in the silence of his office, calamus in hand and a piece of birch bark or dried palm leaf on a plank on his knees, but that he is first of all a narrator, better, a perpetual master. The word is his sole force and his memory is its first guardian. In him he has a universe of sounds, for he is (like all other men and more than them) in the bosom of the †abdabrahman, of the brahman that is sound or word. [...] If we believe it has to do with the beginnings of the epic, such a poem, charged with teaching and at the same time with entertaining, seeks, while not leaving the sacrificial terrain nor the dialogue between the brahman and kßatriya, to replace the debates of the Upanißads between brahmans and kings centered on åtman-brahman. He must probably be a Brahman (or, some would say, would have been one in a previous life) to have the curse of writing and think that only the word is gold. Moreover, the problem appears to me to be secondary here and to require, to comprehend it, competencies –and beliefs– that we do not have in a civilization of the machine at every register. I only wish to forestall an objection: have I forgotten this problem? No, I believe it to be secondary, and still more highly improbable as a problem than all those we have raised from the beginning. We have posed from the beginning, as our fundamental hypothesis, a causal rapport between the conversion of A†oka and the composition of the Mahåbhårata. (Biardeau 2002: vol. 2, 748-49) 5 She begins “there would indeed be other questions to pose. For example, could so vast a poem still belong to an oral literature [...]” (Biardeau 2002: vol. 2, 747). 6 “[...] concevoir une oeuvre à cette échelle”. My italics.

110

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

The last sentence, a whole paragraph in itself, thus punctuates what is not secondary but primary, the conversion of A†oka. But what is the status of secondary problems? Is there not a difference between ‘conceiving’ the scale and dimensions of the Mahåbhårata and producing it, not to mention remembering and reproducing it orally? Granted we must understand the epic poets to value the word in a world of Vedic sound. But the epic poets mention writing and would seem to allude to books in interesting ways. 1. Mahåbhårata References to Writing When Bhœßma tells Yudhiß™hira, ‘sellers of the Vedas, corrupters of the Vedas, and those who write the Vedas, these surely go to hell’,7 he registers one of the reasons why a text calling itself a ‘fifth Veda’ would want to make itself appear oral. Yet according to the bard Ugra†ravas, the Mahåbhårata weighs more on a scale than the four Vedas (Mahåbhårata, 1.1.208), which could suggest a written book. Most interesting, consider this story from the Nåråyañœya section of the Mahåbhårata’s ‡åntiparvan (12.335.21-66), describing how Vißñu-Nåråyaña awakens from his yogic sleep at creation and assumes the form of Haya†iras, the ‘Horse’s head’, to rescue the creator Brahmå, freshly emerged in the primal lotus, from the two demons Madhu and Kai™abha, whose opposition to Brahmå interrupts him as he is in the process of creating the worlds by ‘first emitting the four Vedas’ (sr¢jantam prathama∫ vedå∫† catura† [Mahåbhårata, 12.335.25cd]). What do these demons do to interrupt this ‘Vedic’ manner of creation? Having seen (dr¢ß™vå) the Vedas, those two best of demons, bearing form themselves, then forcibly seized [or stole] the Vedas in Brahmå’s sight (sahaså jagr¢hatur vedån brahmanaΔ pa†yatas tadå). Then those two best of Dånavas, having stolen the Vedas (vedån gr¢hya), quickly entered the Raså8 in the great ocean of the northeast. (Mahåbhårata, 12.335.26-27)

The great Ocean is the Milky Ocean, and its northeast is henceforth to be the location of this Horse’s head manifestation of Nåråyaña. Note that the demons are assisted in stealing the Vedas by being able to see them rather than just hear them –which indeed, perhaps quite appropriately, it is not said that they do. In any case, with the Vedas stolen, Brahmå is stupefied, and heartrendingly deplores their loss. 7 Mahåbhårata, 13.24.70 (vedavikrayiña† caiva vedånå∫ caiva düßakaΔ Ù vedånå∫ lekhakå† caiva te vai nirayagåminaΔ). 8 See also Mahåbhårata, 12.335.3 and 54 on the Raså in the ocean to the NE as where the Horse’s head is also located. In the Vedic story of Saramå, the heavenly bitch, the Raså is a heavenly river (R¢¢gveda, 10.108).

III. How to Produce, Construct and Legitimate a Tradition

111

The Vedas are my supreme eye. The Vedas are my supreme strength. The Vedas are my highest foundation. The Vedas for me are the supreme brahman. All the Vedas are stolen from me by the two Dånavas’ power. The worlds (just now) born are made dark without the Vedas. Without the Vedas, what shall I do, risen up to create the worlds? O alas, great is my woe born from loss of the Vedas. It burns my affected heart, made subject to great pain. Who will lift me up now that I am sunk in this ocean of grief? Who will bring me those lost Vedas? Who will be favorable to me? (Mahåbhårata, 12.335.29-32)

He thinks of Hari Nåråyaña and lauds him for his aid, recalling how they have cooperated through six prior creations, and closes: The eye of the Veda (vedacakßus), transcending age, I am fashioned by you. Those stolen Vedas are my eye (te me vedå hr¢tå† cakßus). I am born blind. Wake up! Give me two eyes (dadasva cakßusœ mahyam)! I am dear to you, you are dear to me. (Mahåbhårata, 12.335.41cd-42)

Nåråyaña then awakens. By his lordly yoga he assumes the vast cosmic form of the Horse’s head (Mahåbhårata, 12.335.43-48); referred to as ‘the repository of the Vedas’ (vedånåm ålayam [Mahåbhårata, 12.335. 44d]), the Head sets forth for the Vedas’ retrieval: [...] the lord, lord of all, disappeared and entered the Raså. Having entered the Raså, he resorted to the highest yoga. Having adopted the tone regulated by the rules, he emitted the sound O∫. And this sound resonated and was smooth everywhere it went, strengthened in the interior of the earth, and risen from the qualities of all the elements. Then the two Asuras, having bound the Vedas together (kr¢två vedån samayabandhanån), having hurled them (vinikßipya) into the Rasåtala, ran to where that sound was (52). In the meantime (etasminn antare), O king, the god bearing the (form of the) Horse’s head, Hari, grabbed all the Vedas that had gone to the Rasåtala (jagråha vedån akhilån rasåtalagatån hariΔ [53cd]), gave them to Brahmå, and then returned to his own nature [which, as we soon learn, is that of the sleeping Nåråyaña as Aniruddha]. After he had established the Horse’s head in the northeast of the great ocean, the Horse’s head then became the repository of the Vedas (vedånåm ålaya†).9 Thereupon the two Dånavas Madhu and Kai™abha, not seeing anything, again came there. The two speeders looked around where they had hurled the Vedas, but that place was absolutely empty (yatra vedå vinikßiptås tat sthåna∫ †ünyam eva ca). Then resorting to high speed, the two best of the powerful again quickly rose up from the abode of the Raså. And they saw that Purußa, the lord, the maker of beginnings, white with a radiance of lunar purity, established in the body of Aniruddha; of immeasurable vigor, under the influence of the sleep of yoga, he was defined upon the waters on a bed abounding in serpent coils prepared to his own measure that was surrounded by a garland of flames. Having seen that he was endowed with a spotless luminosity that had the splendor of solar rays, the two Dånava lords released a 9 As

already in a prior use of this phrase cited above.

112

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

great laugh. Pervaded by rajas and tamas, they said, ‘This white Purußa lies under the influence of sleep. It is he who has taken back (åharañam) the Vedas from the Raså. Whose is this? Indeed, who is it? How come he sleeps with this serpent?’ When the two had spoken so, they awakened Hari seeking to fight. Understanding that the two were seeking battle, the godly Purußottama, having beheld those two Asura lords, put his mind on war. And so there was a war between the two and Nåråyaña. The bodies of both Madhu and Kai™abha were pervaded by rajas and tamas. Gratifying Brahmå, Madhusüdana slew (them). Then, with the death of those two and the quick retrieval of the Vedas (å†u vedåpaharañena ca), Purußottama relieved the grief of Brahmå. Then Brahmå, turned around, his enemy slain, provided (now) with the Veda, fashioned all these mobile and stationary worlds. (Mahåbhårata, 12.335.49-66)

That the four Vedas are seen, stolen, thrown, and bound together by the two Asuras all suggests that they were manuscripts! Moreover, it is highly curious that it is Asuras who are mentioned as binding the Vedas in this form and right away throwing them to hell (Rasåtala is one of the seven hells). Vißñu retrieves them in the Horse’s head form and gives them to Brahmå, for whom having manuscripts of the Veda may not be the same problem they would be for humans. Haya†iras is then made the repository of the Vedas in the northeastern ocean, where their orality is certainly primary, and, indeed, soon enough the passage emphasizes the Horse’s head as the source and inspiration of oral means of Vedic preservation: Having praised with fierce tapas the god bearing (the form of) the Horse’s head, the sequence (of recitation, kramaΔ) was obtained by Pa∞cåla on the path pointed out by Råma.10

Now it is one of Biardeau’s hermeneutical principles, and one with which I concur, to interpret certain Asuras and Råkßasas in the epics as figures of heresy, in particular of Buddhism.11 And this myth is repeatedly recalled whenever Kr¢ßña is called Madhusüdana, Slayer of Madhu12 –albeit that we do not know it is this version of the myth, which one could argue is from a ‘late’ text within the epic, the Nåråyañœya. Clearly the Nåråyañœya is hinting that heretics, Buddhists no doubt the first among them, would have little reason of their own not to throw written Vedas into hell, and would delight in interrupting –indeed, debunking– the proper ‘Vedic’ course of divine creation. Buddhist cosmogonies, of course, make Brahmå deluded in the very thought that he is the Creator.13 10 Mahåbhårata,

12.335.71. See in the same text 12.330.31-39, especially 37. Chåndogyopanißad, 8.7-8 for an early portrayal of the Asuras as materialists. 12 As he is, for instance, repeatedly (twelve times in the three adhyåyas 5.138, 139, and 141) in the episode where Kr¢ßña tempts Karña with the kingship that could follow from Karña’s acknowledging that he is the five Påñ∂avas’ elder brother. 13 Dœgha Nikåya, 3.28ff. (see Embree 19882: 127-128). 11 See

III. How to Produce, Construct and Legitimate a Tradition

113

2. A†oka and the Mahåbhårata In any case, I do not think one can treat a rapport between the conversion of A†oka (who ruled from ca. 269-232 BCE, with his conversion probably some time after the Kaliõga war in ca. 260 [Thapar 1998: 3335]) and the composition of the Mahåbhårata by leaving writing as a secondary, improbable, and virtually inadmissable problem imposed by the machine age. Biardeau hypothesizes a strict correlation in time between the Mahåbhårata and A†oka. Indeed, for her the two ‘successive’ epics, the Mahåbhårata and the Råmåyaña, with the Mahåbhårata the earlier of the two, make up a ‘brahmanical manifesto’ provoked by the ‘imperium of A†oka’, one that ‘dissimulates its real end’, which must be decrypted or decoded.14 Moreover, while the Mahåbhårata is composed under conditions of ‘urgency’ apparently during the A†okan reign itself, the Råmåyaña may, she thinks, have been created under less urgent conditions after the Mauryan empire had ‘crumbled’ (Biardeau 2002: vol. 2, 776). Nicholas Sutton also proposes “a date for the final reworking of the epic narrative some time in the Mauryan era” (Sutton 1997: 335; 339). The period of A†okan rule is probably too early for the epics to have been written, but others –myself and James Fitzgerald included– have proposed a later date, about a century later at the earliest: that is, for me, no earlier than the ‡uõgas and probably about 150 BCE to the year Zero.15 That would make writing far more feasible. I also argue, as does Fitzgerald, that when the Mahåbhårata speaks of nåstikas or ‘heretics’, it has other non-Brahmanical countermovements in mind beside Buddhism alone: i.e. the Jains, whom Biardeau treats as having little if any relevance (Biardeau 2002: vol. 2, 768 n. 23), the Materialists, and the Åjœvakas.16 I do not think, however, that A†oka is the only royal figure of the period to leave traces in the Mahåbhårata, for I think it quite likely that in its problematization of fighting and royal Brahmans, especially Droña,17 the Mahåbhårata may be reflecting on the reign of the Brahman Pußyamitra ‡uõga (ca. 185149 BCE), the overthrower of the Mauryas –a point with which Fitzgerald agrees.18 Further, as Biardeau herself notes, the postMauryan era marks a rise in the significance of the northwest as a place opposed to the more orthodox center. There one finds the Yavanas, 14 While I am cautious of the notion of ‘code’, and would prefer to think of a more flexible semiosis of referencing or allusion (Hiltebeitel 2001: 119 and passim), the test of a code is to see if it works. I can only say that in thinking through the figures and terms that Biardeau decodes in connection with Jaråsandha, I have found several points, to be discussed below, where I believe it is possible to extend her insights. 15 See Hiltebeitel 2001: 18; Fitzgerald 2001: 78-83. 16 See Hiltebeitel 2001: 163 and n. 115 (with various references, several drawn from Nicholas Sutton 2000). Fitzgerald (Fitzgerald 2001) and I are close on these matters. See also the Jåbåli episode in Råmåyaña, 2.100-102.1. 17 But consider also Kr¢pa, A†vatthåman; and see Råvaña. 18 We both consider the epic’s negative evaluation of Brahman rule as evidence for a post –or at the earliest mid– ‡uõga date. See Fitzgerald 2001: 84; Hiltebeitel 2001: 16-17.

114

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

‡akas, and Yue-chi, with a great stüpa at Purußapura/Peshawar, and Taxila/Takßa†œlå as a crossroads of mixed population following destructive wars, while commercial exchanges are open with the West and China (Biardeau 2002: vol. 2, 749) –all of which may have something to do with the way the northwest is represented in both epics as a place where kingdoms (Gandhara [Biardeau 2002: vol. 1, 531], Madra, Bålhœka, Kaikeya) have questionable dharma. Nonetheless, Fitzgerald19 –like Nicholas Sutton (Sutton 1997: 331-341)– would agree with Biardeau that A†oka is the most highimpact historical royal figure on the Mahåbhårata, but the three of them differ over how and where to trace that impact. Biardeau traces A†oka primarily into the career of the Magadha king Jaråsandha; Fitzgerald and Sutton trace him into the figure of Yudhiß™hira. Yet Fitzgerald and Sutton part company over whether Yudhß™hira is a dark or light counterpart to A†oka. For Fitzgerald, Dharmaråja Yudhiß™hira is a grim and somber extension of his father Dharmaråja Yama, the god of death, who must oversee a divine raiding party of the gods that descends to earth to restore Brahmans to privileges denied them by A†oka –Brahmans, that is, who, Fitzgerald thinks, composed an initial ‘main Mahåbhårata’, a first written redaction, out of ‘rage’ at their treatment under A†oka– “a deep and bitter political rage at the center of the Mahåbhårata”.20 For Sutton, on the contrary, Yudhiß™hira is a rather ideal representation of A†oka as a figure of non-cruelty and forgiveness. Sutton argues that “[t]he kßatriya-dharma taught in the dharma-†åstras and the ‡åntiparvan of the Mahåbhårata is rejected by the A†oka of both legend and edict, as it is by Yudhiß™hira in the epic” (Sutton 1997: 334) –a somewhat anachronistic point for A†oka, while for Yudiß™hira it is in unex19 As observed above –mentioning Fitzgerald 1980: 151, n. 1; Fitzgerald 2001: 64, n. 5–, Fitzgerald traces his view back to his 1980 dissertation. There, the note on the subject follows this statement: “A proper king should be a†oka in his execution of dharma” (Fitzgerald 1980: 151). But Fitzgerald supplies no citation for such a description of the king in the epic, and must rather admit in his note: “Actually, more typical language of the text is vi†oka or vœta†oka. I do not get the impression that the late or final redactors of the MBh consciously attempted to draw a parallel between Yudhiß™hira here and the legendary Buddhist king. [...] On the other hand, the parallels between the situations of Yudhiß™hira and A†oka, and the contrast at the doctrinal level (regardless of actual behavior) between the non-violent and renunciatory ideology of Buddhism and the deliberate Hindu sanctioning of violence for dharmic ends and the Hindu attempts to synthesize the renunciatory perspectives of mokßa with the material and social processes of society (in the å†ramadharma and the karmayoga) make it difficult not to see the MBh making some reply to the Buddhist pretense of having an adequate definition of the role of the emperor. The Hindu king must fight and be a†oka at the same time” (Fitzgerald 1980: 151, n. 1). It is not clear whether Fitzgerald at this point was inclined to think the epic’s rebuttle of A†oka would have been only in a late redaction, to which he assigns most of the ‡åntiparvan in his more recent work. 20 See Fitzgerald (2001: 85) thus attributing this ‘rage’ to his first group of epic-writing Brahmans as the motive behind their portrayal of his darker Yudhiß™hira (85-90), a dark underside that I would certainly agree is there (Hiltebeitel 2001: 119-120, 135-139), but not prior to his larger portrayal as a thoughtful, virtuous man and endearing source of occasional light.

III. How to Produce, Construct and Legitimate a Tradition

115

plained contradiction to his being not only the main hearer of the ‡åntiparvan, but his rather small show of resistance to its Realpolitik.21 Moreover, Sutton thinks that Brahmans would have admired A†oka because he patronized them along with Buddhists (Sutton 1997: 340). Sutton and Fitzgerald also have diametrically opposed views of the Mahåbhårata’s take on A†oka’s conversion. For Sutton, A†oka converts to a peaceful rule by dharma, and he sees “the fictional Yudhiß™hira” as “modeled” after “the historical A†oka and other kings of a similar disposition” (Sutton 1997: 334; 339). Fitzgerald, however, regards “the figure of Yudhiß™hira at the beginning of the ‡ånti Parvan, in his attempt to renounce the kingship and go to the forest”, as “deliberately scripted by the authors of the epic to represent what they saw to be wrong with the Mauryan emperor A†oka, to purge and refute whose rule was, I believe, the principal purpose for the creation of the first generation of our written Sanskrit Mahåbhårata” (Fitzgerald 2001: 64-65). On these differences, we could say that Biardeau agrees with both on the pivotal character of A†oka’s conversion, but has greater agreement with Fitzgerald at least to the degree that the A†oka reflected in the text would be the dark one.22 But her Jaråsandha is, as we shall observe further, a figure not of the Hindu god of death, Dharmaråja Yama, but the Buddhist one, Måra. Meanwhile, my attempt23 was also to consider Jaråsandha as a figuration of Buddhism, but of Buddhist cosmological and soteriological ideas rather than of a specific personage, whether mythic or historical. I shall unpack this difference in the next section. Nonetheless, Biardeau and I would agree that whatever the correlations one makes, they require an interpretation tied to the Mahåbhårata’s articulation of bhakti themes and idioms, which set the iconic and narrative conventions for multiple deities and other related figures, but in particular those connected with Kr¢ßña, who leads the opposition to Jaråsandha. As Biardeau demonstrates, while Kr¢ßña, Balaråma, and Vißñu as Purußottama acquire “distinctive attributes that one finds as well on coins, pillars, and in the epic text” –without it being possible to say which is prior–, “the representation of the Buddha and his partisans starting from the monumental aniconic stüpa reliquary” goes on along side that “of the essential personages of postvedic Brahmanism, whom the epic already endows with specific attributes that are acquired henceforth” (Biardeau 2002: vol. 2, 751). In particular: 21 See further Sutton 1997: 336, stating, “Yudhiß™hira utterly rejects this part of his education”, rather overstating the case. Similarly, it is unconvincing to align A†oka’s condemnation of religious festivals and rituals with Yudhiß™hira’s hesitant run-up to his postwar A†vamedha (Sutton 1997: 357). 22 See above all, and most recently, Biardeau 2002: vol. 1, 782 (the theme appears in numerous of Biardeau’s earlier writings). 23 See n. 3 above.

116

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

The Mathurå region, on the Yamunå and all around out to Avanti and Vidi†å and Så∞cœ, is testimony to this. At Mathurå, to our surprise, it is Buddhism that appears dominant there. But has not Kr¢ßña himself taken the precaution in the epic to explain why he and the ‘clans’ that surround it … have had to flee from Mathurå … because of Jaråsandha? They now occupy the extremity of the Gujarat peninsula! But above all, let us keep from confounding epic time and historical time. The epic neither invents the future nor recalls the past, not to mention the present. Apparently history gives another explanation: the Buddhists have invaded Mathurå (more by conversions than by migrations), but despite all there are devotees of Bhagavån Vißñu not far from there as we know, a little more to the south, not far besides from the illustrious Buddhist cult center at Så∞cœ. Kr¢ßña will have his revenge at Mathurå, and it is still there that one finds him today.24 (Biardeau 2002: vol. 2, 751)

Fitzgerald takes a different view here. For him, bhakti is late to the epic and the Jaråsandha episode would probably not be part of his earliest written ‘main Mahåbhårata’.25 In excising bhakti, Fitzgerald’s notion of Brahman rage is quite different from Biardeau’s take on Brahman-Buddhist interactions, in which she proposes two forms of bhakti, Brahmanical and Buddhist, developing along side each other, in the latter case among Buddhists who are “for the most part of Indian origins and inserted in the society of castes”, fully ‘at home’ (chez eux) there, with “no one desir[ing] their departure, despite this sort of Brahmanical manifesto […] that the imperium of A†oka provokes” in the form of the two epic texts (Biardeau 2002: vol. 2, 776). Here, of course, I side with Biardeau, and can see nothing to recommend Fitzgerald’s conception of a pre-bhakti ‘main Mahåbhårata’ in which faith is a matter of afterthoughts. For me, the Mahåbhårata and Råmåyaña are both designed to sustain a subtle, patient, and emerging political theology that propounds a new bhakti framework in which royal patronage and Brahman prestige are provided with new places and meanings. Nonetheless, while I regard Biardeau’s stress on Buddhism and Brahmanism’s mutual at-homeness to be of vital importance, something more needs to be said about the oppositions between them that can be a self-conscious feature of their texts. The Mahåbhårata in particular often reflects greater antipathy toward Buddhism than Biardeau indicates. I shall recall this point in connection with some of the passages she cites. From here on, then, I would like to turn to two points that Biardeau makes central to her interpretation of the relation between 24 Similarly, one finds representation of Rudra-‡iva, Mahißåsuramardiñœ, Skanda, Sürya (fitted with Iranian boots), all “announcing the diversity of Hinduism” (Biardeau 2002: vol. 2, 752-753). See, for discussion of extra details, also Biardeau 2002: vol. 1, 131-133. Further Hiltebeitel 1988: 94-97. 25 Fitzgerald doesn’t mention it among those he thinks to be not ‘main’, but he does include the closely related ‡i†upåla episode as exemplifying the non-originality of “all episodes that elaborate some theme of devotion to Vißñu, ‡iva, or Kr¢ßña” (Fitzgerald 2003: 3).

III. How to Produce, Construct and Legitimate a Tradition

117

Buddhism and the Mahåbhårata: her decoding of the Magadha king Jaråsandha, which I shall compare with my own; and her treatment of Mårkañ∂eya’s prophetic account of the ills of the Kaliyuga. 3. Jaråsandha decoded. Toward the beginning of the Mahåbhårata’s second book, the Sabhåparvan, it is the troubleshooter-saint Nårada who inspires Yudhiß™hira with the idea of the dangerous Råjasüya sacrifice that will make him an ‘emperor’ (sa∫råj). Yudhiß™hira checks with Kr¢ßña, who says they must first eliminate Jaråsandha, Yudhiß™hira’s only rival for paramountcy or empire (sa∫råjya), who ‘has imprisoned eighty-six kings in an ‘enclosure for men’ (purußavraja) at Girivraja, the future Råjagr¢ha in Magadha […] in preparation for a sacrifice of a hundred kings to ‡iva, that is, implicitly, a sacrifice of the entire [årya] kßatra’ excluding his own line and others to be mentioned, ‘for, as Kr¢ßña tells Yudhiß™hira, Jaråsandha sows dissension among the one hundred and one lineages of the Solar and Lunar dynasties’ (Mahåbhårata, 2.13.48 [see Hiltebeitel 1988a: 97]). At Kr¢ßña’s advice, Yudhiß™hira sends Kr¢ßña (for policy or guidance, naya), Arjuna (for victory), and Bhœma (for force) to Magadha, to which the three set off on Kr¢ßña’s chariot with its Garu∂a pennant, Kr¢ßña at the reins prefiguring his role as Arjuna’s charioteer (Biardeau 2002: vol. 2, 775). Meanwhile, we learn of Jaråsandha’s birth. Cañ∂a Kau†ika, a Vai†våmitra R¢ßi, gave a mango to the Magadha king Br¢hadratha, who then gave halves of it to his two wives. The child was born in two halves, discarded near the palace, and then –soon giving him his name– he was ‘put together by Jarå’: a magnanimous Råkßasœ who ‘jauntily’ gave the ‘put-together’ child back to the king –since “she wished to recompense the king for the good treatment she received as a domestic divinity of the palace [...] Oh! Oh! [says Biardeau] Do we dare exclaim: what is this Råkßasœ doing in a royal palace?” (Biardeau 2002: vol. 2, 755-56). Indeed, Biardeau sees her as a precursor of Råvaña and the Råkßasas and Råkßasœs of Laõkå; but let us note that Duryodhana’s ninety-nine brothers are at least Råkßasas incarnate. Later, during the Mahåbhårata war, Kr¢ßña tells how Jarå died in the fight over Mathurå between Jaråsandha and Kr¢ßña when she was slain by Balaråma’s falling mace.26 According to Kr¢ßña, Jaråsandha’s allies are ‡i†upåla as protecteur des sots (that is, ‘protector of fools’); Ha∫sa (whose name ‘Gander’ evokes the supreme self or åtman) and ˘imbhaka (sot, ‘fool’). The last two die as part of a triple death: first a pretext-Ha∫sa is killed in battle; then ˘imbhaka drowns having thought it was the real Ha∫sa, his inseparable friend; then the real Ha∫sa drowns having learned of 26 Mahåbhårata, 7.156.10-14. See Biardeau 2002: vol. 2, 219-220; and further discussion below.

118

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

the drowning of ˘imbhaka. It seems to Biardeau that, along with the initial pretext-death of the first Ha∫sa in battle, this double drowning evokes the Upanißadic two birds as friends in the one tree,27 the åtman and jœva (the embodied ‘living soul’) –leaving ‡i†upåla as a ‘protector of fools (sots)’ who could not, however, protect these fools who represent a Buddhist misrepresentation of these two inseparable dimensions of the soul (Biardeau 2002: vol. 2, 756). Further ‘decoding’ Jaråsandha, Biardeau calls attention to the list of ‘descents’ of gods, demons, and others into the heroic generations, in which Jaråsandha is the incarnation of an Asura named Vipracitti (Mahåbhårata, 1.61.4) –Biardeau suggests “‘sans intelligence’ (?)” (Biardeau 2002: vol. 2, 756). Jaråsandha also appears at two svaya∫varas: that of Draupadœ in Pa∞cåla,28 and that of an unnamed princess of Kaliõga, daughter of the Kaliõga king Citråõgada, who becomes the wife of Duryodhana –a story told only at the beginning of the post-war ‡åntiparvan, where Nårada answers a question from Yudhiß™hira: why was my brother Karña so fated with adversities? When the Kaliõga princess reviewed the kings who had come to win her hand, she passed over Duryodhana, and feeling the snub, Duryodhana abducted her on his chariot –counting on the help of Bhœßma and Droña, but helped especially by Karña who routed the ensuing kings and escorted Duryodhana and his new bride home to Håstinapura. Having admired Karña’s warrior prowess, Jaråsandha then demanded to fight him in single combat –a sporting challenge (where it took place is not said) rather than a duel to the death. The fight began on chariots and ended in wrestling, until Jaråsandha had to call a halt, since the seam left from his two halves being ‘put together by Jarå’ at birth had begun to show signs rupture. Recognizing Karña as the victor, Jaråsandha gave him the city of Målinœ, which is apparently in or near Aõga –where Karña must already rule in subordination to Duryodhana. Says Biardeau: These two little chapters permit us to know that Duryodhana and Karña are in the camp of Jaråsandha, and that Jaråsandha is obliged to recognize the superiority of Karña. But it is difficult to go farther and draw conclusions from this as to their fidelity to Brahmanism, not least because they would have to be deceiving the warrior Brahmans or Kßatriyas who fight on their side against the Påñ∂avas. (Biardeau 2002: vol. 2, 757)

In the Hariva∫†a, Jaråsandha does include Duryodhana and his brothers among his allies (Biardeau 1978: 226; Hiltebeitel 1988a: 97). Moreover, it is highly fascinating that in the Tamil Terukküttu 27 ‡vetå†vataropanißad,

4.4.6-7; Muñ∂akopanißad, 3.1.1-3. Biardeau is relying on the Vulgate here, citing Mahåbhårata, 1.186.23 (= Critical Edition 1.1821*) and Mahåbhårata, 1.187.26 (= Critical Edition 1.1828* lines 5-6), but in both cases with fairly wide manuscript support. 28

III. How to Produce, Construct and Legitimate a Tradition

119

tradition of dramas on the Mahåbhårata, this story from the ‡åntiparvan is clearly known, despite the fact that the drama cycle and the classical Tamil Makåpåratam of Villiputtür Åÿvår that it builds on both stop their narratives with the death of Duryodhana and the end of the war. Moreover, it is not Duryodhana who marries this princess –now named Po¡¡uruvi, the ‘Golden Earth’– but Karña, because Karña touched her during her abduction, and because Duryodhana, despite his great friendship for Karña, always sees him as a man of low caste (De Bruin 1999: 286-90; 314; De Bruin 1998: 238-47).29 Biardeau also calls attention to the other kings who came to this Kaliõga svaya∫vara to vie for the princess. One is named A†oka (Mahåbhårata, 12.4.7c) –a surprise that Biardeau does not do more than note in passing without working it directly into her interpretation (Biardeau 2002: vol. 2, 527; 529-32; 757). But the list is interesting, and bears on her notion of alliances between Duryodhana and a possibly pro-Buddhist camp. Here is the list of kings attending the Kaliñga svaya∫vara –an odd lot that would likely consist quite precisely of kings not penned up in Jaråsandha’s corral at Magadha, the properly årya kings whom Jaråsandha is keeping to sacrifice to ‡iva: ‡i†upåla, Jaråsandha, Bhœßmaka, Vakra, Kapotaroman (Pigeon-Hair), Nœla, Rukmin, Dr¢∂havikrama, the Mahåraja Sr¢gåla who was the overlord of the kingdom of women, A†oka, ‡atadhanvan, and the heroic Bhoja –these and many other kings from the south, and mleccha preceptors and kings from the east and north, O Bhårata. (Mahåbhårata, 12.4.6-8)

One wonders at the company that Duryodhana and Jaråsandha keep, which includes not only this A†oka and Mlecchåcharyas –implying some kind of ‘barbarian preceptors’ from the north and east (the Bhavißyapuråña uses the term mlecchåcharya for Adam, Enoch, Noah, Moses, Buddha, Jesus, and Muhammad) (Hiltebeitel 1999: 277-78)–, but the ‘great king ‡r¢gåla (‘Jackal’), overlord of the kingdom of women’, who may be the ‡r¢gåla Våsudeva mentioned in the Hariva∫†a as one of the ‘pseudo Våsudevas’, another of whom, Pauñ∂ra (i.e. from Bengal) Våsudeva, is explicitly mentioned in the Mahåbhårata as siding with Jaråsandha (Sörenson 1963: 718-19; Hopkins 1969: 217).30 These ‘pseudo-Våsudevas’ perversely claim the title that ‘properly’ belongs to Kr¢ßña, for whom it denotes both his paternity and divinity (Hiltebeitel 1988a: 101, n. 49). As with this A†oka, Biardeau only notes these names, and turns her attention to Bhœßma as, again, one who abducts (or in this case 29 On the name Po¡¡uruvi, following Madeleine Biardeau’s suggestion, see Hiltebeitel 1988b: 399. 30 See further Hiltebeitel 1999: 150 (in the Bhavißyapuråña’s retelling of the Hindœ oral epic Ålhå, ‡r¢gåla, ‘Jackal’, is the demon incarnated by the demonic king Jambuka, who is himself the father of the major demon foe Kåliya, an incarnation of Jaråsandha). See also Hiltebeitel 1999: 164, n. 27: in the Mahåbhårata, Våsudeva of Puñ∂ra is not only a false Våsudeva but a false Purußottama.

120

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

helps to abduct) a bride for another Kuru king: that is, as he did at a svaya∫vara at K农, prior to these other two svaya∫varas, where he abducted three brides for the Kuru prince Vicitravœrya, his own halfbrother. But, notes Biardeau, this current svaya∫vara takes place among the Kaliõgas, that people whom A†oka made famous by the (historical) bloody conquest, regarding it as the regret that he put to the test to the point of inscribing it on stone in several places of India, not including Kaliõga, to be sure, and describing it as the point of departure for his conversion to Buddhism. (Biardeau 2002: vol. 2, 529)

That Duryodhana goes to Kaliõga, confident in Karña’s aid, to get his bride is for Biardeau a confirmation of an earlier suspicion that the name ‘Kaliõga’, interpreted as ‘that which goes toward Kali’ (kali∫-ga), implies affinities between Kaliõga and Duryodhana, who is the incarnation of the demon Kali and who would thus (for Biardeau) be the king Kali of the Kaliyuga.31 Indeed, Biardeau finds more to support the idea that the epic gives a singular attention to Kaliõga, whose people, she suggests, would likely have been converted to Buddhism by A†oka’s conquest (Biardeau 2002: vol. 2, 54). For during the Mahåbhårata war, Kaliõga allies itself with the camp of Duryodhana, “whose Buddhist sympathies one can only guess” (Biardeau 2002: vol. 1, 298); and when Arjuna travels to the east on the first phase of the tour that will bring him three additional wives, the Brahmans accompanying him turn back at the entrance to Kaliõga, determining not to accompany him there (Mahåbhårata, 1.217.10) –on which Biardeau remarks, “so as not to tread on an impure terrain?”.32 For Biardeau, then, the account of the svaya∫vara in Kaliõga gives the poet the opportunity to place at the forefront an alliance between Duryodhana and Jaråsandha that is possible thanks to Karña. Karña’s centrality to this alliance is in evidence again after Kr¢ßña has used ‘stratagems’ to engineer the death of Gha™otkaca, which finally makes the death of Karña possible as the last real hurdle to Duryodhana’s defeat. It is at that very point –at Mahåbhårata, 7.156– that Kr¢ßña tells Arjuna how stratagems have been 31 “Le Kaliyuga ou le roi Kali qu’est Duryodhana” (Biardeau 2002: vol. 1, 298). See also Biardeau 2002: vol. 2, 54. The Bhavißyapuråña and the Ismaili ginån literature would seem to provide supportive evidence that the name can be used in this sense. This puråña, in retelling the Hindœ folk epic Ålhå, completes its profile of the story-launching demonic king (whom Hindœ versions call Karingha or Kariyå) by naming him Kåliya as lord of the Kaliyuga and making him an incarnation of Jaråsandha. Meanwhile, the ginåns simply call him Kaliõga, raising the likelihood that ‘Kåliya’ in the puråña evokes the Kaliyuga by augmentation rather than by connection with the verb gam). In either case, as also in Ålhå, the Påñ∂avas (reincarnated or resurrected) join the divine leader (the reincarnated avatåra or the Imåm) to destroy this king who embodies the Kaliyuga. See Hiltebeitel 1999: 150152; 341-347. Further, on deployment of themes from the Jaråsandha-‡i†upåla sequence, Hiltebeitel 1999: 150-152; 162-164; 346, n. 27. 32 Biardeau 2002: vol. 1, 298: “pour ne pas fouler une terre impure?”.

III. How to Produce, Construct and Legitimate a Tradition

121

used to eliminate other foes, and that had Duryodhana been able to ally himself in particular with Jaråsandha, ‡i†upåla, and Ekalåvya, he would have been unconquerable. Moreover, it is in this context that Kr¢ßña reveals how Jarå died as a result of Balaråma’s falling mace, as cited above. Amid such connections, it is perhaps significant that Kr¢ßña is repeatedly referred to as Madhusüdana in the episode of the ‘temptation of Karña’33 –as if Karña, as the errant figure of the pralayic sun, elicits this aspect of Kr¢ßña-Vißñu as the god who safeguards the Vedic cosmogony (Biardeau 2002: vol. 2, 531-32). But it is still Jaråsandha who remains central to this configuration of alliances. Biardeau’s hypothesis is that for the Mahåbhårata poet, Jaråsandha is a figure of Måra, Death, the Buddhist form of the Tempter who arouses desires and thus leads to “old age (jarå) and death” –that is, to repeated births (Biardeau 2002: vol. 2, 530, n. 5). Jårasandha is thus an ‘apocalyptic’ figure of Måra, recalling the Upanißadic compound jarå-mr¢tyu and the ‘more Buddhist’ jåra-maraña from which the Buddhist teachings of suffering, thirst, and impermanence, including the impermanence of the åtman, proceed (Biardeau 2002: vol. 2, 757). Thus when Jaråsandha is outwrestled by Karña and pronounces the latter victorious, this “‘Buddhist’ king plays a determining role in Karña’s celebrity and in the extension of his power, since he cedes a city to him” –by name Målinœ, which is also a name for Draupadœ in disguise, suggesting what is really coveted by Karña but also forbidden to him (Biardeau 2002: vol. 2, 757). Says Biardeau: Karña now has a rank that is more and more ‘royal’, whatever may be his birth. And the alliance of the three robbers gives precision to the hypothesis that the alliance with the Buddhists has been reinforced even before the intervention of Kr¢ßña, Arjuna, and Bhœma that is commanded from afar by Yudhiß™hira in preparation for his Råjasüya. But at the same time Jaråsandha discovers his limits, his congenital fragility. It serves as a warning of how he will die in combat with Bhœma. (Biardeau 2002: vol. 2, 530-31)

My own reading of Jaråsandha’s relation to Buddhism began with the observation that probably before, or at least in addition to, these alliances between Jaråsandha and the Kauravas, Jaråsandha has, as part of his ‘net of alliances’, one with Kr¢ßña’s first nemesis Ka∫sa, whose anxieties that Kr¢ßña has been born to kill him result in Kr¢ßña’s taking on his ‘cowherd disguise’ among the Gopås and Gopœs. Ka∫sa is allied with Jaråsandha through his marriage to the latter’s daughters Asti and Pråpti (Mahåbhårata, 2.13.30; Hariva∫†a, 2.34.4-6). I proposed that just as the Vålmœki-Råmåyaña probably associates Laõkå, “among its many symbolic associations, […] with Buddhism”, so too would it be “likely for Girivraja, with its caitya peak which 33 See

n. 12 above.

122

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Kr¢ßña, Arjuna, and Bhœma destroy –disguised as brahmans!– upon entering the city on their way to killing Jaråsandha” (Mahåbhårata, 2.19.2, 17, 41) to be associated with Buddhism: Girivraja and Råjagr¢ha are of course prominent in the early history of both Buddhism and Jainism, and a center of early Buddhist kings –most notably Bimbisåra and Ajåta†atru– whose throne supposedly descends from Jaråsandha (see e.g. Vißñu Puråña 4.23). And the region of Magadha is later the base of the first great Buddhist emperor, A†oka. But most curious are the names of Jaråsandha’s two daughters –Asti and Pråpti– whom he marries to Ka∫sa. It is these two women who prompt their father’s revenge against Kr¢ßña after the slaying of Ka∫sa. Unusual names for Indian girls, they both evoke prominent features of Sarvåstivådin Buddhism: Asti (sarvam asti, the phrase which gives the school its name) and Pråpti (‘obtention’, the Sarvåstivådin ‘pseudo-soul’). No other explanation for their names seems likely. (Hiltebeitel 1988a: 98)

As I still would not do today, I did not single out A†oka to the extent that others have come to do, but would rather place him in a long history of Brahman dissatisfactions not only with heterodoxies and heterodox rule, including not just the Buddhists but the Jains, but further with rule by Brahman kings. But of course Biardeau’s tracing of Jaråsandha’s Kaliõga connections would justify giving A†oka a certain prominence. From there, like Biardeau (Biardeau 2002: vol. 2, 751), I attempted to relate the opposition between Kr¢ßña and Jaråsandha to ground conditions, with Mathurå being “connected not only with [Kr¢ßña’s] need to operate in the epic from Dvårakå, but with the prominence in Mathurå during the period of the Mahåbhårata’s composition of both Jainism and Buddhism” (Hiltebeitel 1988a: 98). But my greatest emphasis was on Jaråsandha not as a figure who could be decoded with reference to one figure, one ‘signified’, whether A†oka or Måra as implying A†oka, but as a figure in juxtaposition with Kr¢ßña through whom –or rather, through both of whom– the Mahåbhårata poets could bring into relief a contest between Brahmanical and Buddhist cosmological and soteriological ideas. Thus, although I take Jaråsandha’s “curious name [...] ‘Put together by jarå’” much as Biardeau now does34 (that is, as implying the Buddhist connection between old age and death), I also take it to further allude to [...] the Buddhist ‘wheel’, the bhåvacakra [which] is precisely ‘put together by old age and death’. The twelve nidånas are drawn into a circle that ‘puts these two together’ with ‘ignorance’: [...] avidyå. But more than this, the Buddhist bhåvacakra is precisely a closed circle, without 34 Biardeau had an earlier explanation –Biardeau 1978: 227– relating the sandha in his name to either ‘pact’ or ‘twilight’ (as in sandhyå). See Hiltebeitel 1988a: 98 and n. 54.

III. How to Produce, Construct and Legitimate a Tradition

123

periods of crisis and renewal, yugas. That is, it is a circle that does not admit the intervention of the avatåra who ‘comes into being from yuga to yuga’ (Bhagavad Gœtå 4.8). ( Hiltebeitel 1988a: 98-99)

Here I refer back to a passage from the Mahåbhårata’s Udyogaparvan where Sa∫jaya tells Dhr¢taråß™ra how Kr¢ßña rules the ‘Wheel of the Yugas’ in relation to two grander wheels, the ‘Wheel of Time’ (probably implying kalpas) and the ‘Wheel of the Universe’ (probably implying space): As if sporting, the supreme being (purußottama) Janårdana keeps the earth, atmosphere, and heaven running. Having made the Påñ∂avas his pretext, and as if beguiling the world (loka∫ sammohayann iva), he wishes to burn your deluded sons [the Kauravas] who are disposed toward adharma. By his self’s yoga, the Lord Ke†ava tirelessly keeps the Wheel of Time, the Wheel of the Universe, and the Wheel of the Yugas revolving (kålacakra∫ jagaccakra∫ yugacakram […] parivartayate). I tell you truly, the Lord alone is ruler of Time and Death (kålasya ca hi mr¢tyo† ca), and of the mobile and the immobile. Yet ruling the whole universe, the great yogin Hari undertakes to perform acts like a powerless peasant (kœnå†a iva durbalaΔ). (Mahåbhårata, 5.66.10-14)35

I thus conclude that Kr¢ßña’s opposition to Jaråsandha and other “wheel-evoking foes36 may represent a confrontation of cosmologies: the bhakti cosmology of Hinduism which admits ruptures of time –twilights– for the sake of the world’s renewal, and images of Time without the possibility of such divine intervention, such as occur in Buddhism and Jainism” (Hiltebeitel 1988a: 99). On this type of contrast, I was then as I am now indebted to the inspired work of Randy Kloetzli on Buddhist, Hindu, and Jain cosmologies (Kloetzli 1983; 1985; 1987). The important differences concern the avatåra and the yuga, which thus points the way to Biardeau’s discussion of references to Buddhism in Mårkañ∂eya’s account of the Kaliyuga. Be it emphasized, however, that the avatåra and the yuga are both concepts in formation in the epics: the one not yet used as a substantive;37 the other used with great fluidity;38 and neither yet codified and ordered in the way they are to be in the puråñas. 35 Slightly

modified from the translation in Hiltebeitel 1988a: 95. as Kålanemi, see Hiltebeitel 1988a: 98. 37 As has been long and widely recognized. See Hiltebeitel 2001: 109, n. 56 –a note ovelooked even though it is back-referenced in a note on a page (236 n. 36) cited to criticize one of my approaches to this topic as ‘implausible’ on the sole stated grounds that I ‘show no recognition of the fact that the term [avatåra] is later than the epics’ (Brockington 2002: 601). 38 See Luis Gonzales-Reimann 2002. Also Hiltebeitel 2001: 152, n. 82 (I would stand by my point that Gonzales-Reimann’s atomization of yuga references is not successful, but add that his work is valuable in demonstating that uses of the yuga concept are fluid in the epic, and that a consistent or even overriding view of their applications is not easy to arrive at); Biardeau 2002: vol. 1, 1007-1008 (with a view different from mine on the prevaling yuga context of the Mahåbhårata war). 36 Such

124

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

4. The degradation of dharma in the Kaliyuga Just after Mårkañ∂eya has wound up his account of how he survived the pralaya or dissolution of the universe on the remaining single ocean and has revealed that Janårdana Kr¢ßña, who is sitting there with the Påñ∂avas listening, is none other than the primal god who not only sleeps on that ocean and wakes at creation but is the very god whom he, Mårkañ∂eya, met as a baby on a branch of a banian tree on that vast ocean, and who swallowed him then and there to bring him back to this world (Mahåbhårata, 3.186.77 ff), Yudhiß™hira asks Mårkañ∂eya to tell him next about ‘the future course of the world under imperial rule’ (så∫råjye bhavißyåm jagato gatim [Mahåbhårata, 3.188.3]) –or ‘under empire’– in the Kali age (Mahåbhårata, 3.188.5). I take some liberty in translating sa∫råjye by ‘under imperial rule’ here, but do so bearing three points in mind. First, there is nothing to support van Buitenen’s making Yudhiß™hira ask “about the future course of the world under his sovereignty” (Van Buitenen 1975: 593) –that is, under Yudhiß™hira’s own sovereignty or imperial rule. Second, that it is a question about future sa∫råjya in the abstract allows one to consider it as a back-reference to, or reminder of, Jaråsandha’s own past rule itself. As I indicate in Rethinking the Mahåbhårata, the epic “[…] construes the whole episode of Yudhiß™hira’s assertion of paramountcy through the elimination of his rival, the Magadha king Jaråsandha, and his performance of a Råjasüya sacrifice around the issue of empire”. This sequence provides in a flurry most of the Mahåbhårata’s usages of the terms sa∫råj, ‘emperor’, and så∫råjya, ‘empire’ –to which I note, “Within the Poona C[ritical] E[dition], there are, between 2.11 and 2.42 [that is, in the Jaråsandha-‡i†upåla sequence], eight out of the fourteen such usages in the entire epic”39 (Hiltebeitel 2001: 8 and n. 37). Thus, as regards Jaråsandha, “Once Yudhiß™hira learns from Nårada that he should consider the Råjasüya as a means to empire (så∫råjyam; 2.11.61), Kr¢ßña says that he has the qualities (guñas) to be emperor (sa∫råj) and to make himself emperor of the Kßatra (kßatre sa∫råjam åtmånam kartum arhasi; 2.13.60), but must first defeat Jaråsandha who has obtained empire by birth (så∫råjya∫ jaråsa∫dhaΔ pråpto bhavati yonitaΔ; 2.13.8)”40 (Hiltebeitel 2001: 8). Third, I do not follow Biardeau’s argument that the epics do not mean empire by the term sa∫råjya (Biardeau 2002: vol. 1, 299; 845; vol. 2, 771). Indeed, it would seem she would have to concede that they use the term to talk about it if she considers Jaråsandha a cryptic figure for A†oka (Hiltebeitel forthcoming). In any case, Mårkañ∂eya addresses this topic under the heading of what will happen ‘at the end’ or ‘with the destruction of the yuga’ 39 My

source for such epic word-counts was Tokunaga 1994. do not agree –as indicated in Hiltebeitel 2001: 8, n. 36– with those who argue that the whole Råjasüya episode is late. 40 I

III. How to Produce, Construct and Legitimate a Tradition

125

(yugånte, yugakßaye).41 Among these things to come, here are some of the passages that Biardeau finds most interesting, along with some others that bear on our concerns in this essay. I take them in their textual order. The Brahmans shall find fault with the Veda and abandon their vows; seduced [deceived]42 by argumentation/logic,43 they will neither offer worship nor sacrifice (na vratåni carißyanti bråhmañå vedanindakåΔ Ù na yakßyanti na hoßyanti hetuvådavilobhitåΔ). (Mahåbhårata, 3.188.26)

The Vulgate adds here the following line, which Biardeau emphasizes as a double reference to the key phrase: ‘Deceived by argumentation/logic, they will make effort in the lowlands’ (nimneßvœhå∫ karißyanti hetuvådavimohitåΔ). (Mahåbhårata, 3.948*)44 Biardeau glosses hetuvådavilobhitåΔ as “[…] led into error by discussions that pretend to be logical”.45 And she continues: But let us not fall asleep over this appeal to logical discussion that draws the Brahmans into error: discussions with whom? On what subject? Let us keep in mind these logical discussions that, this time, cannot situate themselves in the sacrificial arena during the pauses of the ritual process, but which can no longer be purely mythical inventions, for it is the first time that one hears them spoken of in a Brahmanical text. (Biardeau 2002: vol. 2, 760)

This notion of ‘pauses in the ritual’ –that is, the intervals in which debates could be held among Brahmans and Kßatriyas and (though she doesn’t mention it) stories could be told– is important in her contrast between Brahmanical and Buddhist theories of karman, the act: the Brahmanical act defined in relation to the rite and specifically to sacrifice; the Buddhist rite taken out of that context and explained in relation to impermanent aggregates of a moment, that is, dharmas (Biardeau 2002: vol. 2, 763, 768). Here we find Biardeau at her most pithy: “Faced with Brahmanical society [...], the Buddha had only several words to say about it: Brahmans are inferior to Kßatriyas (the 41 Note that this suggests for Yudhiß™hira, who goes on speaking in the future tense, that the Kaliyuga is an age yet to come. In what follows, Mårkañ∂eya distinguishes Kr¢ta etc. with imperfect or present tense verbs (Mahåbhårata, 3.188.10 ff) from Kaliyuga traits described by future tense ones (Mahåbhårata, 3.188.15 ff). 42 Where the Critical Edition has vilobhitåΔ here, the Vulgate has vimohitåΔ. 43 See Biardeau 2002: vol. 2, 760; 778. 44 Biardeau does not in this context draw attention to the mention of the ‘lowlands’ here, but elsewhere she notes that the marshy delta areas of Kaliõga-Orissa, classifiable as lowlands, are off-bounds to the Brahmans accompanying Arjuna (Biardeau 2002: vol. 1, 298, 516, 528, 530). This interpolated passage is found in the manuscripts B3, Dn1, n2, D6, and seems to have made its own connection between seduction by hetuvåda and the ‘lowlands’, which are mentioned in the next verse as a place where people commit malpractices in plowing (Mahåbhårata, 3.188.27) 45 Biardeau 2002: vol. 2, 760: “induits en erreur par des discussions prétenduement logiques”.

126

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Buddha being himself a Kßatriya) and their rites are perfectly useless” (Biardeau 2002: vol. 2, 768). In this vein and further along in Mårkañ∂eya’s prophesy, The entire world will be barbarized,46 without rites and sacrifices, without joy, and also without festivals (mlecchabhüta∫ jagatsarva∫ nißkriya∫ yaj∞avarjitam Ù bhavißyati nirånandam anutsavam atho tathå). (Mahåbhårata, 3.188.29)

It is worth recalling Gregory Alles’s attention to the ‘repressive’ centralization that marked imperial policies of the Mauryas, among whom A†oka “explicitly forbade popular religious assemblies” (Alles 1994: 65; also Hiltebeitel 2001: 16). Still further along these lines, and now most decisively, Biardeau brings up Mårkañ∂eya’s references to e∂ükas, the oldest term for Buddhist reliquaries and to begin with those for the bones of the Buddha after his cremation: a term found both in Sanskrit and Påli (where it occurs ‘eventually also under the form e¬üka’) and the early form of what are to become stüpas.47 The world will be totally upside down: people will abandon the gods and will offer püjå to [Buddhist] reliquaries; ‡üdras will refuse to serve the twiceborn at the collapse of the yuga (viparœta† ca loko ’yam bhavißyatyadharottaraΔ Ù e∂ükån püjayißyanti varjayißyanti devatåh Ù †üdråΔ paricarißyanti na dvijån yugasa∫kßaye). In the hermitages of the great R¢ßis, in the settlements of Brahmans, at the gods’ temples (devasthåneßu), in the Caitya sanctuaries, and in the abodes of the Någas, the Earth will be marked by [Buddhist] reliquaries and not adorned by houses of the gods. At the expiration of the yuga, that will be the mark of the yuga’s end (å†rameßu maharßœñåm bråhmañåvasatheßu ca Ù devasthåneßu caityeßu någånåm ålayeßu ca ÙÙ e∂ükacihnå pr¢thivœ na devagr¢habhüßitå Ù bhavißyati yuge kßœñe tad yugåntasya lakßañam). When men become ever-gruesome and lawless meat-eaters and liquor-drinkers, the yuga will collapse […]. Then the earth will soon be overrun by barbarians while Brahmans, out of fear of the tax burden, will flee to the ten directions (mahœ mlecchasamåkœrñå bhavißyati tato ’ciråt Ù karabhårabhayåd viprå bhajißyanti di†o da†a). (Mahåbhårata, 3.188.64-67, 70)

Note that such explicit reference to Brahmanical temples is rare in the Sanskrit epics.48 Yet more horrors are to follow (including the appearance of six suns [Mahåbhårata, 3.188.75]), until Mårkañ∂eya tells how the Kr¢ta age begins anew (Mahåbhårata, 3.188.87): 46 Mårkañ∂eya makes this a recurrent phrase; see Mahåbhårata, 3.188.37a (variant) and 45a. 47 Biardeau 2002: vol. 2, 759-760. The e∂üka references are among the “various Buddhist terms and concepts” that E.W. Hopkins thought made it “impossible to suppose that during the triumph of Buddhism such a poem [as the Mahåbhårata] could have been composed for the general public for which it was intended” (Hopkins 1969a: 399). See further Hopkins 1969a: 391 (with the e∂üka references); 475. 48 Although there is a R¢ßi named Devasthåna in the Mahåbhårata, this prophetic passage provides not only the sole instances of e∂üka in the epic, but those as well of devasthå-

III. How to Produce, Construct and Legitimate a Tradition

127

A Brahman by the name of Kalki Vißñuya†as will arise, prodded by time (kålapracoditaΔ), of great prowess, wisdom, and might. He will be born in the village of Sambhala in an auspicious Brahman dwelling, and at his mere thought all vehicles, weapons, warriors, arms, and coats of mail will wait on him. He will be king, a Turner of the Wheel, triumphant by the law, and he will bring this turbulent world to tranquility (sa dharmavijayœ råjå cakravartœ bhavißyati Ù sa cema∫ sa∫kula∫ lokam prasådam upaneßyati). That rising Brahman, blazing, ending the destruction (kßayåntakr¢d), noble minded, will be the destruction of all and the one who makes the yuga turn. Surrounded by Brahmans, that Brahman will extirpate the lowly Barbarian hosts (mlecchagañån) wherever they are.49 After destroying the robbers he will ritually make over this earth to the twiceborn at a great celebration of the horse sacrifice (tata† corakßaya∫ kr¢två dvijebhyaΔ pr¢thivœm imåm Ù våjimedhe mahåyajne vidhivat kalpayißyati). He will reestablish the auspicious limits that Svayambhu has ordained (maryådåΔ svayambhuvihitåΔ †ubhåΔ). And when he has grown old in works of holy fame, he will retire to the forest. People who live in the world will follow his morality (†œlam). And with the robbers (cora) destroyed by the Brahmans, safety will prevail. Establishing black antelope skins, spears, tridents, and emblematic arms in the conquered territories (de†eßu vijiteßu), that tiger-like Brahman Kalki, praised by the chief Brahmans and honoring their leaders, shall walk the earth forever bent upon slaughter of the Dasyus. The Dasyus will wail piteously, ‘Ah father, Ah son!’ as he leads them to destruction. Adharma will decline and dharma increase, Bhårata, and the people will observe the rites when the Kr¢ta age arrives (bhavißyati kr¢te pråpte kriyåvå∫† ca janas tathå). ņråmas (restingplaces), caityas (sanctuaries), temple tanks, wells, and the many ceremonies (kriyå† ca vividhå) will reappear in the K¢ta yuga. Brahmans will be strict, Munis will do tapas, hermitages [that were formerly filled] with heretics will be firm in truth; people will be subjects (å†ramåΔ sahapåßañ∂åΔ sthitåΔ satye janåΔ prajåΔ) […]. (Mahåbhårata, 3.188.89-3.189.9)

Here we have a number of terms that provide further reminders of an interface with a Buddhism compounded by mlecchification. Kalki, like Kr¢ßña, is a ‘wheel-turner’, but moreover a militant Brahman Cakravartin King who conquers by the dharma. His rule allows ‘conquered territories’ to be reconquered. He restores a ‘morality’ (†œla) of ‘auspicious limits ordained by Brahmå’ that would be nothing other than a morality ordained by the Veda, a Veda which, as we have noted, certain thieving demons, probably cryptoBuddhist, would have liked to hurl into hell at the moment of creation –according to the Mahåbhårata. In a world restored by Kalki, ‘robbers will be destroyed by the Brahmans’. I believe we must recognize in such a prophesy a greater antipathy, indeed, the projection of an ongoing antipathy, between the na and devagr¢ha in the sense of ‘temple’ (to which the Råmåyaña adds only one usage of devasthåna at 2.94.3) –as derived from Tokunaga 1994 (see n. 39 above). Since the Mahåbhårata rarely turns prophetic, and is normally portraying an heroic past, this is what one could expect in this passage. 49 An adhyåya break occurs here, resumed with ‘Vai†ampåyana said’ as the next adhyåya begins.

128

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Brahmans who composed the Mahåbhårata and their heterodox others, than Biardeau stresses. And I believe the feelings and strategies were probably reciprocal, and that counterpart expressions are not difficult to find in Buddhist texts. Nonetheless, I agree with Biardeau that the texts of these two main traditions have an answering relationship to each other, with bhakti as one of the their shared vocabularies –though certainly not the only one. More specifically, I would agree with her, as I do with my starting point from Jaråsandha’s daughters Asti and Pråpti, that it is pertinent to think of the Sarvåstivådins as one locus of Buddhist bhakti in formation that would have been familiar to the epic poets, along with the trends that would have led to the inclusion of the Mahåså∫ghikas as another early formation that included Buddhist bhakti-prone sects50 among the so-called eighteen Hœnåyana schools –counted (let me suggest) from a perspective familiar with both totalistic and oppositional senses of this epic number.51 I also like Biardeau’s formulation that bhakti is the epic-composing Brahmans’ ‘trump’ (atout), in that their bhakti is a bhakti of acts centered on relations between God and the king (Biardeau 2002: vol. 2, 763; 777).52 And I further concur that contending ideas about compassion (Biardeau 2002: vol. 2, 762), the gift,53 caste (Biardeau 2002: vol. 2, 773), and descent (Biardeau 2002: vol. 2, 776) are among the basic areas in which to explore this interface, to which I would add formulations about friendship,54 and the very basic shibboleths of what is a Bhagavat and what is a self –in the latter case, the very issue that would seem to be at stake in the choice of names like Asti, Pråpti, Ha∫sa, and ˘imbhaka. Indeed, if Biardeau is right about such twists, as I believe she is, 50 See Étienne Lamotte 1988: 526-527 (the prominence of both in Mathurå, and of the Mahåså∫ghikas in Magadha); 622-629 (‘Sarvåstivådin and Mahåså∫ghika Buddhology’); John Strong 1979: 221-237 (on bhakti in the Sarvåstivådin avadåna literature). Of course Theravåda texts must be studied as well (Bailey 2003), but as texts canonized in a more distant milieu. 51 I go further than Biardeau here; see her more cautious treatment of some of these factors (Biardeau 2002: vol. 2, 760-71). On the count of eighteen sects, see Lamotte 1988: 529, 533, 535, 547: a number ‘fixed by convention’ that is dated by Buddhist sources to the period from the second to third century after the Buddha’s parinirvåña, and related to the great schism at the time of A†oka. 52 Biardeau 2002: vol. 2, 750-752. Along with exchanges in the scientific, medical, astronomical, and artistic domains would be transformations in the religious domain, with the Mahåbhårata being the principal witness to such from the brahmanical side, while “the throne of the Buddha no longer remains empty” (Biardeau 2002: vol. 2, 750). I.e, these transformations can be read as countering positions on kingship. 53 Biardeau 2002: vol. 2, 763-764. For instance, Mårkañ∂eya prophesies that at the end of the yuga, ‘people will be delighted with gifts in name only [or perhaps ‘telltale gifts’?] even from the wicked’ (tat kathådånasamtuß™å duß™ånåm api månavåΔ) (Mahåbhårata, 3.188.31cd). 54 This is an old interest of mine, especially with reference to Karña; see Hiltebeitel, [1976] 1990: 254-266; 1982: 85-112; 1984: 1-26. See n. 27 above on the Upanißadic image of the two birds as reflected, according to Biardeau, in Ha∫sa and ˘imbhaka. As I hope to show in a future study, this whole cluster of themes –especially those of caste, friendship, and the gift– comes together in Karña.

III. How to Produce, Construct and Legitimate a Tradition

129

the Mahåbhårata’s answers to Buddhism are much more diverse and diffuse than than others have imagined.55 I believe that the Buddhist and Brahmanical texts, while preferring not to make it obvious, make such twists prominent in the ways they refine positions over and against ‘each other’. As Frits Staal put it during the discussion that followed the initial presentation of this paper,56 whereas everyday worshipers would seem to have moved rather easily between Brahmanical and Buddhist discourses and objects of devotion, it was the business of writers to make precise and subtle distinctions.57 From my perspective, which would stress polysemy, it is also less determinative to make one-to-one political or religious readings –such as a construction of Yudhiß™hira on A†oka, Yudhiß™hira on Dharmaråja Yama, or Jaråsandha on Måra or A†oka– than it is to see that political readings must be tied in further with the larger cosmological and soteriological idioms through which the texts sustain their narratives. From this perspective, all these one-to-one readings may have something persuasive about them without being as singly correct as their proposers propound. If Buddhism was in the air for the poets of the Mahåbhårata, they could very well have referenced it in multiple ways, and –as they do with many other classes of beings, including, as we have noted, Råkßasas– through figures allied with both sides of the epic’s main rivalry.

55 See my discussion of ‡aunaka’s instruction to Yudhiß™hira, upon the latter’s entering the forest, about an ‘eightfold path’ (aß™åõgenaiva mårgeña): from number one, ‘right binding of intention’ (sa∫yak sa∫kalpasambandhåt) through various Vedic and yogic ‘right’ (sa∫yak) procedures and proficiencies (numbers two through seven) to the eighth, ‘right stopping of thought’ (sa∫yak cittanirodhåt) (Mahåbhårata, 3.2.71-75d). Given what ‡aunaka calls Yudhiß™hira’s ‘eight-limbed awareness’ (aß™åõgam buddhim) (Mahåbhårata, 3.2.17), this path will allow him to conquer sa∫såra, into which one ‘falls [...] womb after womb, and is moved round like a wheel by ignorance (avidyå), karma, and thirst (tr¢ßñå)’ (67). I conclude: “Considering the echoes of Buddhist language here, it would seem that ‡aunaka’s instructions for the forest life combine a preemption and subversion of Buddhist teachings about forest enlightment and the eightfold path with a strongly Vedic interpretation (or anticipation?) of an eight-limbed yoga” (Hiltebeitel 2001: 172). 56 The initial draft of this paper was prepared for Professor Frits Staal’s Buddhism Class at the University of Leiden, the Netherlands, and presented November 6, 2003. I thank Professor Staal for his kind invitation to bring my thoughts together for that occasion on the topic that remains the lead title of this paper. 57 My italics, of course; and since apart from the quoted phrase this is a paraphrase, I allow myself to add the word ‘subtle’.

130

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

BIBLIOGRAPHY Alles, G.D. (1994), The Iliad, the Råmåyaña, and the Work of Religion: Failed Persuasion and Religious Mystifications, Pennsylvania State University Press, University Park, PA. Bailey, G. (2003), The Mahåbhårata as Counterpoint to the Påli Canon, paper presented at the Conference on Religions in the Indic Civilization, Center for Developing Societies, New Delhi, December 18-21. Biardeau, M. (1978), Études de mythologie hindoue: 5. Bhakti et avatåra, in «Bulletin de l’École Française d’Extrême Orient», 65. ______ (2002), Le Mahåbhårata: Un récit fondateur du brahmanisme et son interprétation, Seuil, Paris, 2 voll. Brockington, J. (2002), Review of ‘Rethinking the Mahåbhårata’, in «Bulletin of the School of Oriental and African Studies», vol. 65, n. 3. De Bruin, H.M. (1998), Karña Moksham - Karña’s Death: A play by Pukalentipulavar, École Française d’Extrême Orient, Institut Française de Pondichéry, International Institute of Asian Studies, Pondicherry. ______ (1999), Ka™™aikküttu: The Flexibility of a South Indian Theatre Tradition, in «Gonda Indological Studies», Egbert Forsten, Groningen, n. 7. Embree, A.T. (ed.) (19882), Sources of Indian Tradition, Columbia University Press, New York. Fitzgerald, J.L. (1980), The Mokßa Anthology and the Great Bhårata, Ph.D. dissertation, University of Chicago. ______ (2001), Making Yudhiß™hira the King: The Dialectics and the Politics of Violence in the Mahåbhårata, in «Rocznik Orientalistyczny», vol. 54, n. 1. ______ (2003), The Position of Brahmins in the Mahåbhårata: New Perspectives on the Development and Growth of the Epic Between the Empires, Between the Empires Conference, University of Texas at Austin. Gonzales-Reimann, L. (2002), The Mahåbhårata and the Yugas: India’s Great Epic Poem and the Hindu System of World Ages, Peter Lang, New York. Hiltebeitel, A. (1979), Kr¢ßña in the Mahåbhårata (A Bibliographical Essay), in «Annals of the Bhandarkar Oriental Institute», vol. 60. ______ (1982), Brothers, Friends, and Charioteers: Parallel Episodes in the Irish and Indian Epics, in Polomé, E.C. (ed.), Homage to Georges Dumézil: Journal of Indo-European Studies, Monograph, n. 3, pp. 85-112. ______ (1984), The Two Kr¢ßñas on One Chariot: Upanißadic Imagery and Epic Mythology, in «History of Religions», vol. 24, pp. 1-26. ______ (1988a), Kr¢ßña at Mathurå, in Srinivasan, D. (ed.), Mathurå: A

III. How to Produce, Construct and Legitimate a Tradition

131

Cultural Heritage, Manohar and American Institute of Indian Studies, New Delhi. ______ (1988b), The Cult of Draupadœ. Mythologies, From Gingee to Kurukßetra, University of Chicago Press, Chicago, vol. 1. ______ ([1976] 1990), The Ritual of Battle. Krishna in the Mahåbhårata, State University of New York Press, Albany. ______ (1999), Rethinking India’s Oral and Classical Epics: Draupadœ among Rajputs, Muslims, and Dalits, University of Chicago Press, Chicago. ______ (2001), Rethinking the Mahåbhårata: A Reader’s Guide to the Education of the Dharma King, University of Chicago Press, Chicago. ______ (fothcoming), Weighting Orality and Writing in the Sanskrit Epics, in Koskikallio, P. (ed.), Proceedings of the Third Dubrovnik International Conference on the Sanskrit Epics and Puråñas (DISCEP), Croatian Academy of Sciences and Arts, Zagreb. Hopkins, E.W. (1969a) [1901], The Great Epic of India: Its Character and Origin, Punthi Pustak, Calcutta. ______ (1969b) [1915], Epic Mythology, Biblo and Tannen, New York. Kloetzli, W.R. (1983), Buddhist Cosmology (From Single World System to Pure Land: Science and Theology in the Images of Motion and Light), Motilal Banarsidass, Delhi. ______ (1985), Maps of Time - Mythologies of Descent: Scientific Instruments and the Puråñic Cosmograph, in «History of Religions», vol. 25, n. 2. ______ (1987), Buddhist Cosmology and Hindu and Jain Cosmologies, in Eliade, M. (ed.), The Macmillan Encyclopedia of Religion, Macmillan Free Press, New York, vol. 4. Lamotte, E. (1988), History of Indian Buddhism, translated by Sara Webb-Boin, Institut Orientaliste, Louvain-la-Neuve. Sörenson, S. (1963) [1904], An Index to the Names in the Mahåbhårata, Motilal Banarsidass, Delhi. Strong, J. (1979), The Transforming Gift: An Analysis of Devotional Acts of Offering in Buddhist Avadåna Literature, in «History of Religions», vol. 18, n. 3. Sutton, N. (1997), A†oka and Yudhiß™hira: A Historical Setting for the Ideological Tensions of the Mahåbhårata, in «Religion», vol. 27, n. 4. ______ (2000), Religious Doctrines in the Mahåbhårata, Motilal Banarsidass, Delhi. Thapar, R. (1998), A†oka and the Decline of the Mauryas, Oxford University Press, Delhi. Tokunaga, M. (1994) [1991], Machine-Readable Text of the Mahaabhaarata Based on the Poona Critical Edition, First revised version, Kyoto. Van Buitenen, J.A.B. (1975), The Mahåbhårata, University of Chicago Press, Chicago, vol. 2.

LAURIE L. PATTON Trita’s Tumble and Agastya’s Ancestors: On The Narrative Construction of Dharma*

There are two major gifts which Wilhelm Halbfass has bestowed upon the Indologist who works outside of the strictly defined Indian philosophy and the six dar†anas as such. The first is a renewed interest in the intellectual value of word study, outside of the more narrow realm of etymology. Time and again his work has focused on the history of a particular word as a kind of micro-intellectual history, and with felicitous results. His essay on dharma in India and Europe,1 and his treatment of the word bhåva and abhåva in Being and What There Is,2 are two more of the salient examples of this kind of writing. In addition, Halbfass has cautioned us that the medium of cultural transition is as important as the content, no matter what period of Indian thought we are examining. In his epilogue to India and Europe, Halbfass wonders whether the exchange between the two cultures has determined not only the content, but the medium of the message. He is concerned here with both the vehicle and conditions of cultural transmission in this stormy, painful, and enriching relationship.3 The present article takes up these two major contributions –the intricacies of dharma and the medium in which cultural transmission occurs. In his ‘Dharma in the Self-Understanding of Traditional Hinduism’, mentioned above, Halbfass takes pains to note particular nuances of the term. For instance, he argues with Robert Lingat’s assertion that in the Mahåbhårata, dharma is dhåraña, ‘supporting’, * Sections of this paper have been published elsewhere. The discussion of Trita appears in chap. 8 of Patton 1996a. The discussion of Agastya and Lopåmudrå appears in Patton 1996b. Translations are my own, unless otherwise noted. 1 As Dharma in the Self-Understanding of Traditional Hinduism in Halbfass 1988: 310-344. 2 See Halbfass 1992. 3 See In Lieu of a Summary in Halbfass 1988: 441.

134

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

‘maintenance’, in the sense of cosmological, eternal law. Rather, it is a maintenance and preservation which is incumbent to man, particularly involved with ahi∫sa, or non-injury.4 The various meanings of dharma begin of course with Vedic meanings, including the physical connotation of ritual limit or boundary, as in R¢gveda, 10.90.16. There are also significant usages in the ‡atapathabråhmaña to the Taittirœyårañyaka, 10.29, which lists dharma as one of the principles of human order. Yet Halbfass argues that dharme pratiß™hitam tasmåd dharmam parama∫ vadanti is a stereotyped recurrent phrase, a difficult thing to understand as cosmic lawfulness and interpreted by Såyaña as maintaining or installing, including walls and water reservoirs. Even the development between a ‘maintaining’ duty, and a ‘cosmic law’, variations all throughout the Vedas, upanißads, and the Mahåbhårata, leaving out the classical period, shows the inner variety and flexibility of dharma, to use Halbfass’ words.5 The present essay takes up where Halbfass leaves off and focuses on the ways in which proper behavior of dharma is narrated. In other words, it considers, as Halbfass would want us to, the medium of cultural transmission. In particular, this essay examines the specific narrative factors and strategies which change some Vedic characters into exemplary Epic characters. Let us begin with a glimpse at those exemplary Epic characters. In Mahåbhårata 1.50.16, the r¢ßi Åstika praises the sacrifice of Janamejaya, saying that Janamejaya generates as much heat as the Vedic r¢ßi Trita. In Mahåbhårata 4.20.11, Bhœma exhorts Draupadœ to persevere, telling her of the dharmic behavior of the Vedic r¢ßi Lopåmudrå, who gives up all her finery to follow her husband Agastya into the forest. Throughout the epic, one sees Vedic figures being used as exemplars of dharma –indeed, used with such frequency that the stories of their lives seem to be assumed. The Epic itself tells many of them –giving the r¢ßis’ background, family pedigree and heroic exploits in colorful detail. Yet these narratives (called itihåsas and åkhyånas), are also told in the earlier Vedic material, with many intriguing similarities. There are striking parallels in content between Vedic and Epic narratives, as illustrated by the tales of the r¢ßis Trita, Lopåmudrå and Agastya, the birth and exploits of Dœrghåtamas, the birth of the r¢ßis Agastya and Vasiß™ha from a pot, the legend of Puråravas and Urva†œ, and others. As I have discussed elsewhere,6 much scholarly ink has been spilled on how to account for these parallels, and determine whether the Vedic tradition contained the ‘original’ or ‘prior’ collection of itihåsas, or whether the later Epic and Puråñic traditions developed their own narratives based on the ‘hints’ given in the Vedic hymns.7 4 See Halbfass 1988: 317. Cfr. Mahåbhårata, 7.49.49 ff; 12.10.10 ff. Also, see, Lingat 1967: 17 (in eEglish ed. 1973: 3). 5 See Halbfass 1988: 321. 6 See Patton 1993; Patton 1996a.

III. How to Produce, Construct and Legitimate a Tradition

135

1. Origins Beginning with the first blossoming of interest in the itihåsa/åkhyåna material in the mid-nineteenth century, most Indologists have studied itihåsas with a concern for empirical-historical evidence of Vedic sources. In the late nineteenth century, Hermann Oldenberg put forth his ‘åkhyåna-theory’, asserting that many of these later legends were the degenerated form of prose ‘frame narratives’ which originally connected the metrical part of the R¢gveda itself. Many Vedic scholars agreed with Oldenberg’s basic idea, and much of the next half-century was taken up with an analysis of individual legends in an attempt to show their specific function in framing the hymns of the R¢gveda to which they were attached.8 Not only did this theory prove an ‘original’ form of Vedic composition, it also engendered further debate about origins. Karl Geldner, Richard Pischel, and Emil Sieg,9 took issue with Oldenberg’s dismissal of the åkhyånas in later Vedic commentary as degenerations of an unrecoverable but nonetheless ‘purer’ form of åkhyåna. Preferring the term itihåsa to åkhyåna, they attempted instead to prove that the later stories were part of an uninterrupted chain of an itihåsa tradition –an idea which Oldenberg dismissed. They remained unified, however, in their view that such legends provided the ‘key’ to the hows and whys of the earliest form of Vedic composition. Not surprisingly, in the first part of the twentieth century scholars reacted against this theory about the origins of Vedic hymns, in all of 7 Ernest Windisch, in a study of Irish literature (Windisch 1879: 15 ff), made a conjecture that the song of Purüravas and Urva†œ might well be a poem detached from its narrative context. Hermann Oldenberg picked up on this possibility in his study of the Suparña legend, and later (Oldenberg 1883; 1885). Basing his theory on the literary style of the Påli Jåtaka tales, he argued that the dialogue hymns of the R¢gveda postulate a prose narrative as the connecting medium of the metrical parts. This resulted in a combination of prose and poetry, in which poetic verse was the high point of emotion. The older portion of the Mahåbhårata and later Vedic texts, such as th Suparña åkhyåna, also preserved this pattern. Yet Oldenberg saw the later itihåsas as useless historically, the fanciful creation of the commentators to describe and rationalize the hymns of the R¢gveda which were embarrassing or puzzling to them. Karl Geldner, in his important study of Purüravas and Urva†œ (Geldner 1889: 243-295), argued against Oldenberg’s denigration of later itihåsas. Following his teacher Geldner, Emil Sieg’s Die Sagenstoffe des ¥gveda Veda und die indische Itihåsa tradition (Sieg 1902) is still regarded as the most thorough treatment of Rgvedic legends available. 8 The ensuing twenty years resulted in specific studies of those particular stories for which Såyaña gives no ritual employment (viniyoga), such as the V®ßåkapi hymn (R¢gveda, 10.86), Saramå and the Pañis (R¢gveda, 10.108), the recovery of Agni (R¢gveda, 10.51-53), Mudgala’s race (R¢gveda, 10.102), the dialogue between Purüravas and Urva†œ (R¢gveda, 10.95), Lopåmudrå and Agastya (R¢gveda, 10.179), and Indra, the Maruts, and Agastya (R¢gveda, 10.165, 10.170, 10.171). Bloomfield treated the stories of Indra and Namuci, the two dogs of Yama, and again the marriage of Sarañyü, as well as Trita. See Bloomfield 1898; 1896. Albrecht Weber contributed an important study (Weber 1891), which focused on the ritual use of Vedic legends and the relationship between the legends in ritual literature and in the Mahåbhårata. 9 See Geldner 1889: 243-295. More generally, Geldner 1889; 1897; 1901. Also Sieg 1902.

136

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

its variations. The ‘anti-åkhyåna’ reaction believed that the R¢gveda hymns were coherent enough on their own –defined as some sort of cohesive dramatic scene,10 independent of any supporting prose apparatus;11 as full ritual dramas in their own right;12 or as epic poetry without the addition of prose.13 Whatever side one takes in this debate, it is clear that the question of origin was paramount. The force of the åkhyåna theory and its adherents postulates an ‘ur-principle’ of original stories, from which even the R¢gveda itself derived; it explains in the most elegant fashion even the most difficult of dialogical hymns. The anti-åkhyåna theory only substitutes a number of other ur-principles for that of the first scholarly hypothesis, whether it be that of ‘original coherence’, ritual drama, or a combination of these and other elements. To be sure, some scholarship of the last few decades has balanced these excessive concerns with origins. In an attempt to move away from the exclusively text-critical concerns, some authors, such as H.L. Hariyappa and S.A. Dange, concentrate on the telling of religious history through changes in themes and the treatment of characters in the various versions of these legends, focusing particular attention on the relationship between Epic, Puranic and Vedic sources.14 Yet even in his thoroughgoing recent treatment of itihåsa, Jan Gonda15 treats the subject solely from the perspective of origins. He pays considerable attention to whether the authors of such tales originally drew on an oral tradition connected to the R¢gveda, or on their own imagination. He also claims that some itihåsas which, upon first appearance, seem to be ‘suitable exegeses’ of a Vedic hymn, are when subject to further scrutiny found to be ‘concoctions’ which are ‘secondary’ to the rgvedic hymn itself. 10

See Levi 1963. In addition to Levi 1963, various other scholars took positions in various degrees of extremity against the åkhyåna theory. J. Hertel, working independently of Levi, rejected the åkhyåna theory in several articles. Moriz Winternitz tried to reconcile the idea by postulating that some of the sa∫våda hymns were ballads, in which everything is told in versified speeches, some were åkhyåna, poetic fragments which did have the non-surviving prose element, and some were strophes which belonged to ritual dramas. See Winternitz 1909. In a similar vein, A.B. Keith (1911), in one great sweep rejects both the åkhyåna and drama theory, stating that there is not enough evidence for either. The debate about the åkhyåna theory died down in the early 20’s, and the fourth period is marked by a notable silence on the subject of Vedic legends. Interestingly, in the later 60’s there do exist attempts to reconsider and reaffirm the åkhyåna theory. Ludwig Alsdorf (1964) attempted to vindicate the åkhyåna theory from the perspective of Jain literature. 12 See von Schroeder 1908. 13 See Charpentier 1922. 14 In his monumental work, H.L. Hariyappa (1953) began the trend by examining the evolution of several Vedic stories, especially those of the Rgvedic sages. The works of S.A. Dange (1969) and Robert Goldman (1969) are some other important examples of a more thematic treatment of these itihåsas. Paul Horsch (1966) attempts to give a thorough intellectual history of the åkhyåna theory as a subject of the larger problem of genre in Vedic literature. See also Ram Gopal 1969, and Mehta 1971. 15See Gonda 1975: 125-126. 11

III. How to Produce, Construct and Legitimate a Tradition

137

2. The Epic Connection Revisited: New Frames of Analysis Despite this concern with origins in the discussion of Veda and Mahåbhårata, new perspectives are emerging which have a slightly more complex frame of analysis: these approaches focus on both continuity and discontinuity between the two traditions over time, as Halbfass would want us to. In a recent article,16 Christopher Minkowski –following Michael Witzel–17 has shown the ways in which the pattern of embedding narratives within ritual contexts, hinted at by earlier Vedic texts such as the Bråhmañas, is born out by the frame tales of the Mahåbhårata. There, both Janamejaya’s and ‡aunaka’s sattras (extended ritual actions) provide the moments for narrative episodes. Although he does not explicitly discuss itihåsa or åkhyåna, Minkowski argues that this ritual setting for narrative is a crucial organizating principle in the development of Indian epic literature. One important implication of Minkowski’s work is that we can see new ways in which the ritual setting –itself paramount evidence for dharmic behavior– provides a frame for particular meditations upon dharma and the dharmic life. The Vedic ritual setting, then, provides the continuity for the slightly less Vedic, ‘dharmic’ world view presented in the Mahåbhårata. Relatedly, it would be fair to say that most of the Vedic characters in the Mahåbhårata provide a particular kind of continuity: as mentioned above, they are cast as the ‘early exemplars’ of dharmic virtue. But what of the discontinuities? Despite their status as Vedic exemplars, the Mahåbhårata tales told are re-workings of particular Vedic themes which are not presented in the same way in the earlier Vedic texts. How might we describe the difference in emphasis between Vedic and Epic versions of particular tales? In other words, what are the Mahåbhårata’s strategies of modifying and elaborating Vedic tale in order to teach dharma? I would like to suggest, through a discussion of two examples, that there are four main strategies: 1) the elaboration of a simpler Vedic presentation of a plot to include a more detailed familiar structure; 2) greater emphasis upon the implied crisis family coherence –a crisis usually caused by the act of an adharmic relative; 3) greater focus upon a single Vedic individual whose heroism, while present in the earlier versions of the tale, in the Mahåbhårata consists of the fact that he alone is able to make the family cohere once again; 4) greater stress being laid on that heroism being demonstrated not solely by 16

See Minkowski 1989. See Witzel 1987; 1986. Witzel historical arguments about the intent of the stories are indeed convincing, and inform the overall spirit of my own study. O’Flaherty (1985) has also treated the Jaiminœyabråhmaña from a more symbolic, psychoanalytic perspective. While less historically persuasive about the development and intent of the narratives, her interpretations of individual stories are useful for the purposes of symbolic analysis, and will be referred to on occasion. 17

138

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Vedic mantra, but by a singular act of will, usually in the form of an insight or mental feat that shows supreme knowledge of dharma. 3. Getting the Gods to Listen: Precursors to the Mahåbhårata Story of Trita Aside from being an exemplar of sacrificial heat, the story of Trita is told in some detail in the Mahåbhårata itself. Yet the Epic version has very specific Vedic antecedents. It harkens back to R¢gveda 1.105, a series of petitions to the gods by Trita, pleading with them, and more specifically, with Heaven and Earth, to ‘know me in such a state’. The hymn goes as follows:18 1. The moon shines in the water, the swift-moving one runs across the heavens. They do not find your location, you gold-wheeled glistening stars. Know me in such a state, O Heaven and Earth. 2. The ones who have a need, achieve their need. The woman pulles her husband toward herself. Both overflow in the courting wetness. She gives her juices and milks the husband. Know me in such a state, O Heaven and Earth. 3. You gods, may that sun not fall from heaven; may we never be without the healing Soma drink. Know me in such a state, O Heaven and Earth. 4. I ask about my most recent sacrificial offering. May the messenger Agni explain: to what end did my earlier good work come? Who owns it nowadays? Know me in such a state, O Heaven and Earth. 5. You gods who are there in the three light realms of heaven, what is justice for you; what is injustice? Where did the earlier sacrificial offering go for you? Know me in such a state, O Heaven and Earth. 6. What is eternal in you law? What is Varuña’s command? Can we, walking in the way of the great Aryaman, avoid the evil-wisher? Know me in such a state, O Heaven and Earth. 7. I am still the same one who talked earlier with Soma about this and that. Now sorrows befall me like a wolf torments the thirsting deer. Know me in such a state, O Heaven and Earth. 8. Bricks torment me like rival wives. O (Indra) of a hundred powers, oppressing cares devour me, your praiser, as mice the tails. Know me in such a state, O Heaven and Earth. 9. Those seven rays, my origin goes back this far. Trita Åptya knows that; he raises his voice in favor of the relatives. Know me in such a state, O Heaven and Earth. 10. Those five stars which were standing in the center of the high heavens –the gods be thanked– they are vanished altogether. Know me in such a state, O Heaven and Earth. 11. The swift-moving ones are sitting in the middle of the ladder to heaven. They are chasing the wolf away from the path, the wolf who comes from the youthful waters. Know me in such a state, O Heaven and Earth. 12. The praiseworthy one, the one to be lauded, is solidified anew, O gods. The rivers stream accordingly; the sun has truly spread himself all over. Know me in such a state, O Heaven and Earth. 18 Cfr. R¢gveda, 1.105.1-19. I have consulted the work of Karl Geldner (1923: vol. 1, 123126) for this translation.

III. How to Produce, Construct and Legitimate a Tradition

139

13. Agni, you have the praiseworthy acquaintance among the gods. Sit down here as if next to Manu and sacrifice as the most knowledgeable one to the gods. Know me in such a state, O Heaven and Earth. 14. As if sitting next to Manu as the invoker, invoke the gods to be present, you as most knowledgeable one. Agni makes the sacrifice delicious; the wise god among the gods. Know me in such a state, O Heaven and Earth. 15. Varuña enacts the words of praise; we address ourselves to him as the pathfinder. He unlocks the thought in the heart. Anew righteous [song] shall be made. Know me in such a state, O Heaven and Earth. 16. The way of the Ådityas, which is made –heavens be thanked– no one can step on it, you gods; the mortal ones don’t see it. Know me in such a state, O Heaven and Earth. 17. Trita in the well calls the gods for help. Br¢haspati heard that. The liberation out of the narrow [well] is achieved. Know me in such a state, O Heaven and Earth. 18. The red wolf had seen me once as I walked on my way. Since he noticed me, he erects himself like a carpenter with aching backbones. Know me in such a state, O Heaven and Earth. 19. Through the song of praise we want to keep the upper hand, with Indra on our side, with holy men in the struggle. May Mitra and Varuña grant us that. Know me in such a state, O Heaven and Earth.

In this hymn, Trita appears as a kind of Vedic version of the Biblical Job. In the first few verses, he seems to be reminding the gods that he is in a state where he needs to be ‘found’, and that he must, like others, achieve his needs. He entreats the sun to remain on its path (v. 3), and yet also reminds the gods of his sacrificial offering, and wonders what has happened to it (v. 4-5). He also muses upon the nature of the gods’ justice and injustice; wondering at his state since he is ‘still the same one’ who offered Soma to the gods, as all good sacrificers are supposed to (v. 6-7). Trita then laments of his being surrounded by bricks who torment him like rival wives (sapatnir iva par†avaΔ) (v. 8), and reminds them of his lofty origins (v. 9). He goes on to remark that the stars seem to have vanished altogether (v. 1011), and then goes back to praising the gods (v. 12). He especially invokes Agni, who has a special relationship with the gods (v. 13-14), and Varuña, who seems to put his words of praise to good effect (v. 15). He praises the hidden-ness of the gods’ path (v. 16 [na sa devå atikrama; tam martaso na pa†yatha]). In the final verses, the hymn switches to the third person narration, which describes Trita’s asking the gods for help (v. 17), then switches back to a first-person narration of an encounter with a red wolf (v. 18), and lastly, the triumphant celebration of holy men in the struggle, with Indra on their side (v. 19). All the while, in almost each refrain, he is pleading to be recognized: vittam me asya rodasi! The Nirukta19 version of the story behind the hymn (4.6) should provide an interesting contrast. Because the Nirukta is an etymologi19

See Sarup’s edition (1920-27).

140

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

cal dictionary from the fifth century B.C.E., it is not surprising that in that text Trita serves an entirely different function than implied by the Rgvedic narrative. His legend is alluded to only as it sheds light upon the meanings of particular words. The r¢ßi’s power, while mentioned, is not emphasized in terms of the actual mantras used to get him out of the sticky situation. Trita in the well quotes other verses in the Nirukta: Bricks torment me like rival wives. O (Indra) of a hundred powers, oppressing cares devour me, your praiser, as mice the tails.20 Know me in such a state, O Heaven and Earth. (R¢gveda, 1.105.8; 10.32.2. Cf. Br¢haddevatå, 7.34) Bricks, i.e. bricks of the well, torment me on every side, like rival wives. As mice devour the greasy tails (si†na). Si†na may mean one’s own limbs, i.e. they devour their own limbs, so oppressing cares torment me, the singer of your praises, O lord of a hundred powers […] […] Realize, O heaven and earth, this [state] of mine. This hymn was revealed to Trita when he had fallen into a well. With reference to this, there is an invocation, accompanied by a legend, a verse, and a gåtha. Trita was one most eminent in wisdom. Or else the word may have been intended as a synonym of a number –i.e. ekata, dvita, trita, and thus the three were produced.

The Nirukta does not tell us how or why Trita was one most eminent in wisdom. Moreover, it intends to explain only the obscure similes of the sükta, such as that of bricks and rival wives, and mice devouring their own tails. Dürga, the commentator on the Nirukta, explains the simile of the mice by commenting that it is the habit of a mouse to besmear its tail with grease and to lick it afterwards, just as some devour their own tails. The bricks thus consume Trita in a similar fashion. In an amusing set of verses, a slightly later commentary, the Br¢haddevatå (c. 400 B.C.E.),21 emphasizes the power of the r¢ßi Trita in several ways rather distinct from the Nirukta: The cruel sons of a she-wolf expelled Trita, who was following the cows, into a well, and then carried off all of them from there. There he presses Soma, that best knower of mantras among knowers of mantras, and summoned all the gods. Br¢haspati heard that. Now having seen them approaching, he reproached [them]: ‘where truly dwells the all-seeingness Varuña and of Aryaman? My limbs were wounded by the bricks of a well. Having seen all [the gods], I praise [them], although not one [of them] sees [me]’. The three troops of the all-gods, urged on by Br¢haspati, went to the sacrifice of Trita and took shares of it altogether. As a r¢ßi, Br¢haspati sang of that knowledge and wisdom of Trita here with the last triple of the sükta (R¢gveda, 1.105.16-18) [beginning,] ‘The way of the Ådityas, which is made –heavens be thanked– no one can step on it, you gods; the mortal ones don’t see it. Know me in such a state, O Heaven and Earth’. (Br¢haddevatå, 3.132-137) 20 According to Såyaña, †i†na can be translated as ‘threads’ or ‘tails’. Another prevalent meaning is ‘penis’ –which would yield a very different reading indeed. 21 See Br¢haddevatå, in Macdonell’s edition (1904).

III. How to Produce, Construct and Legitimate a Tradition

141

The r¢ßi Trita is shown in this story to be a canny sage in three distinct respects. In his anger, Trita brilliantly turns the phrase ‘having seen’ as well as the cliché of the ‘omniscient all-seeing gods’ against the gods themselves, playing upon it as Sophocles does in Oedipus Rex: ‘Where were you looking’, Trita says, ‘when I was looking for you?’ Second, in the Rgvedic sükta itself (R¢gveda, 1.105), there is no resolution whereby the gods come and take shares of the sacrifice all together, as the Br¢haddevatå suggests. While the last three verses of the hymn are about Trita, the Br¢haddevatå supplies the resolution of its own accord, thus showing Trita as a competent actor in his own right. Jaiminœyabråhmaña (1.184) contains a version of the Trita story which also emphasizes the actual words of the r¢ßi, and a god –this time Parjanya– coming to his aid. Yet, unlike the Br¢haddevatå, it does not concentrate upon the gods’ acquiescence to the r¢ßi, and their subsequent praise of him.22 Traveling with wealth in the forest the Åptyas became thirsty. They found a well in the desert. Dvita and Ekata were unable to climb down in the well. Trita climbed down in it. When the other two had drunk water and become refreshed, they went away with the cows after covering him with a chariot wheel in the well. He saw this såman. He praised with it. When he reached the nidhana [the concluding passage of a såman], ‘[…] With drops’, by means of heavy rain Parjanya flushed him up to the chariot wheel with which he was covered. This is the way-finding, protection-finding såman. Indeed, found the way, the protection. He followed [them] on foot. Having seen them, one of the two became a bear and the other a monkey, and they leaped into the forest. This is also a rival-killing såman. They whom he turned into a bear and a monkey who had indeed reached the state of rival. Now Parjanya becomes rainy for him. This såman is also for cattle. Indeed, he made the cattle his own. He who knows this keeps and becomes rich in cattle. Trita Åptya saw it, therefore it is called Traita-såman.23 22 Såyaña’s version is similar to the Bråhmaña’s versions: In olden times there were three seers named Ekata, Dvit, and Trita. Once being tormented by thirst in a jungle in the desert they found a well. Trita entered the well to drink water. Having drunk water, he brought water from the well for the other two. After drinking water they threw Trita into the well and covered it with a chariot wheel. Having robbed him of all his wealth, they went away. Being unable to come out of the well, Trita thought in his mind, ‘May all the gods rescue me’. Then he saw this hymn in praise of the gods. Seeing the rays of the moon in the well at night, he wails. 23 Translation of these passages with reference to Tokunaga 1979: 290-294, and O’Flaherty 1985: 54-56. Even later texts, Nœtima∞jarœ and Veõka™amådhava (ascribing to the ‡åtyåyañabråhmaña) give this version: “Traveling with wealth in the forest, the Åptyas became thirsty. They found a well in the desert. Dvita and Ekata were unable to climb down in the well. Trita climed down in it. When the other two had drunk water and become refreshed, they went away with the cows after covering him with a chariot wheel (in the well). He desired: ‘May I get out of here (ud itah iyam) and find a patron’. Fallen in the well he saw at night the rays of the moon”. Here, the details of the såman are not included in the story, and the tale ends ambiguously with a vague reference to the rays of the moon. Trita proves himself as a r¢ßi, but his brothers are not conspicuously punished, and Trita does not get his full revenge via the agency of a god.

142

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

In the Jaiminœyabråhmaña, the focus is upon the efficacy of the såman as a mantric rite for several purposes: finding one’s way, summoning rain, procuring cattle, as well as the killing of rivals. The animals at the end of the tale are the brother/rivals, punished for their behavior. 4. The Mahåbhårata’s Dharmic Elaborations All of the stories above serve to give us some background to the epic version of the tale. The Mahåbhårata legend (9.36) transforms the story into a familial story about the behavior of three sons in relationship to their father, and plays upon the Indo-European theme of the eventual pre-eminence of the youngest son. Here, as we shall see, all four of the strategies named above, used to achieve the ‘dharmicization’ of the Veda, are fulfilled. The story might be summarized as follows: In ancient times there were three sages named Ekat, Dvit, and Trita. Their father Gautama was highly respected by all the kings, and after his death his three sons were also held in high esteem. Among the three sons, Trita was the best and most respected like his father. Once, the three brothers got a lot of cattle-wealth by performing sacrifices for their rich patrons. When they were proceeding towards the east with the herd of cattle, Trita was moving in front and the other two brothers were at the rear. Having surveyed the vast herd of cattle, the two brothers at the rear conferred together: ‘How can both of us appropriate all these cows? Since Trita is an adept in sacrifice and learned in the Vedas, he will be able to earn many more cows. So let us decamp with the cows’. In the course of their journey at night a wolf suddenly appeared in their way. Having seen the wolf standing in his way, Trita ran away in panic and fell into a deep well not far away from the bank of the river Sarasvatœ. There he loudly cried for help within the hearing of his two brothers. They did hear his cries, but went away leaving him there to his fate on account of the wolf’s fear and their greed for cows. Deserted by his brother in a dry and dusty well, Trita apprehended death without having performed a Soma sacrifice, and thought: ‘How shall I drink Soma here?’ Thinking thus he by chance sighted a creeper there. Envisioning there himself as sacrificial priests in the presence of sacrificial fires, the creeper as Soma, pebbles as Somapressing stones, waters, and thinking of the relevant verses belonging to the R¢gveda, Yajurveda and Såmaveda, Trita mentally performed a Soma sacrifice and raised a loud voice which reached heaven and perturbed the gods. Br¢haspati heard his voice and said to the gods, ‘Trita’s sacrifice is going on. Let us all go there, otherwise he may forsake us in anger’. On arrival there all the gods partook of Trita’s offerings and granted him boons. Trita asked for the boons that they should rescue him from there, and that any one who performs ablutions in that well may attain the same heavenly reward as the performance of the Soma sacrifice. The river Sarasvatœ came there with a strong current and flushed him out of the well. Reaching home, Trita cursed his two brothers: ‘On account of your greed for cattle you deserted me! There fore due to my curse both of you shall become ferocious, fanged, and

III. How to Produce, Construct and Legitimate a Tradition

143

wandering animals assuming the forms of wolves, and your progeny shall become langurs, bears, and monkeys’. As soon as he pronounces his curse, both of them became so.24

Let us return to our four criteria outlined above. First, the familial emphasis is made paramount in the Mahåbhårata version. Neither the Nirukta not the Br¢haddevatå make such of the family connection, if they mention it at all. The Nirukta only mentions that they were drowned in a well; the Br¢haddevatå mentions that Trita had been expelled by the ‘sons of a she-wolf’, but neither go into the family connection in more detail. The Jaiminœyabråhmaña does make more of the brother relationship, but does not mention a father. The Mahåbhårata, on the other hand, portrays a full-fledged family with respected father Gautama (abhavad gautamo nityam pitå dharmarataΔ sadå [9.35.9]), three well-respected sons, all of whom behave dharmically by performing sacrifices and acquiring cattle (hence the sense of habitual sacrificing suggested by the text, with the terms pujayan and apujayan contrasted). Trita, of course, behaves the most dharmically, both in life and in the episode, and it is thus stated explicitly (trita saΔ †reß™hå∫ pråpa yathaivåsya pitå tathå [9.35.12]). Thus the first strategy, the expansion of the family scenario, is clearly in operation in the Mahåbhårata version. Notice too the elaboration of the question of crisis in the family –the second strategy mentioned above. First, the Mahåbhårata makes the emotions explicit, whereas the Bråhmaña and the Br¢haddevatå leave them to be inferred from the action. Note, in this mode, tad bhayåd apasarpan vai tasmin küpe papåta (9.35.15) to describe Trita’s actions. The text also includes the panic and fear that Trita feels. The crisis ‘stage’ is therefore more dramatically set. The sentences are explicitly added to focus on the different roles of the brothers, even their spatial relationships in the episode (mahåråjå puraståd åti hr¢ß™avat [9.35.19]). Moreover, we see the crisis caused by the adharmic brothers; the ones who assume that, because Trita is so skilled in sacrificing, he will be able to take care of himself (trito yaj∞eßu ku†alastrito vedeßu niß™hitah [9.35.22]). Moreover, the brothers are seen as greedy for cattle, both in the beginning of the story and at the end. (For example: katham na syur imå gåva åvåbhyåm vai vinå tritam [9.35.20]) In the earlier versions of the tale, no motivations of the brothers were given. Thus, crisis is expanded further than in the earlier versions; with the motivations of the brothers being given, we see the perfectly-manicured Vedic family has been ruptured at the roots, all over the adharmic greed for cattle. Third, in the Mahåbhårata version, we see a clearer representation of the heroism of Trita. The family coheres solely because he, the 24 Translated in O’Flaherty 1985: 54-56. The itihåsa quoted by Skandasvåmin in the beginning of his commentary on R¢gveda, 1.105 follows to a large extent the above quoted version of the Mahåbhårata.

144

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

dharmic one, can continue, whereas the others must be cursed, and the family cleansed of the negative elements. The Mahåbhårata reintroduces the Vedic wolf (vr¢ka) and Trita’s righteous curse to explain the presence of the bears and monkeys at the end (uvåca parußa∫ våkya∫ †a†åpa ca mahåtapåΔ Ù pa†ulubdhau yuvåm yasmånmå∫ utsr¢jya pradhåvitau [9.35.48]; and later, prasava†caiva yuvayorgolåõguœlarkßavånaråΔ [9.35.50]). The Jaiminœyabråhmaña only implies that the brothers became physically transformed into beasts. In addition to the more elaborately familial emphasis, the sacrifice of Trita becomes a mental one in the Mahåbhårata version, thus exemplifying the fourth elaborative strategy above. In the Mahåbhårata Trita not only recites a verse, but performs an imaginary Vedic sacrifice (for a lovely example: r¢ßi yaju∫ßi såmåni manaså cintayan muniΔ [9.35.33]). The focus of the Mahåbhårata does not carry the Nirukta’s emphasis on the phrases of the mantra, nor the Br¢haddevatå’s focus on the autonomous power of the individual words uttered by the r¢ßi Trita. Instead, it focused on Trita’s dharmic internal knowledge of sacrificial procedures –so correct, in fact, that the gods are fooled by it. And Trita’s curse, although powerful, is uttered again in the service of dharma, as a curse upon his bothers’ greed. A comparison of theses versions shows several significant changes in the Mahåbhårata. The Nirukta sees the story as an opportunity for the derivation of Vedic words. The point of the story in the Br¢haddevatå is to explain the Rgvedic verses, and by so doing, further emphasizes the close relationship between the gods and the r¢ßi. Just as Trita praises the gods, Br¢haspati in turn praises Trita in the end. The Bråhmaña also focuses upon the production of the mantra, and, characteristically, on the tension between human and divine. Yet in the Bråhmaña, the gods are often the more powerful, and thus they portray a helpless Trita who is ‘aided’ by the generous deities. In the Mahåbhårata, it is the family, the family crisis, and Trita’s heroic mental sacrifice that is eulogized. Note, too, that the simpler, dharmic act of performing ablutions in that well may help one attain the same heavenly reward as the performance of the Soma sacrifice. While the Bråhmaña and the Br¢haddevatå versions emphasize the ‘deeds’ which gave us the powerful mantra which we have today, in the Mahåbhårata, one hears a more didactic tone: ‘These are deeds which are great no matter how, when, where, or why […]’. The Bråhmaña and Br¢haddevatå versions give us a kind of etiological narrative, and the Mahåbhårata presents narrative for the purpose of teaching dharma. 5. Progeny and Celibacy: Precursors to the Mahåbhårata Story of Agastya and Lopåmudrå While the story of Trita deals with deadly sibling rivalry, the Mahåbhårata also deals with the rivalry between husband and wife. One of the more multi-leveled itihåsas in the Mahåbhårata is the story

III. How to Produce, Construct and Legitimate a Tradition

145

of Agastya and his wife, Lopåmudrå. The Mahåbhårata tale, to be discussed below, involves the plight of a r¢ßi to save his ancestors from condemnation, and the intricate relationship between husband and wife.25 In our exploring the Vedic hymn antecedents, let us begin, as we did with the story of Trita, with the Vedic hymn itself. Hymn 1.17926 of the R¢gveda is an intriguing negotiation between a r¢ßi torn between asceticism and his duty to produce progeny. 1. For many autumns [and for] age-producing dawns I have worked; day and night. Age distorts the glory of bodies; [thus] virile men should go to their wives. 2. Even those men of the past, who acted according to the sacred order, and discussed the sacred order with the gods, stopped [their asceticism] when they did not find the end. Wives should unite with virile men. 3. Not in vain is all of this work (†rånta∫) of asceticism, which the gods themselves encourage. Let us two take on all battles; and through this we will win the contest of a hundred ways when we two, as a pair, drive against [the enemies]. 4. Desire of my swelling reed overwhelms me, desire engulfing me from this side, that side, all sides. By this Soma which I have drunk in my innermost heart I say, ‘Let him forgive whatever wrong we have committed, for a mortal is full of many desires’. 5. Agastya, digging with spades, wishing for children, progeny, and strength, nourished both ways, for he was a powerful sage. He found fulfillment of his real hopes among the gods.

This Rgvedic himn opens within the context of asceticism (the fourteenth century commentator Såyana implies that both Lopåmudrå and her husband are engaging in ascetic practice). In the first three verses, Lopåmudrå is claiming an ideology that is quite clear; she describes herself as ‘toiling’ (†a†ramaña) –a word which could mean either her own work of asceticism, or her common labor. In either case, such work delays the basic fulfillment of their gender roles; Agastya is vr¢ßan, or fertile, bursting with seed, and should go to his wife. Lopåmudrå wishes to persuade her husband out of his asceticism, since even the archetypal man, the r¢ßis of old, did the same when they failed to become successful in their practices. Agastya thus argues with her that there is more than one way to achieve happiness, and implies that their union will be all that much happier if they remain in their separate, ascetic state now. Notice here that Agastya accepts her understanding of the respective duties of their gender.27 25 See Sårvånukramañœ on R¢gveda, 1.179; Såyaña on R¢gveda, 1.179.5. Nœtima∞jarœ quotes Br¢haddevatå, 4.57-60 on R¢gveda, 1.179. Oldenberg (1885) also quotes them (p. 68), and the whole story is treated in a portion of his article (pp. 65-68). Sieg (1902: 120-126) also discusses the various issues of this itihåsa. 26 As mentioned above, my translation of the hymn is in part inspired by Wendy Doniger O’Flaherty’s (1985). However, it departs from hers in several respects. 27 It is important to note here that it is ambiguous whether the contest requires one hundred stratagems by the people who are contesting, or whether the contest has one

146

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

The fourth verse has been attributed both to Lopåmudrå, to Agastya, and to the poet: Like the verse above, interpretations of this physically graphic verse also vary. Several Indian commentaries, one of which I will be discussing below, suggest that Agastya is the speaker.28 Thus, one should read nadasya må rudhataΔ as ‘[desire] of my swelling reed’. The verse would then be Agastya’s acquiescence to his wife’s voraciousness. However, other interpreters29 believe that Lopåmudrå speaks the verse. In the mouth of Lopåmudrå, this verse would read, ‘Desire has overwhelmed me for the bull who roars’, with a pun on the phrase, nadasya rudhataΔ. Translated in this way, the phrase would show Lopåmudrå’s ultimate triumph over his will. Although the details are not central to our argument here, Paul Thieme has provided persuasive evidence that the verse should be read as Agastya’s.30 Read in this way, it shows the implications of the valiant, heroic struggler who is abandoned by his partner to the call of desire. Instead of initiating sex, he first rejects, then submits to the will of Lopåmudrå. Lopåmudrå acts as the less valiant warrior, the persuasive agent of desire who, in keeping with her character, has left herself aging and questioned the value of celibacy. A similar interpretive dilemma faces the reader of the fifth verse. Some interpreters have proposed that Agastya speaks this verse, and others, to be discussed below,31 have argued that a brahmacårin (celibate) student of Agastya is the speaker. If the verse is Agastya’s, it is a kind of acquiescence to the husbandly role that he should play –a role that is in direct conflict with his asceticism. Therefore he must drink Soma in order to replace the semen that he has lost. hundred stratagems of its own. Paul Thieme (1971), in a persuasive interpretation of the hymn, chooses the latter reading. In a psychologically realistic portrayal of the struggle of the pair, he has suggested that the hymn compares the battle against temptation that celibacy involves to a chess game (or even a real battle) with changing situations, surprise attacks and defenses: “Mann konnte geradezu auf Gedanken kommen, dass der Dichter bei seinem Bild vom ‘Streit der hundert Strategeme’ an ein Brettspiel wie das Schach denkt, bei dem es sich um ein lang augedehntes Ringen mit wechselnden Situationen, mit berraschenden Angriffen und Verteidigungen handelt […]”. Thieme also compares this hymn with other dialogue hymns, such as that of Saramå and the Pañis (R¢gveda, 3.33), and Vi†våmitra’s conversation with the rivers (R¢gveda, 10.108). 28 See Sårvånukramañœ and Såyaña on R¢gveda, 1.179; also Br¢haddevatå, 4.57-61, to be discussed below. 29 See Nirukta, 5.2; Sieg 1902: 120 ff; Geldner 1923: vol 1, 232; Hillebrandt 1980: 136 ff; Oldenberg 1988: 65 ff. 30 Siding with the Indian tradition, Paul Thieme (1971: 209) again provides several more persuasive arguments that the verse is Agastya’s. First, if this verse is given to Agastya, the hymn is a more symmetrical dialogue where each speaker utters two verses. Second, it would be unlikely that Agastya would end with verse three, an explicit rejection of Lopåmudrå’s advances. More importantly, since Lopåmudrå has already articulated her desire in verses one and two, it makes no sense for her to claim that desire has suddenly overtake her, from an indeterminate place. It is only Agastya who struggles against desire, and would thus be overtaken so suddenly by it –victim of a ‘sudden attack’ in the contest of a hundred ways. 31 Såyaña on R¢gveda, 1.179.5; Sårvånukramañœ, 1.12; Br¢haddevatå, 4.58-60.

III. How to Produce, Construct and Legitimate a Tradition

147

Finally, it is agreed by all that in the sixth verse, the poet ends the narration: ‘Agastya, digging with spades, wishing for children, progeny, and strength, was nourished both ways, for he was a powerful sage. He found fulfillment of his real hopes among the gods’ (ubhau varñåv r¢ßir ugraΔ pupoßa; satyå deveßu å†iso jagåma). Thus, Agastya has achieved immortality through children and through asceticism. Turning again to the Br¢haddevatå, we see that the commentary provides an intriguing twist. It presents the hymn as follows: The r¢ßi, through desire for a secret intercourse, began to chat up (upa jalp) [his] wife, the illustrious Lopåmudrå, when she had bathed after menstruation. With two verses, ‘For many autumns [and for] age-producing dawns I have worked; day and night. Age distorts the glory of bodies; [thus] virile men should go to their wives. Even those men of the past, who acted according to the sacred order, and discussed the sacred order with the gods, stopped [their asceticism] when they did not find the end. Wives should unite with virile men’. (R¢gveda, 1.179.1-2) she declared her intention. Then Agastya, lusting, pleased her with the two following verses: ‘Not in vain is all of this work (srantam0 of asceticism, which the gods themselves encourage. Let us two take on all battles; and through this we will win the contest of a hundred ways when we two, as a pair, drive against [the enemies]. Desire of my swelling reed overwhelms me, desire engulfing me from this side, that side, all sides’. (R¢gveda, 1.179.3-4) The student, having known by tapas of the existence of the two lusting [after each other], [thinking] ‘I have committed a sin [by] having listened’, sang he last two verses; By this Soma which I have drunk in my innermost heart I say, ‘Let him forgive whatever wrong we have committed, for a mortal is full of many desires. Agastya, digging with spades, wishing for children, progeny, and strength, nourished both ways, for he was a powerful sage. He found fulfillment of his real hopes among the gods’. (R¢gveda, 1.179.56) The teacher and his wife praising and embracing him, kissing him on the head; smiling they both said to him, ‘You are without fault, son’. (Br¢haddevatå, 4.57-61).

A close reading of this passage reveals the addition of distinct elements, such as the embarrassed brahmacårin who utters the apology, not Agasya. Note too, that both the element of the couples’ embrace of the embarrassed brahmacårin and the words of ‘forgiveness’ from the two lovemakers (‘you are without fault, son’ [anåga asi putraka]) are not derived from the Rgvedic hymn at all, but supplied by the Br¢haddevatå. In contrast to the hymn itself, in the Br¢haddevatå the imbalance of desire is not a result of the desire, or excess of the r¢ßi himself –but that of his student. Thus, the mantra takes away the ‘wrong’ caused not by the r¢ßi, but by the outside character of the brahmacårin, who feels that he has committed a sin by having listened (†rutvainaΔ kr¢tavån asmi). Through a very deft use of exegetical techniques, so important in a time when the maintenance of brahminical norms was paramount, the author of the Br¢haddevatå ‘cleans up’ the offensively free sexual dialogues in an effort to make them acceptable. As would be expected, the Br¢haddevatå gives a fourth century BCE brahmin’s version of

148

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

what the female r¢ßis in the R¢gveda are supposed to say, in addition to what the R¢gveda actually says. Moreover, the manipulation of the character of the brahmacårin, discussed above, is evident. As can be seen from R¢gveda 1.179, there is nothing in any of the verses about a student –or about a conversation– being overheard at all.32 To be more specific here: by telling the story the way it does, the text omits a certain possibility. If the hymn is taken as a dialogue, then verse five, which the Br¢haddevatå argues is uttered by the brahmacårin, could in fact also be uttered by Agastya, as Geldner suggests. Yet if that were the case, then Agastya would be explicitly asking for pardon in this verse. Because the r¢ßi has committed the sin of excess of lust in the hymn, he would then be asking pardon, or, to be more accurate, would be replenishing his own virility, via Soma. This is not, however, the appropriate conduct for a r¢ßi according to 1st Century BCE brahminical norms. The Br¢haddevatå, being a product of these norms, thus does not entertain the possibility of Agastya’s apology. If one were to consider the idea that the r¢ßi was apologizing (or at least, as the Sanskrit says, he needs to claim that he is anåga, ‘blameless’), then the r¢ßi would be less perfect. He would have a fault. On effect of the Br¢haddevatå story, then, is to augment the power of the r¢ßi and the gods, at the expense of the woman. To put the verses of apology in the mouth of a brahmacårin is an ingenious way of allowing the r¢ßi to remain virile, as r¢ßis should be, as well as without fault.33 32 See Sieg 1902: 125. Sieg has taken notice of this interpolation, and traces it to Nirukta, 5.2, where the Sanskrit nadasya må rudhata¶ kåma agan of R¢gveda, 1.170.4, is translated as ‘desire has come upon me for the self-controlled seer’. Nirukta interprets this ‘as one who is celibate (brahmacårin) and who has controlled himself with regard to procreation’, and attributes these words to Lopåmudrå. Sieg attributes the tradition which supplies a ‘brahmacårin student overhearing’ to a ‘misunderstanding’ of the word brahmacårin, used by Yåska to explain the verse. He writes, in a delightful inversion of this present perspective, “How does it happen, above all things, that the tradition brings a schoolboy into the drama? Does there really appear here a reconstruction, which the old interpreters make for their justification? I think not. I think, on the contrary, that we have here proof for a real tradition. Surely this tradition has been misunderstood. The word brahmacårin in the old itihåsa, must have existed. That indeed, as we saw, gives the key to the entire song. This word was then mistakenly taken in the special sense of ‘Brahman student’. […] Through that however, the whole itihåsa had to become incomprehensible, and we have in our case in question, only an example where the native interpreters attempted to bring comprehension to that which was incomprehensible” (Wie kommt es vor allen Dingen, dass die Tradition noch einen Schuler in die Handlung hineinzieht? Liegt hier wirklich nur eine Rekonstruktion vor, welche die alten Erklärer auf ihre Verantwortung machen? Ich glaue nicht. Ich glaube im Gegenteil, dass wir hier wieder einen Beweis fr eine wirkliche Tradition haben. Allerdings ist diese Tradition missverstanden worden. In dem alten itihåsa muss namlich das Wort brahmacårin gestanden haben, das uns ja, wie wir sahen, den Schlssel fr das ganze Lied giebt. Dieses Wort wurde nun missverständlich in dem speziellen Sinn ‘Brahmanenschuler’ gedeutet […] Dadurch musste aber der ganze itihåsa unverständlich werden, und wir haben, in dem uns vorliegenden Falle nur ein Beispiel dafr, wie die einheimisschen Interpreten Verstand in das fur sie Unverständliche zu bringen suchten). 33 The Br¢haddevatå could be simply following Indian tradition in claiming that the verse was uttered by the brahmacårin. As mentioned above, the commentary of the Sårvånukramañœ, a text probably earlier than the Br¢haddevatå, introduces a brahmacårin as the author of verse five of the hymn. However, as Thieme argues, the earlier text does not

III. How to Produce, Construct and Legitimate a Tradition

149

In the story of the Br¢haddevatå, then, the roles of Agastya and Lopåmudrå become even more circumscribed; by placing the verse of atonement in the mouth of the brahmacårin, the scribe achieves a certain effect: he keeps Agastya’s reputation intact, and thus he implicitly maintains Agastya’s hierarchical relationship over Lopåmudrå and the student. Lopåmudrå is caught in the middle, however. She is the lustful woman who is contrasted to Agastya, who attempts to resist her. She is also contrasted to the Agastya-to-be –the brahmacårin. For him, her sheer presence, let alone her sexually explicit words, is taboo. Lopåmudrå is no longer simply the negotiator of sexual and gender roles, but is surrounded by males who are forbidden from any interaction with her.34 6. The Mahåbhårata Elaborations In the Mahåbhårata, 3(33)94-97 the tale of Agastya and Lopåmudrå is significantly more elaborate. While the entire tale covers several pages of the Årañyakaparvan, I will summarize the important elements here. Agastya discovers a number of his ancestors hanging upside down in a cave; when he asks them why they have been given this fate instead of progressing to heaven, the ancestors say that they are waiting for future progeny to be born to perform the appropriate rites to release the from a curse. It is thus the necessarily imply that the brahmacårin was Agastya’s student, or that he was living in the house with the couple at the time of the lustful conversation. However, Thieme (1971: 210211) again proposes the most reasonable solution to the problem. The last two verses are thus spoken as a kind of response to the mise en scène that has been created in the first four verses. They are verses of atonement which are spoken by the brahmacårin poet. Building on the wording of one commentary (Sårvånukramañœ, 1.12), Thieme argues that the brahmacårin does not need to be the student of Agastya living in the same house as his teacher, but simply a brahmacårin who, after learning the first four verses, saw the last two as a kind of revelatory incarnation which would counteract the breaking of celibate vows. This view is supported by the fact that the consumption of Soma accompanied by a såman, or chant, is used elsewhere in the R¢gveda as a method of atonement, as Såmavidhånabråhmaña, 1.7.9 suggests. Moreover, there is further evidence for the introduction of a new speaker in the fact that the meter changes from triß™ubh in verses one to four, to br¢hatœ in verse five. As Thieme also notes of R¢gveda, 3.33.13 and 10.108.11, the changes of meter implies that a new speaker has entered the scene. Thus, the first four verses act as a kind of artistic portrait of a kind of human truth –that celibate people break their vows, and must atone for the rupture. The fifth verse acts as an incantation that counteracts the effects of such a wrong, and could be used by later brahmacårins as a kind of ritual of self-absolution after a sexual transgression. As Stephanie Jamison (correspondence, 1994) suggests, this ritual use formulated by Thieme is also supported by the fact that verse five begins with a strong deictic imam. Such a linguistic construction suggests a ‘here-and-now’ kind of situation which would be used in ritual performance. 34 It is worth adding a note that the embarrassment at the couple’s lovemaking is not only that of the Br¢haddevatå’s brahmacårin. One nineteenth century French Indologist, Abel Bergaigne, gives the hymn a ‘mystical interpretation’, similar to that given by Biblical exegetes to the Song of Solomon. For Bergaigne (1878-1883: vol. 2, 394), Agastya is the god Soma, and Lopåmudrå, in a ‘prayer’, draws him down from his dwelling place. His approach is perhaps a subtle reminder that modern as well as ancient interpretations of uncomfortable material can be just as invested, if not more so, than Indian ones.

150

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

sage Agastya’s responsibility to produce the appropriate progeny in order to overcome the plight of his ancestors (3[33]94.11-15). He fashions for himself a woman whose limbs are matchless, and gives her to the king of Vidarbha in order to keep him safe for himself (3[33]94.16-28). After Lopåmudrå has grown up as a princess, King Vidarbha gives the full grown Lopåmudrå to Agastya for fear of his terrible ascetic’s curse. Lopåmudrå marries Agastya only to be ordered to throw away her finery and retire with him to the forest to practice austerities. Having proved, in a fashion similar to Pårvatœ, that she is his equal in her powers of asceticism, or tapas, Agastya approaches her for intercourse (3[33]95.1-14). Instead of giving in to his demands, Lopåmudrå manipulates Agastya into giving her a significant amount of wealth. Just as he is about to approach her, she demands that before they make love she must be given all the wealth that she has been accustomed to as a princess. In the stereotype of the manipulative wife, she demands things like the bed which she was accustomed to in her father’s house, and jewels to her taste. Agastya must go and find this before she will give in to him (3[33]95.15-25). Agastya searches the realms of Kings ‡rutarvan, Vadhrya†va, and Trasadasyu, and concludes that the taking of income from these realms would mean hardship for the subjects (3[33]96.1-20). After searching in these three kingdoms, Agastya then goes to the kingdom of the Asura Ilvala, and is presented with the brother of Ilvala, Våtåpi, cooked on a plate as a test. The great sage Agastya eats him, and digests him successfully, and thus is given the wealth he needs. As a result of this demonstration of his power in both the secular and the sacred realms, Agastya is allowed into Lopåmudrå’s bedroom and, as a result of their union, Dr¢∂hasyu, is born –a seer of tremendous power, who knows the Vedas in the womb and comes into the world reciting them (3[33]97.1-25). Let us examine this Epic story in terms of the Mahåbhårata’s strategies of elaboration outlined above –the narrative medium in which dharma is exemplified. First, the elaboration of the family structure in the Mahåbhårata version is clearly evident. Neither the Br¢haddevatå nor the R¢gveda itself, mention the ancestors as hanging upside down in a cave, their plight in urgent need of amelioration (3.84.12 [pitr¢ün dadar†a garte vai lambhamånånadhomukhån so ’pr¢cchallambamåna∫stånbhavanta iha ki∫paråΔ]). Neither earlier text mentions the complex relationship between Agastya and his father-in-law, the king who acquiesces to Lopåmudrå’s safe-keeping for fear of the r¢ßi’s terrible curse (3[33]95.4 [mahår¢ßivœryavåneßa kruddhaΔ †åpågninå dahet]). While the Br¢haddevatå does add the family-like figure of the embarrassed brahmacårin, the complexities of in-laws, lineage (whether instrumental or familial para∫parås) are simply not explored in the earlier texts. Moreover, the Mahåbhårata version adds a familial coloring in two more significant ways. First, the couple’s lustful lovemak-

III. How to Produce, Construct and Legitimate a Tradition

151

ing is not presented for its own sake, as it was in both the R¢gveda and the Br¢haddevatå, but instead becomes the occasion for Lopåmudrå to demand of Agastya that he provide for her as she was provided for in her father’s house. Thus, although Lopåmudrå acts as a negotiator here, her action is not an expression of her sexual desire (as she puts it, in her negotiations in 3[33]95.23 [etat tu me yathåkåmam sampådayitum arhasi]). Second, the Mahåbharatå’s familial relationships have the added dimension of negotiation between different classes, or varñas. In her demand, Lopåmudrå sets up the contrast not between r¢ßi and brahmacårin, as the Br¢haddevatå does, but between king and sage, testing Agastya to see of he can match the wealth of a kßatriya (as she subtly puts it in 3[33]95.20 [ar¢ßo ’si tapaså sarva∫ samårhatum e†vara Ù kßeñena jœvaloke yad vasu ki∫cana vidyate]). Secondly, we see clearly the context of crisis that is emphasized in the Mahåbhårata –evidence for the second elaborative strategy. In the earlier texts, the crisis is far more simple: it is about sexual temptation, and Agastya’s attempt to resolve it (R¢gveda), or the student’s attempt to excuse himself from the embarrassment of overhearing their lustful conversation (Br¢haddevatå). Yet in the Mahåbhårata version, the issue is not only about the production of progeny through sex, as it is in the earlier texts, it is about the freeing of earlier progeny through asceticism. The tension is not between the celibate and the householder, it is between the tending to and the neglect of the ancestors. Thus, the issue of lineage, the coherence of the line, becomes paramount. This perversion of family coherence is underscored by the image of the ‘up-side-down’ relatives in the cave. Relatedly, once again the Mahåbhårata version focuses on the heroism of a single individual –Agastya– thus demonstrating the third elaborative strategy outlined above. Agastya is clearly the one whose actions matter, and who alone can make the family lineage cohere once again by freeing his relatives. Agastya is the one who ‘discovers’ the plight of his ancestors, and they explicitly state that they are waiting for someone like him to release them from their curse (9.34.13 [te tasmai kathayåmåsurvaya∫ te pitaraΔ svakåΔ Ù gartam evam anupråptå lambåmaΔ prasavårthinaΔ]). Moreover, the entire plot of the tale revolves around Agastya’s attempt to solve this crisis, and overcome the acts of various adharmic, or less dharmic activities of the kings around him. His father-in-law, who spoils Lopåmudrå, and as the texts indicates wonders in his mind how he will ever give her away (3[33]94.27 [manaså cintayåmåsa kasmai dadyå∫ sutåmiti]). So too, the king Asura Ilvala tests Agastya in treacherous ways. At one point in the narrative, as if to underscore the difficulty, Våtåpi is described as ‘sa∫skr¢ta∫’ –which implies that the demon is both cultured, as most demons are, and ‘well cooked’. Finally, the fourth elaborative strategy –the mental feat– is exemplified in the creation of Lopåmudrå herself. Unlike the

152

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

R¢gveda and the Br¢haddevatå, the mantras are not even mentioned in the Mahåbhårata, but rather Lopåmudrå herself becomes the achievement of the r¢ßi’s concentration. Like a mantra is created in earlier texts, she is entirely Agastya’s creation; she is the mental ‘family’ that will save the ‘real’ family in the cave. Moreover, through this ‘cardboard cutout’ Lopåmudrå, a further dharmic element is added to Agastya’s personality; he not only creates progeny, and is a truly powerful ascetic, but he can also garner wealth like the best of kings. He does this not through the polluted processes that kings employ, but through his vision and guile, as a bråhmaña would. The desire for progeny is also no longer Lopåmudrå’s complaint, as it was in the Vedic genre, but rather Agastya’s altruistic act for his ancestors. And the progeny itself, Dr¢∂hasyu, is himself performing of mental feats (såñopanißadånvedå∞japanneva mahåya†åΔ) which will continue the legacy. 7. Final Thoughts The Mahåbhårata’s embroidery of the stories of Trita and Agastya are thus strikingly similar. Both Trita and Agastya are responsible for far more complicated families –Trita’s father and brothers, and Agastya’s father-in-law and ancestors. Both are beset by the crisis of adharmic relatives –Trita’s cruel brothers’ greed for cattle, Agastya’s predecessors’ negligence of their duties to the ancestors. Both are heroes concerned with family coherence –Trita’s continuity of the family line while his brothers are turned to animals, and Agastya’s reestablishment of his upside-down relatives. Both are capable of extraordinary cerebral feats to achieve their ends –Trita’s mental sacrifice and Agastya’s mental wife. The stories of Trita and Agastya are but two of the many possible examples of the Mahåbhårata’s four-fold mode of expansion on Vedic materials. One could explore the stories of Dœrghåtamas, or Purüravas and Urva†œ, and come up with similar kinds of changes rung into the earlier Vedic narratives.35 While space does not permit such a further exploration here, it is important to emphasis that this kind of analysis differs from its predecessors. It does not emphasize the question of origins, as in the earlier debates about Veda and Epic åkhyånas. Rather, it provides a set of narrative parallels to the points that Christopher Minkowski has made about ritual: just as Vedic ritual settings furnish the frames for stories about dharma, so too Vedic characters within the narratives furnish the vehicles for the expansion and exemplification of dharmic behavior. They are the media through which changes of dharma take place. In this sense, Trita and Lopåmudrå are not simply anachronisms, but brilliantly crafted archaic vehicles for the elaboration of Epic purposes. 35

See my discussion of these stories in Myth as Argument (1996a), Chapters 9, 14, 15.

III. How to Produce, Construct and Legitimate a Tradition

153

There is something perhaps even more significant to be learned here: narrative form may not simply be a vehicle, but in its own right may be part of the content of the Mahåbhårata’s dharmic message. Characters such as Trita or Agastya, and plots such as those outlined above, can be viewed as ‘formal properties’ of the narrative. It stands to follow that the ways in which Trita and Agastya enact dharma, in both the specifics of their character as well as the plots in which they are mapped, are thus also part of the meaning of dharma. Emily Hudson, following Martha Nussbaum, remarks that the form of the Mahåbhårata as much as its content may well be worth investigating for an idea of the epic’s moral teaching –and that, to date, such an investigation has indeed been wanting.36 Narrative form also tells the reader or listener how to live a life –one of Nussbaum’s great contributions. However, recent studies on dharma in the Mahåbhårata have tended (understandably) to focus on philosophical assessments rather than literary ones.37 Elaine Scarry puts the same point in another way. In her recent work, Dreaming by the Book,38 she argues that narratives are forever giving instructions to the reader on how to imagine, and describes the global features of these instructions that help us to imagine. For instance, Scarry emphasizes that repetition can lend to imaginary fullness: a friend smiling in the mirror, and then smiling again in full face, can give richness and vivacity to the image in a way a single description of the smile cannot.39 So, too, repetition might well be part of the same imaginative technique used by the authors of the Mahåbhårata. We see repetition within the text: for example, the anger that Trita feels is repeated in different narrative contexts throughout the Mahåbhårata’s version of his story. Repetition also works across the texts: a character is first known in less elaborate detail, and then is more fully fleshed out in later renditions in later texts. Trita as a ‘character repeated’ only gives the audience a deeper sense of the nature of the character. The refusal to introduce much about any of these characters also gives them a certain assumed, familiar quality so that, if one hasn’t heard of them, one understands that other listeners have. And, yet at the same time, these elaborations are presented not as if they were additions, but always there. None of the dilemmas in which the Mahåbhårata characters find themselves would seem odd to the listeners; rather, they would seem familiar because they are cast in a mode contemporary to them. 36

See Hudson forthcoming. See also Nussbaum 1992: 3-53; 168-194. See, among others, Bhattacharya 1992; Bailey, Brockington 2000; Sutton 2000; Hill 2001; Gonzales Reimann 2003. Alf Hiltebeitel’s Rethinking the Mahåbhårata (2001) is an exception in its broader attention to compositional issues in history. So, too Ayyappa Paniker’s very recent Indian Narratology (2003) has some important contributions via the epics and narrative style. 38 See Scarry 1999. 39 See Scarry 1999: 237-238. 37

154

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

This form of repetition may well be one of the compositional ways in which the mythical quality of the ‘always already given’ element, about which Hayden White writes so eloquently, is achieved. The ‘always already given’40 is technically achieved by the insertion of a well known character, presumed to be known, into a situation which is new to the character, but not new to the audience. Agastya’s ancestral situation, and his creation of Lopåmudrå, was not introduced in the Vedic dialogue between the two in the hymn, but would nonetheless be a recognizable dilemma to a second century BCE audience, concerned as it was with lineage and its maintenance. It would make sense, therefore, that Agastya would be the one to address the issue. So, too, the mental sacrifice of Trita was an innovation for him, but not for the audience which was by now used to the idea and practice of svadhyåya, or Vedic self study. These small thoughts may well be a beginning for longer studies in which we consider the ways in which the ‘mythical’ quality of the ‘always already present’ is actually achieved in texts we call mythical, such as the Mahåbhårata. And we can do so grammatically, compositionally, technically, without losing, as Elaine Scarry does not, the excitement about the imaginative act itself. We might explore uses of the passive and active moods in these inquiries as well as manipulations of tense and aspect. Such new inquiries might lead us farther down the road that Halbfass would have wanted us to go: whether the various media of the idea of dharma, in all their colorful technicalities, remain as crucial as dharma’s philosophical content.

BIBLIOGRAPHY Alsdorf, L. (1964), The Åkhyåna Theory Reconsidered, in «Journal of the Oriental Institute of Baroda», 13, n. 3, pp. 195-207. Ayyappa Paniker, K. (2003), Indian Narratology, Indira Gandhi National Center for the Arts, Delhi. Bailey, G., Brockington, M. (ed.) (2000), Epic Threads: John Brockington on the Sanskrit Epics, Oxford University Press, Delhi. Bergaigne, A. (1878-1883), La Religion Védique, F. Vieweg, Paris , 3 vols. Bhattacharya, A. (1992), Dharma, Adharma and Morality in the Mahåbhårata, S.S. Publishers, Delhi. Bloomfield, M. (1896), Contributions to the Interpretation of the Veda 7. 6. Trita, the scape-goat of the gods, in relation to Atharva Veda 6.112 and 1.113, in «American Journal of Philology», 17, pp. 430-437. 40

See White The Content of the Form (1987)

III. How to Produce, Construct and Legitimate a Tradition

155

_______ (1898), Contributions to the Interpretation of the Veda 3. 1. The Story of Indra and Namuci. 2. The two dogs of Yama in a new role. 3. The Marriage of Sarañyü, Tvaß™å’s daughter, in «Journal of the American Oriental Society», 15, pp. 143-188. Br¢haddevatå, edited and translated by A. Macdonell, Harvard Oriental Series, Harvard University Press, Cambridge 1904, 2 vols. Charpentier, J. (1922), Die Suparñasage. Untersuchungen zur Altindischen Literature und Sagen-geschichte, A.-b. Akademiska Bokhandeln i Kommission, Uppsala. Dange, S.A. (1969), Legends in the Mahåbhårata, Motilal Banarsidass, Delhi. Geldner, K. (1889), Vedische Studien, Verlag von W. Kohlhammer, Stuttgart, vol. 1. _______ (1897), Vedische Studien, Verlag von W. Kohlhammer, Stuttgart, vol. 2. _______ (1901), Vedische Studien, Verlag von W. Kohlhammer, Stuttgart, vol. 3. _______ (1923), Der Rigveda, Vandenhoeck and Ruprecht, Göttingen, vol. 1. Goldman, R. (1969), Mortal Man and Immortal Woman: An Interpretation of Three Åkhyåna Hymns of the ¥g Veda, in «Journal of the Oriental Institute of Baroda», 18, pp. 273-303. Gonda, J. (1975), A History of Indian Literature. Vol. 1 fasc. 1. Vedic Literature (Sa∫hitas and Bråhmañas), Otto Harassowitz, Wiesbaden. Gonzales Reimann, L. (2003), The Mahåbhårata and the Yugas, Peter Lang, New York. Gopal, R. (1969), Vedic Sources of the ‡årñgaka Legend in the Mahåbhårata, in «Journal of the Ganganatha Jha Research Institute», 25, pp. 397-401. Halbfass, W. (1988), India and Europe, State University of New York Press, Albany. _______ (1992), On Being and What There is: Classical Vai†eßika and the History of Indian Ontology, State University of New York Press, Albany. Hariyappa, H.L. (1953), ¥g Vedic Legends Throughout the Ages, Deccan College Dissertation Series 9, S.M. Katre, Poona. Hill, P. (2001), Fate, Predestination, and Human Action in the Mahåbhårata, Munshiram Manoharlal, Delhi. Hillebrandt, A. (1980), Vedic Mythology, (trans. by S.R. Sarma), Motilal Banarsidass, Delhi. Hiltebeitel, A. (2001), Rethinking the Mahåbhårata: A Reader’s Guide to the Education of the Dharma King, University of Chicago Press, Chicago. Horsch, P. (1966), Die Vedische Gåtha und ‡loka Litteratur, Francke Verlag, Bern. Hudson, E. (forthcoming), The Roll of the Dice and the Roll of the Story, in Proceedings of the International Conference on the Mahåbhårata.

156

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Keith, A.B. (1911), The Vedic Åkhyåna and the Indian Drama, in «Journal of the Royal Asiatic Society», part 2, pp. 979-1009. Levi, S. (1963), Le Théâtre Indien, College de France, Libraire Honoré Champion, Paris, 2 vols. (S. Levi, The Theatre of India, [trans. by N. Mukherji], Writers’ Workshop, Calcutta 1980, 2 vols.). Lingat, R. (1967), Les Sources du Droit dans le System Traditionnel de L’Inde, Paris (R. Lingat, The Classical Law of India, [trans. by J.D.M. Derrett], Berkeley 1973). Mehta, M. (1971), The Evolution of the Suparña Sage in the Mahåbhårata, in «Journal of the Oriental Institute of Baroda», 21, n. 12, pp. 41-65. Minkowski, C. (1989), Janamejaya’s Sattra and Ritual Structure, in «Journal of the American Oriental Society», 109, n. 3, pp. 401-420. The Nighañ™u and the Nirukta. The Oldest Indian Treatise on Etymology, Philology, and Semantics, Critically Edited From Original Manuscripts and Translated by Lakshman Sarup, Oxford University Press, London - New York 1920-1927. Nussbaum, M. (1992), Love’s Knowledge, Oxford University Press, New York. O’Flaherty, W. (1985), Tales of Sex and Violence, University of Chicago Press, Chicago. Oldenberg, H. (1883), Das altindische Åkhyåna, mit besonderer Ruecksicht auf das Suparñåkhyåna, in «Zeitschrift der Morgenlandischen Gesellschaft», 37, pp. 54-86. _______ (1885), Åkhyåna-Hymn im R¢g Veda, in «Zeitschrift der Morgenlandischen Gesellschaft», 39, pp. 52-83. _______ (1988), The Religion of the Veda, (trans. by S.B. Shrotri), Motilal Banarsidass, Delhi. Patton, L.L. (1993), Beyond the Myth of Origins: Types of Tales Telling in Vedic Commentary, in S. Biderman (ed.), Myth and Fictions: Their Place in Philosophy and Religion, E.J. Brill, Leiden. _______ (1996a), Myth as Argument: The Br¢haddevatå as Canonical Commentary, DeGruyter Mouton (in collaboration with Harvard University, Center for the Study of World Religions), Berlin. _______ (1996b), The Fate of the Female R¢ßi. The Changing Face of Lopåmudrå, in J. Leslie (ed.), The Making of India Myth, Curzon Press, London. Scarry, E. (1999), Dreaming by the Book, Princeton University Press, Princeton. Sieg, E. (1902), Die Sagenstoffe des ¥g Veda und die indische Itihåsatradition, Druck und Verlag von W. Kolhammer, Stuttgart. Sutton, N. (2000), Religious Doctrine in the Mahåbhårata, Motilal Banarsidass, Delhi. von Schroeder, L. (1908), Mysterium und Mimus im ¥g Veda, H. Haessel Verlag, Leipzig.

III. How to Produce, Construct and Legitimate a Tradition

157

Thieme, P. (1971), Agastya und Lopåmudrå, in Kleine Schriften, Franz Steinder Verlag GMBH, Wiesbaden, vol. 1. Tokunaga, M. (1979), Text and Legends, Harvard University Dissertation. Weber, A. (1891), Espisches im vedischen Ritual, in «Sitzersberichte der Preussischen Akademie der Wissenschaft», 38, pp. 769-819. White, H. (1987), The Content of the Form: Narrative Discourse and Historical Representation, Johns Hopkins University Press, Baltimore. Windisch, E. (1879), Über die altirische Sage der Tain Bo Cualnge, Verhandlung d. 33. Philologenversammlung in Gera. Winternitz, M. (1909), Dialogue, Åkhyåna, und Drama in der Indischen Literatur, in «Wiener Zeitschrift fr die Kunde des Morgenlandes», 23. Witzel, M. (1986), JB palpülanœ: The Structure of a Bråhmaña Tale, in R.K. Sharma, et al. (ed.), Felicitation Volume B.R. Sharma, Kendriya Sanskrit Vidyapeetha, Tirupati. _______ (1987), On the Origin of the Literary Device of the ‘Frame Story’ in Old Indian Literature, in H. Falk (ed.), Feschrift U. Schneider, Freiburg.

ROMILA THAPAR Creating Traditions through Narration. The Case of ‡akuntalå 1

The manner in which we construct the past is now acknowledged as an important process in the writing of history. This involves appropriating the past, an act in which the concerns of the present are apparent. Historical sources are used to construct a link between an event in the past and how we view it today. I would like to argue that there are in addition many representations of an event between the point at which it happened and the present, and that these representations are significant to the eventual understanding of the past. Such representations in the form of a narrative may either be fictional or may claim to embody an event, but in both cases they address themselves to a historical moment. This brings the relationship between narrative and history to the forefront. I will be looking at this relationship through the different versions of a fictionalized narrative, illustrating my argument with the story of ‡akuntalå in its variant forms. Does the retelling of the same narrative help our understanding of historical change in as much as the retelling reflects change in both society and ideology? Can we treat the act of narrativization or the making of a narrative, as constituting an event? Every narrative has a context which is consciously or subconsciously derived from a world view and an ideology. Let me hastily add however, that this is not to authenticate a story as history, for a story remains fictional. But it can reveal perspectives of a time and a society. I am suggesting that it be analysed as representing such a perspective, which emerges all the more clearly through a comparison of its retellings. A fictionalized narrative cannot be treated as history 1 This essay is a slightly modified version of the one published in R. Thapar, Narratives and the Making of History, Oxford University Press, Delhi 2000. The editor of this volume thanks both the author and the publisher for their kindness.

160

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

but it can be an indicator of a past condition. What I am arguing for is the analysis of narratives which become constituents of a historical perception and have therefore a contextual location. A narrative can have its own biography and the changes it manifests can provide us with a view of historical change. By historical change I do not mean just chronology but rather, the manifold dimensions of the historical context. A narrative frequently recreated over time becomes multi-layered like a palimpsest. One can attempt to reveal the many pasts which went into the making of its present. Where the retellings of a narrative or where narratives implying an event, become contesting versions, the differing perspectives also provide evidence for historical constructions. This relationship has been the subject of lively discussion among historians. Best known perhaps was the discussion between Lawrence Stone on the revival of narrative in history and its critique by Eric Hobsbawm, published in the 1980s in the British historical journal, Past and Present. The discussions focused largely on whether there was a shift away from social and economic history, drawing on the disciplines of the social sciences, towards directing attention to language, culture and ideas and a focus on micro-events. Was this a new way of viewing the structures of the story and of society? The suggested duality was found to be untenable since there was a considerable over-lap in both sources and interpretations. Even narrative history as it has developed in recent times, was not just a bald telling of a story. The new use of narrative incorporated analytical history and the analyses of the micro-event illumined the macro generalization. The discussion has taken a different form in this decade with the introduction of what has been termed ‘the linguistic turn’ (Evans 1997). Some have stated that history as a discipline has no future given the kind of analyses of narrative which are possible. History in this argument becomes a kind of pointillist history –rather like the style of painting– a collection of unconnected dots which taken together compose a picture. Historians have reacted with the logical argument that even these dots have to be contextualized as indeed does the picture itself. However significant the understanding of the fragments may be, history attempts to look at the larger whole. What ‘the linguistic turn’ has done is to make historians more aware of the nuances of language and words, which far from terminating historical investigation, have added to its precision. The writing of history has had a continuous interface with literature. Historians have culled literature for information on what may have happened in the past, the statements being juxtaposed with other kinds of evidence. This is a legitimate activity. I would however suggest a sharpening of this interface by changing the focus somewhat, by searching for the historical perspectives which this interface provides, through examining the representations present in the narrative. The same narrative or approximately the same, can occur in

III. How to Produce, Construct and Legitimate a Tradition

161

variant forms as different genres of literature –in this case, the story of ‡akuntalå in the Mahåbhårata, the play of Kålidåsa, the prose-poem in Braja-bhåßå. From a different perspective but with a bearing on the narrative, are the many translations of the Kålidåsa play where the act of translation in itself becomes a cultural negotiation, and there is also the commentary in the form of an essay by Rabindranath Tagore. These are significant moments in the biography of a narrative. But there is more that just an interface between literature and history. The narrative of ‡akuntalå, highlights the gender perspective. The same character is depicted differently in the variant forms. Does this reflect different social perceptions, the understanding of which requires some familiarity with the historical context? The form which the variants take –epic fragment, drama, poetry– and the cultural interpretations which they encourage, makes the narrative an item in cultural history. Choosing a particular item from the past and recreating it as a variant is, in part, an act of historical significance. The past is viewed from the present, wherever the present may be located, and that which is selected from the past goes into constructing a tradition or constructing a history. A tradition is never handed down intact from generation to generation, however appealing this idea may seem. Innovation is what gives it vitality. The items selected from the past are often so chosen as to legitimize the values and codes of the present. In selecting and recasting cultural items we highlight some and marginalize others. The act of selection becomes a dialogue with the past. The point in time at which the selection is being made gives a different value to the selection as a cultural symbol, as an idiom, as an icon. This has happened throughout our cultural history, although our awareness of this process is perhaps more apparent now. Where the narrative is culturally central to our own present today, we have also to see it as a part of the intervention of the colonial period and recognize the disjuncture this may produce. The concept of culture in relation to the early past, implies an intersecting of disciplines of which history, it seems to me is foundational. This involves the original text and its historical contexts, as also frequently the Orientalist reading of it and equally frequently, the internalizing of this reading by commentators of the last century or two. And more recently, the questioning of this reading. Inevitably there is a contextualizing of the Orientalist representation and European perspectives brought to bear on the reading. A single item can therefore have multiple identities which change at historical moments. Understanding a cultural item historically requires some comprehension of the world-view which it represented. Each version has some relation with those which preceded it: a relation ranging from endorsement to contestation of earlier versions. I would like to touch on some of these ideas using the narrative of ‡akuntalå. My focus therefore is not on the Kålidåsa play, but on the

162

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

treatment of the central figure which transforms the narrative in its variant versions; and on the possible historical explanations for the variants and the commentaries. Let me now turn to the narrative. The åkhyåna or narrative of ‡akuntalå as given in the Ådiparvan of the Mahåbhårata (1.62-69) is one among the many bardic fragments which were stitched together in the making of the epic. In many of these fragments the morphology of the folk tale is evident. There are other sections of the Mahåbhårata, such as the ‡åntiparvan, which have been labelled as didactic. These have less to do with the story and more with theories of the ideal society, of social obligations –dharma of government– rajadharma, of ideas about the liberation of the soul –mokßadharma, and such like. The ‡akuntalå story occurs in the narrative section. Råjå DuΔßanta, with the title of goptå, a protector of cows, has conquered widely. One day he goes on a hunt accompanied by a large entourage of soldiers. The hunt turns into a fierce killing of tigers and deer, the wounding of elephants, the uprooting of trees and a general devastation of nature. DuΔßanta follows a deer deep into the forest which brings him to the lush and secluded å†rama of Kañva. On calling out, a young woman answers and performs the ritual of welcome for the guest. She introduces herself as ‡akuntalå, the daughter of the r¢ßi Kañva. On DuΔßanta asking her how a r¢ßi could have daughter, she explains her parentage in detail. Indra, disturbed by the powers which the r¢ßi Vi†våmitra was accumulating through tapasya, sent the apsarå Menakå to seduce him. ‡akuntalå was born but discarded by Menakå and brought up as a foundling by Kañva in his å†rama. DuΔßanta, deeply attracted by what he calls ‘the flawless girl of the beautiful hips’, proposes a gåndharva marriage. This was a marriage by mutual consent, appropriate it is said, to kßatriyas. ‡akuntalå agrees but sets a condition that she will only marry him if the son born of this marriage is declared his successor. After a three year pregnancy she gives birth to a boy, Bharata. She takes him at a young age to Hastinåpura from where DuΔßanta rules, and demands that DuΔßanta recognize him as his heir. DuΔßanta pretends not to recognize her and rejects them both. ‡akuntalå in extreme anger, explains why a wife and son are necessary to him, particularly a son to continue the lineage. The exchange is heated with much down-to-earth abuse. Menakå is called a slut. Vi†våmitra a lecher and ‡akuntalå a whore. ‡akuntalå stands her ground and insists that the boy be given his status and to that end she decides to leave him with DuΔßanta. As she is about to return to the å†rama, a disembodied celestial voice proclaims that the boy is indeed DuΔßanta’s son. DuΔßanta explains that he had remembered his meeting with her and had no doubt about the veracity of ‡akuntalå’s claim, but was waiting for this public legitimation of the relationship. Subsequently he accepts them both. Bharata when he comes to rule is acclaimed as a great ruler.

III. How to Produce, Construct and Legitimate a Tradition

163

The story in the epic is the origin myth of Bharata and therefore also tied into the ancestry of the Kauravas and the Påñ∂avas, central to the events in the Mahåbhårata. Divine proclamation establishes status and legitimacy because the relationship has also to be accepted by the clansmen. It is a society of clans and heroes, a lineage-based society, where ancestry, genealogy and origins are vital (Thapar 1984). It is also a cattle-keeping society requiring extensive grazing grounds. Hence the respect for the title of goptå. The clearing of land and of forest for agriculture was recognized as a source of wealth. The hunt is a surrogate raid, a war against nature but also a means of establishing claims to territory. So dominance over the forest is beginning to assume importance. The depiction of ‡akuntalå is central to the story. She is forthright, free, high-spirited and assertive. She makes her marriage conditional and then demands that the promise be honoured. She accuses DuΔßanta of behaving unrighteously. She is the reverse of the pativratå, the ideal wife as described in the didactic sections of the epic. The dispute is clearly over the paternity of the child. The condition she imposed at the time of the gåndharva marriage hinged on the status of her son, characteristic of a patriarchal society. This was also crucial to the status of the woman in such a society even if it was a clan-based society: she was the link to kinship and alliances, and her son ensured her membership of the clan. The celestial voice describes the mother as the receptacle, for it is the father who begets the son, and the son frees the father from the abode of the dead. Implicit in this utterance is the statement that DuΔßanta accept responsibility for the child. The period of the composition of the epic remains controversial but generally it is thought that the composition and the interpolations can be placed between 400 BC and AD 400, the narrative sections possibly being earlier than the didactic sections (Sukthankar 1964). The epic continues to have an audience well into the centuries AD. It is part of ancestral mythology and provides links with the heroes of old. The epic was added to often enough, and presumably when it was converted to sacred literature it became part of brahmanical high culture. However, the hierarchy in this high culture would have placed the epics and Puråñas in what some regarded as the notso-high culture, perhaps because of their links with folk culture. I would now like to turn to the play, the Abhij∞åna†åkuntalam of Kålidåsa.2 It reflects a different historical scene. It was written subsequent to the story in the epic and is generally dated to about the fourth century AD although the date is controversial. Kålidåsa selects a fragment from the epic, converts the narrative into a play (nå™aka), which is a different genre of literature from the poetry of the epic. To the original narrative he adds other subthemes. One is the story of 2 See Kale 1961. A discussion and translation of the play is included in Stoler Miller 1984.

164

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

the ring as a token of recognition which seems to have come from the Buddhist Ka™™ahåri Jåtaka (Cowell 1969; Stoler Miller 1984). The other is the theme of the curse which is frequent in folk literature. There is, as a result, the creating of a new tradition. An item, selected from the past, is moulded to suit the cultural expression of the later time. It could be seen almost as a contestation with the epic version, the norms of which undergo changes in the play. The play is no longer concerned with lineage-based societies and clans but carries the rhetoric of the political power of monarchical states. These were well established, legitimizing the concentration of power in a single family and the authority of upper caste society. The state had its appurtenances of administration, revenue, coercive agencies and such like. There is also the visibility of brahmanical high culture which was dominant in the construction of classicism and therefore familiar to Kålidåsa. It is evident in the use of language and in the nuanced relationship between the characters. Kingship is approximate to deity and kings and gods intermingle. The å†rama of the Kañvas carries traces of a new incipient institution which was to develop into the agrahåras of post-Gupta times, institutions which changed the socio-economic landscape. Tax-free land was donated by the king for settlement by bråhmañas which could be in areas already under cultivation or newly opened to cultivation. These were to become powerful nuclei and networks of brahmanical culture. The play itself is intended for performance at the court before a small, sophisticated, urban audience and not as part of a popular recitation. It reflects the values of upper caste society although there may implicitly on occasion be some questioning of these. Intended as entertainment, the theme was inevitably romantic. The changes introduced by Kålidåsa are significant to more than just the story-line. DuΔßanta/Dußyanta leaves his ring with ‡akuntalå as a token of his promise to send for her on his return to Hastinåpura. Deep in thought one day, ‡akuntalå neglects to receive with appropriate ceremony an irascible r¢ßi Durvåsas who therefore spews out his curse that the person she is thinking of will not remember her. Her friends plead for at least a modification of the curse and the r¢ßi then says that the ring will provide the remembrance. ‡akuntalå leaves for the court and on the way loses the ring. On arriving there, she is not recognized by DuΔßanta and no amount of persuasion convinces him that she is his legally wedded wife bearing his son. ‡akuntalå in despair calls upon Mother Earth and there is a flash of lightening and she is whisked away to the å†rama of Marœca. Here she gives birth to her son Bharata. Meanwhile the ring is found in the belly of a fish, and since it is his signet ring, it is brought to DuΔßanta. On seeing it he recollects his relationship with ‡akuntalå. He is now full of remorse at having lost both a wife and a son. The eventual happy outcome occurs when the

III. How to Produce, Construct and Legitimate a Tradition

165

king is called to Indra’s aid in a campaign against the demons. On his return he stops at the å†rama of Mårœca where he is united with his wife and son. The story of the play is an elaboration of the skeletal story in the epic. Courtly drama requires a romantic mood and dramatic effects. The teasing out of the narrative is done through the sub-plots of the curse and the ring. There is a contrapuntal relationship between the two; the curse impedes action and is a barrier, the ring resolves the barrier so that the action can move. The curse and the ring gloss over the tension between DuΔßanta and ‡akuntalå, both over the paternity of the child and the responsibility of the father. But ‡akuntalå in the play cannot defend the right of her son because the flow of events is beyond human control and she had made no conditions to the marriage. DuΔßanta cannot be blamed for rejecting her as he is under a spell. Is Kålidåsa therefore avoiding the moral issue of condemning DuΔßanta’s action in rejecting ‡akuntalå? Or would this not have been regarded as irresponsible in those times and in that society? The epic version does at least raise the issue through the celestial voice. The structure of the play seems to be based on a duality which comes to be associated with an increasingly common view of the world. It is expressed in terms of the dichotomy of the gråma and the arañya or the kßetra and the vana –the settlement and the forest (Malamoud 1996: 87-88; 1976). It is generic to the epic where the broader action moves back and forth from settlement to forest. But it is strongly indented in the play as well. One of the reasons for this may be that by the Gupta period attitudes towards the forest were beginning to change. Whereas earlier the settlement was the ordered society and the forest the habitat of the unknown, and the wild, now the forest was beginning to be seen differently: as a source of revenue through its natural products of timber and elephants; its potential as agricultural land after clearing; and as the location of bråhmaña settlements, in the form of agrahåras. The society of the forest was no longer entirely unknown, but it was still different from village settlements and the difference continued to be emphasized. The dichotomy is highlighted in the play between the å†rama of Kañva and the court at Hastinåpura. It is further underlined in the depiction of ‡akuntalå as the woman of the å†rama and DuΔßanta as the man of the court. The å†rama in the play is the liminal area, the threshold between the settlement and the forest, for although it is set deep in the forest, the people who live there attuned to nature, are nevertheless also aware of the mores and customs of the settled society, from where they have come. They are not å™avikas, forest dwellers, in origin. Kålidåsa seems to use this duality to reverse the activities associated with each. The å†rama becomes the location for what has been called love-in-union –sa∫bhoga †r¢õgåra, generally not associated

166

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

with å†ramas. The court is the location for love-in-separation –vipralamba †r¢õgåra, where she is rejected and leaves, although most romances achieve fruition at the court (Stoler Miller 1984). From the epic narrative to the play there is a change in the conceptualizing of the woman. ‡akuntalå is now the child of nature and identifies with plants and animals. She dresses in bark clothes, adorns herself with flowers which miraculously turn into jewels at the time of her departure. Nature weeps at her going away. Her innocence is heightened by her grappling with the emotions of romantic love, leading her to the gåndharva marriage. She is shy, retiring, modest and generally submissive. In the last act she excuses DuΔßanta’s action because of his being under a spell, and instead explains to herself that she is reaping the consequences of some wrong doing on her part in a previous birth. If ‡akuntalå claims to be the wife of DuΔßanta she has to conform to the pativratå ideal. Although both Kañva and DuΔßanta refer to her as the lawfully wedded wife, one of Kañva’s disciples hints at the gåndharva marriage being a seduction. One wonders whether this is resentment against a woman’s transgression of patriarchy and her taking an independent decision, for he insists that she must suffer the consequences of such a decision. She is told that she cannot return to the å†rama and has to remain at the court because the husband’s authority over the wife is unlimited. He has the right to accept her or abandon her. It is better that a wife be a servant in her husband’s home than live away from him. The epic version had underlined the centrality of the son and the empowerment of the woman, both in herself and as the mother of a son. In the play romantic love seems to supersede this, and the question of empowerment fades away. The king does not taunt her for her illegitimacy but is uncomplimentary about women in general. Eventually the desire for an heir drives the king to as much grief as the disappearance of his beloved. Subsequent to the Kålidåsa play there were now two versions of the story in circulation. Briefly narrated in the Puråñas as an ancestral myth of the Pürus it was important to the legitimation of dynasties of the post-Gupta period (Bhågavatapuråña, 9.20.7-32; Matsyapuråña, 49.11-15). The recitations of the pauråñikas and the kathåkåras kept these stories alive among audiences more comfortable with the oral tradition. That it became something of a folk stereotype is evident from the Kathåsarœtasågara (32.306-390 [ed. Tawney 1968]) which includes a charming story using the same theme but replete with folk motifs. Interpretations of visual forms as pictorial representations of the story have also been suggested (Rapin 1996; Agrawala 1946: 102109; Marshall 1915: 29-49). At a somewhat later period the play becomes an item for discussion in a variety of theoretical works on literature and aesthetics. Taking off from the Nå™ya†åstra, there were wide-ranging views on

III. How to Produce, Construct and Legitimate a Tradition

167

what constitutes good poetry and drama, discussed in the works of theoreticians such as Abhinavagupta and Ånandavardhana at the end of the first millennium AD. More specific to the Kålidåsa play is the commentary of Råghavabha™™a in the sixteenth century. Much of the discussion was in the context of the evolving theories of rasa, central to Indian aesthetics. Gradually the Kålidåsa play became central to analysing both poetry and drama and was judged as the exemplar in the Sanskrit nå™aka tradition. It was doubtless both its reputation as the finest Sanskrit play and the popularity of the story, that led to its being adapted to yet another literary form which was to reach a still wider audience. In 1716, the Mughal emperor Farrukh Siyar bestowed a title on a nobleman at the court. To celebrate this, the court poet, Navåz Kavœ†vara, was asked to render –not to translate– the story of ‡akuntalå from Sanskrit into Braj-bhåßå, the language of much of the Hindi poetry at the time. The story now becomes a kathå in verse. The theme of love and separation and the style of the rendering, gives it a quality which recalls the dominant form in Braj poetry at that time –the bårahmåså. This is not to suggest that it was actually a bårahmåså, but shorn of the borrowings from the play, it was a kathå concerned with lovers, partings and reunions, characteristic of this kind of Braj poetry (Vaudville 1986; 1959: 129-134). The language is earthy, the poetry sounds like doggerel verse at times. ‡akuntalå emerges as less given to romanticism and more down-to-earth, a distinct echo of the ‡akuntalå of the epic. In some ways this is a mediation between the epic version and the play. In 1806, the Braj-kathå was translated into an Urdu prose-poem, Shakuntala, by Mirza Qasim Ali Dehlavi, an Urdu poet teaching at the recently established Fort William College at Calcutta. This brings an infusion of the Persian dåstån style with its world of fables and exaggerated emotions. ‡akuntalå, in embarrassment, constantly hides behind her ghunghat, and the king in true Majnu or Farhad style, swoons almost every time he sees her. But the dialogue remains earthy and the exchanges between the king and ‡akuntalå make for racy reading. The narrative moves away from being a court play and its more accessible language gives it a greater universality. Presumably its performance was accompanied by music, dance and mime. The feel of eighteenth century late Mughal society pervades this version. At this point, the biography of this narrative takes another turn. There are no further literary genres for the retellings, but it entered the world stage through translations. And translation changes the cultural role of the narrative, for it introduces into the play, the culture and the world views of the society using the language of the translation and of its ideologies. William Jones, often described as the father of British Indology, was an officer of the East India Company at Calcutta, and spent much time in reading and translating Sanskrit texts. He was enthused by the play, and translated it, first into Latin which was linguisti-

168

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

cally closer to Sanskrit, and then from Latin into English. In 1789 it was published as Shakoontala or the Fatal Ring (Canon, Pandey 1976: 530-537; Canon 1977; Mukherjee 1987). He gave currency to the phrase that Kålidåsa was the Indian Shakespeare. He maintained that the play demonstrates the height of Indian civilization, all the more remarkable because it was written at a time when the Britons were as unpolished and unlettered as the army of Hanuman. His more significant comment was that he had been disturbed by some of the more erotic passages which would be unacceptable to European taste. And for the first time, the erotic in the play became a matter for debate. Nevertheless, the play took Europe by storm. It was translated into German and acclaimed by the German poet Goethe in a verse which has since been repeated ad nauseam. There followed a succession of ballets and operas on the theme, including an incomplete attempt by Franz Schubert. In each decade of the nineteenth century there was yet another translation in yet another language, even Icelandic. The experimental theatre of Tairoff in Moscow made it the opening presentation with an enthusiastic reception from the Symbolist poets, just prior to the Bolshevik revolution. Throughout the nineteenth century in European literary circles, and most particularly in the German Romantic Movement, ‡akuntalå was projected as the child of nature and the ideal Indian woman encapsulating the beauty of woman kind (Willson 1964; Sedlar 1982; Drew 1987). Her closeness to nature was particularly important to literary Romanticism distancing itself from the formalism of neo-classicism. This was also in part a response to what was referred to as the ‘discovery’ of the orient or the Oriental Renaissance. European Romanticism was inter-twined with Orientalism. To understand the construction of Orientalism and its fusion with European Romanticism, requires a familiarity with the images, created as part of the intellectual history of Europe in the nineteenth century, and the politics of these images. The Oriental Renaissance it was believed, would provide new visions of how man should perceive the world (Schwab 1984). But the images were what Europe projected onto the Orient. These were crystalized as the duality of the Orient and of Europe as expressed in the preference of Romanticism for the less orderly aspect of the past and its search for the exotic, the irrational, and the imaginative as against the rational and the real, thought to be typical of European classicism. The creation of what has been called the ideal of India in German Romanticism was also conditioned by early Greek stories of Alexander of Macedon’s meetings with Indian philosophers. This was said to explain the presence in substratum European thought of ideas on metempsychosis, the unity of man and nature, and the meaning of renunciation. These were central to the theories of the NeoPlatonists who believed that much of the philosophy alternative to

III. How to Produce, Construct and Legitimate a Tradition

169

the Judaeo-Christian tradition in Europe came from Indian sources. Romanticism therefore was also questioning the theories of the European mainstream. With the growth of notions of race and the wide acceptance of what came to be called ‘race science’ in the later nineteenth century, a touch of racism entered the idyllic picture of a closeness to nature (Hanson 1997). The children of nature were the primitive peoples, at the foot of the evolutionary ladder. Eroticism therefore was an aspect of their unawareness of the need for moral laws. But the not-so-idyllic relationship between colonizers and the colonized in the nineteenth century contributed to a fading out of the enthusiasm for Romanticism. If in the early nineteenth century there was a concern to reform the native to the ways of the colonizer, by the latter part of the century this was seen as an impossibility because the native was believed to be racially inferior. By the end of the nineteenth century, ‡akuntalå had become a collector’s item in Europe. Not so in India. It was in the nineteenth century that the play became important both to debates on colonial cultural policy and to the self-definition of the Indian middle-class. James Mill, writing as a liberal utilitarian, in the early nineteenth century, saw little that was worthwhile in Indian culture, opposed Orientalism, and argued that Sanskrit literature was the literature of a self-indulgent society. It is only nations in their infancy who produce literature which is in praise of the pastoral, for such societies are fettered by despots and they can only indulge in light romances, rather than analysing their condition. The gåndharva marriage, the curse, the authority of the bråhmañas, were for him, signs of Indian degradation (Mill 1823: 111). But there was a tradition among British administrators with a bent for scholarship and working in India, of a more ambiguous view. They felt that those who governed India had to be familiar with its culture and this coincided with forms of exercising power. The so-called ‘rediscovery’ of the Indian past was in part directed towards this end. But it was also an attempt to revive Indian culture in the format of Orientalist scholarship. This is perhaps best stated in the introduction to yet another translation of the play, published by Monier-Williams in 1855, which superseded the translation of Jones. Where Jones in his writing was representing India both to the Indians and to Europe, now there was a subordination of cultural representation to the politics of governance. The attempt was to mould the Indian understanding of its cultural past in the way in which the colonizer intended (Visvanathan 1989: 121-127). Monier-Williams states in the Introduction to the eighth edition of his translation published in 1898, that it was intended for a variety of purposes. It would enable the British to familiarize themselves with the life of the Hindus. It was also part of British policy to rediscover the Indian past for the Indian, to revive Indian culture as defined by Orientalist scholarship, to make the Indian middle-class

170

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

aware of this culture and to imprint on the mind of the Indian middle-class, the interpretation given to the culture by Orientalist scholarship. The impression conveyed is that the acclamation for the play should be attributed to Orientalist scholarship, thus forgetting or ignoring, the extensive analyses of earlier literary theorists who wrote some centuries before. There was now a shift of emphasis and the play was viewed as an item of Hindu culture, explaining the condition of the Hindu subjects of the empire. The reading of the play was moving from ‡akuntalå being the child of nature to her being what Monier-Williams calls, the ‘rustic maiden’. Nature and culture were no longer juxtaposed for nature had receded and the mores of ‘civilization’ had become essential to assessing the actions of the play. Initially the play was not selected as a text for the teaching of Sanskrit at college level because it was said to support immorality. Eventually the supposedly erotic passages were deleted and it came to be prescribed. Implicit in this argument is the question of morality –but it is not a comment on the moral decision on which the earlier tradition had focused, that of DuΔßanta’s rejection of ‡akuntalå. The question of morality as related to eroticism, which had not been a concern earlier, was now made the central issue and impinged on the projection of ‡akuntalå. These ideas had an influence on the emerging Indian middle class. Nineteenth century nationalism in India is thought to have fostered a conservative attitude towards tradition, because to question it was a concession to western ideas (Chatterjee 1993; 1989; Jayawardena 1995). The broader middle-class codes were also being forged with the emergence of a new class, associated with the upper castes. These drew from both the new historical situation of colonialism and what was described as the Indian tradition. But in relation to the perspective on women in society, the particular conservatism of Victorian morals had also entered Indian society. There was an appropriation of some of the attitudes of the Judaeo-Christian tradition, attitudes generally absent in early Indian texts. Gradually the definition of womanly virtues focused on modesty, chastity, selfsacrifice, devotion and patience. These were the virtues recognized in the ‡akuntalå of the play but these would have been unfamiliar to the ‡akuntalå of the epic. In a later phase of nationalism, a certain liberalism towards women was encouraged and women began to tentatively assert what they saw as their rights. Participation in the national movement was not intended to emancipate women but to encourage a sense of partnership. With rare exceptions, most women remained the subordinate partners. Victorian attitudes and social conservatism could not be set aside so easily. It was only a matter of time therefore, before someone would declare ‡akuntalå’s actions as ‘the fall of ‡akuntalå’. What is surprising is that this comment comes from Rabindranath Tagore. In 1907

III. How to Produce, Construct and Legitimate a Tradition

171

he published an essay in Bengali which was later translated into English with the title, ‡akuntalå: Its Inner Meaning (Tagore 1911). He takes Goethe’s verse on ‡akuntalå as his starting point and argues that the play is a series of developments from the lesser to the finer, from the flower to the fruit, earth to heaven and matter to spirit. From a young, passionate woman ‡akuntalå becomes the model of a devoted wife with qualities of reserve, endurance of sorrow, rigid discipline and piety. According to him, the play focuses on two unions: one is the gross, earthly, physical union with desire contributing to the fall of ‡akuntalå –and he uses the words patana and patita in association with her actions; and the other is the moral union when both DuΔßanta and ‡akuntalå have been cleansed through a long period of separation. Their tapasya takes the form of grief, remorse and penance and is necessary to a true and eternal union. Love is not its own highest glory, for goodness is the final goal of love. This is Tagore’s reading of the inner meaning of the play and he sees it as an allegory. Tagore’s reading reflects the moral concerns of his time, influenced it would seem by the perspectives of Indian nationalism and also Orientalism. In this reading the empowerment of a woman through the birth of her son, which was significant to the epic story, now becomes unimportant. The woman’s morality is the central question. Let me return to the relationship of narrative and history. If I am reading history into the context of the different versions and commentaries, it is because they are distinct in form and ideology, and when seen in sequence, represent historical changes. I have tried to demonstrate the interface between literature and history not by limiting myself to garnering historical information from the texts but by trying to see the texts as representing historical contexts. I have tried to show that the narrative of ‡akuntalå changes, either in itself, or through the many translations of one version, and it becomes an icon of varying concerns. Underlying the sequence is what seems to me to be a transformation of these concerns from earlier times to colonial times: a transformation which shifts the focus quite strikingly. Its visibility is clearest in the treatment of gender. This is evident in the portrayal of ‡akuntalå. She is ostensibly the same character in the variant versions but is in effect, perceived differently in each. The perception is not unrelated to a shifting social and moral focus of the story, shifting in accordance with historical demands. She is the mother of an epic hero in the Mahåbhårata where the main issue is the paternity of her child and the father’s responsibility in recognizing this. In the play she is the romantic ideal of upper-caste high culture, where moral responsibility is misted over by the introduction of the extraneous factors of the curse and the ring. In the Braj-bhåßå kathå, she is not cowed down by the king –if anything it may be the reverse– and insists on his behaving in a just manner. German Romanticism sees her as the child of nature, the

172

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

personification of innocence and pays little attention to problems of paternity and responsibility. The ‘rustic maiden’ from the colonial perspective, becomes enmeshed in colonial readings of the erotic in the culture of the colonized. The ideal wife within a nationalism reaching back to what it sees as tradition, raises the question of morality but the problem now devolves around the woman always having to exercise restraint. This is a middle-class perspective since subaltern perspectives remain outside the picture. I have tried to show that each version comes out of a process of selection and implicit in this is the contemporizing of the icon. We select from the past those images which we want from the present. These contribute to the construction of the self-image of our contemporary culture and its projection back into what is believed to be ‘tradition’. From the gender perspective we have in the last two centuries, ignored the ‡akuntalå of the Mahåbhårata, the liberated woman demanding to be justly treated, and have endorsed the ‡akuntalå of Kålidåsa, the woman waiting patiently for a recognition of her virtue.

BIBLIOGRAPHY Agrawala, V.S. (1946), Våsavadattå and ‡akuntalå: Scenes in the Ranigumpha cave in Orissa, in «Journal of the Indian Society of Oriental Art», 14. Evans, R.J. (1997), In Defence of History, London. Canon, G. (1977), The Life and Mind of Oriental Jones, Cambridge. ______ Pandey, S. (1976), Sir William Jones Revisited: On His Translation of the ‡åkuntalå, in «Journal of the American Oriental Society», vol. 96, n. 4, pp. 530-537. Chatterjee, P. (1993), The Nationalist Resolution of the Women’s Question, in Kumkum Sangari, Sudesh Vaid (eds.), Recasting Women, Delhi. ______ (1989), Colonialism, Nationalism and Coloured Women: The Contest in India, in «American Ethnologist», vol. 16, n. 4, pp. 622-633. Cowell, E.B. (ed.) (1969), The Jåtakas, London, vol. 1, n. 7. Drew, H. (1987), India and the Romantic Imagination, Delhi. Hanson, T.B. (1997), Inside the Romanticist Episteme, in «Thesis Eleven», 48, pp. 21-41. Jayawardena, K. (1995), The White Woman’s Other Burden, London. Kale, M.R. (1961), The Abhij∞åna-†åkuntalam of Kålidåsa, Bombay. Malamoud, C. (1996), Cooking the World, Delhi. ______ (1976), Village et Fòret dans l’ideologie de l’lnde Brahmanique, in «Archives Sociologie Européene», July.

III. How to Produce, Construct and Legitimate a Tradition

173

Marshall, J.H. (1915) [1911-1912], Excavations at Bhita, in «ASIAR», Calcutta, pp. 29-49. Mill, J. (1823), History of British India, London, vol. 2. Mukherjee, S.N. (19872), Sir William Jones, Delhi. Rapin, C. (1996), Indian Art from Afghanistan, Delhi. Schwab, R. (1984), The Oriental Renaissance: Europe’s Rediscovery of India and the East, 1680-1880, New York. Sedlar, J. (1982), India in the Mind of Germany, Washington. Stoler Miller, B. (ed.) (1984), Theater of Memory, New York. Sukthankar, V.S. (1964), On the Meaning of the Mahåbhårata, Bombay. Tagore, R. (1911), Sakuntalå: Its Inner Meaning, in «Modern Review», vol. 9, pp. 171-184. Tawney, C.H. (ed., tr.) (1968), The Kathåsarœtasågara, (repr.). Thapar, R. (1984), From Lineage to State, Delhi. Vaudville, C. (1986), Barahmasa in Indian Literature, Delhi. ______ (1959) A Note on the Ghataparkara and the Meghaduta, in «Journal of the Oriental Institute (Baroda)», vol. 9, n. 2, pp. 129-134. Visvanathan, G. (1989), Masks of Conquest, New York. Willson, A.L. (1964), A Mythical Image: The Ideal of India in German Romanticism, Durham.

JONARDON GANERI A Dynamic Tradition of Truth-telling: Moral Innovation in the Mahåbhårata

1. The Idea of a Dynamic Tradition The kernel of the idea of tradition consists in the idea of a practice whose present performance is causally explained by the fact of its having been performed in the past. Traditions instantiate the scheme ‘X is done because X has been done’, where the ‘because’ stands for a relation of causal explanation (the ‘because’ as is found in statements such as ‘the river is swollen because it rained’). The term ‘practice’ in this definition is of broad extention, referring to any species of intentional human activity, and including all but not only the following: matters of personal, familial and social custom, ritual activity, assent to particular and systematic doctrine, the transmission of apprenticeship, skills and training, research programmes and methodologies of inquiry, valuational practices and the cultivation of particular virtues. A tradition is a practice whose later stages are causally self-explained by its earlier stages. We should note that the idea of causal self-explanation is consistent with the existence of external causal factors acting on the tradition, just as citing the rain as the causal explanation of the river’s being swollen does not preclude the existence of other causal factors, such as the soil being dry. Causal explanation involves the selection, from among the totality of causes, one that is especially salient or relevant (and in that sense, explanatory). There is no implication, then, that traditions are insulated, self-sufficient, or immune to external causal influence. The core idea of tradition sustains elaborations and extensions which result in richer but historically contingent conceptions of tradition. The history of the conception of tradition in Europe has extended the core idea of tradition in at least two respects. The Enlightenment deprecation of tradition as an enslavement of rea-

176

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

son, forcibly represented in Diderot’s Encyclopedia, derives from a stricter reading of the ‘because’ in the schema that grounds the core idea of tradition: that ‘because’ is now read as implying both causal explanation and rational justification. The Enlightenment’s conception of tradition commits one to thinking that the past provides one with reasons for action and belief, a proposition which is inconsistent with the new ideal of individual rational autonomy. The core idea of tradition is expanded in a different direction when tradition is made to contrast with modernity. Now the ‘because’ in the schema is read as implying both causal explanation and axiological appraisal, a traditional society being one in which the past performance of a practice automatically instills it with value or worth. Both these developments of the idea of tradition encourage the idea that it is sufficient for a tradition to perpetuate itself that it remains numerically identical across times. The application of the idea of tradition in South Asia ought not to be assumed to exhibit either of these two historically contingent extensions to the core notion of tradition as a causally self-explanatory practice. If a tradition is any practice the later performance of which is causally explained by its earlier performance, then nothing precludes the further idea that the form of that practice might change over time: all the core idea of tradition requires is that the present form the practice takes is causally explained by the forms it took in the past. Let us say that a tradition which meets this requirement is dynamic. The term ‘living tradition’ implicitly assumes an organic model for the dynamics of tradition, but this is not the only possible model. When a tradition is dynamic, a further question arises, namely, by what crtieria are we to distinguish those changes in the form of the practice which constitute ‘developments’ from those which are ‘corruptions’. These terms are taken from John Henry Cardinal Newman (Newman 1845), whose treatment of the problem of development in Christian doctrine provides a rich resource for the more general study of dynamics in tradition, as we will see below. Let us first note that it will not do to offer a causal criterion of development versus corruption, to say, for example, that any change brought about by processes within the tradition is a development, and any change brought about by external causes acting upon the tradition is a corruption. This is for two reasons. First, dynamic traditions are typically able to absorb external influence, and, indeed, such influences often perform an important function in spurring the development of the tradition. Second, dynamic traditions are typically able to persist through discontinuous changes of practice: they undergo ‘paradigm shifts’ in the language of Thomas Kuhn. We need an account of the dynamics of tradition rich enough to allow for paradigm shifts within a tradition. For both these reasons, no merely causal criterion of development will be adequate. What we may well want to insist upon is that a change in a tradition can constitute a

III. How to Produce, Construct and Legitimate a Tradition

177

development only if it can be made sense of as a development according to the immersed critical standards of the tradition itself (see Nussbaum, Sen 1989; Williams 2002). 2. Newman’s Analysis of Dynamic Traditions A recent commentator has said of Newman’s account that although specifically dealing with theology and the Church, Newman’s essay is a profound analysis of the continuity through discontinuity present in any long-lasting tradition, with implications for any field of human endeavour which manifests creative interplay between inherited tradition, rational reflection and the wider social circumstances in which it is located […] Newman shows that the success of a tradition is related to its ability to assimilate new data, while conserving its past principles and achievements, and also to its ability to develop complex sequences of thought and practice while anticipating future development. He brings to the study of tradition a subtlety and a comparative perspective often lacking in the blanket statements of self-professed traditionalists and antitraditionalists alike. (O’Hear 1998)

Newman conceives of the development and corruption of Christian doctrine as akin to the growth and decay of a living body. Just as, in the case of a body, there is a “process towards perfection” followed by “the reversal and undoing of what went before”, where “till this point of regression is reached, the body has a function of its own, and a direction and aim in its action, and a nature with laws”, so too one may “discriminate healthy developments of an idea from its sate of corruption and decay” (Newman 1890: 171). The marks of a healthy development are summarised as follows: There is no corruption if [the idea] retains one and the same type, the same principles, the same organisation; if its beginnings anticipate its subsequent phases, and its later phenomena protect and subserve its earlier; if it has a power of assimilation and revival, and a vigorous action from first to last. (Ibidem)

Let us consider these several marks in turn, following Newman’s elaboration of them. 2.1. Identity of Type, Continuity of Principles, and Logical Sequence (i.e. preservation of organisation). By the ‘type’ of an idea is meant something that underlies whatever particular form in which the idea finds expression. Newman comments, on the one hand, that “ideas may remain, when the expression of them is indefinitely varied”, and, on the other, that “one cause of corruption in religion is the refusal to follow the course of doctrine as it moves on, and an obstinacy in the notions of the

178

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

past” (Newman 1890: 176-177). This amounts to an important criticism of fundamentalism –the ossification of an idea is the sign of unhealth and decay. Fidelity to a religious belief implies an acknowledgement of its underlying idea, but this is in no way antithetical to a respect for ways in which the articulation or “external image” of the idea can change. How do we identity the underlying ‘type’ of an idea? This is a problem of epistemology Newman does not explicitly address, but it is certainly consistent with his position that the later developments of the idea are in fact our best evidence for the underlying type, and that would point to a second reason why fundamentalism is mistaken –the original expression of an idea might be a poor indication of its true form. As for the continuity of principles, a “principle”, in Newman’s terminology, is distinguished from a “doctrine”. The same doctrine, for example the transitiveness of worldly goods, might lead an Epicurean to hedonism and an ascetic to mortification (Newman 1890: 180). A principle is a maxim of conduct, justified but not entailed by a correlative belief. A tradition which abandons its principles, Newman says, is like a people who have lost their spirit. A third mark of a genuine development is logical sequence: “a doctrine professed in its mature years by a philosophy or religion, is likely to be a true development, not a corruption, in proportion as it seems to be the logical issue of its original teaching” (Newman 1890: 195). Newman’s illustration is Lutheranism, which “issued in various theories in Pantheism, which from the first was at the bottom of Luther’s doctrine and personal character” (Newman 1890: 193), an issue therefore constituting a faithful development of the original idea. 2.2. Anticipation and Preservation The idea that a dynamic tradition anticipates its own future development is extremely insightful. Newman says: Since, when an idea is living, that is, influential and effective, it is sure to develop according to its own nature, and the tendencies, which are carried out on the long run, may under favourable circumstances show themselves early as well as late, and logic is the same in all ages, instances of a development which is to come, though vague and isolated, may occur from the very first, though a lapse of time be necessary to bring them to perfection […] and it is in no wise strange that here and there definite specimens of advanced teaching should very early occur, which in the historical course are not found till a late day. (Newman 1890: 195-196)

If the genuine development of a tradition consists in the ‘perfection’ of its underlying idea and its principles, and if the possibility of such a perfection has existed from the first, then we might well expect to find, albeit in an incohate and undeveloped form, anticipations of such later developments in the earlier strata of the

III. How to Produce, Construct and Legitimate a Tradition

179

tradition. On the other hand, of a shift in the tradition which is not a fulfilment of its underlying idea, there will be no antecedent anticipation. This idea provides a powerful rationale for a strategy of legitimisation which is prominent in the Indian literature: a later author will justify an innovation by seeking to demonstrate that the new idea was anticipated in the earlier literature, for example by referring to a sütra as a ‘proof text’. The strategy is evident, for example, in the work of later Navya-Nyåya, where the followers of the innovative thinker Raghunåtha wrote commentaries on the Nyåyasütras with the explicit intent of showing that Raghunåtha’s ideas were the perfection of doctrinal implications already implicit in and anticipated by the earliest texts. Of course, the clarity of this criterion for a genuine development is always in danger of being clouded by what we might call creative hermeneutics, and it is for this reason that the criterion is hard to apply in practice. But the principle, that a tradition has a proleptic unity (much as some claim for the dialogues of Plato) is an important one. The idea behind the idea that a tradition acts conservatively on its part, is another expression of the principle that genuine developments are perfections and ‘illustrations’ of the basic principles of the tradition (Newman 1890: 199). 2.3. Assimilation and Vigour According to Newman, it is the very essence of a healthy tradition that it has the ability to absorb new ideas from outside itself. He says: [M]athematical and other abstract creations […] are solitary and selfdependent; but doctrines and views which relate to man are not placed in a void, but in the crowded world, and make way for themselves by interpenetration, and develop by absorption. Facts and opinions, which have hitherto been regarded in other relations and grouped round other centres, henceforth are gradually attracted to new influence and subjected to a new sovereign. They are modified, laid down afresh, thrust aside, as the case may be. A new element of order and composition has come among them; and its life is proved by this capacity of exansion, without disarrangement or dissolution. An eclectic, conservative, assimilating, healing, moulding process, a unitive power, is of the essence […] of a faithful development. (Newman 1890: 186) The stronger and more living is an idea, that is, the more powerful hold it exercises on the minds of men, the more able is it to dispense with safeguards, and trust itself against the danger of corruption. As strong frames exult in their agility, and healthy constitutions throw off ailments, so parties and schools that live can afford to be rash, and will sometimes be betrayed into extravagances, yet are brought right by their inherent vigour. […] Thus Saints are often characterized by acts which are no pattern for others; and the most gifted men are, by reason of their very gifts, sometimes led into fatal inadvertences. Hence vows are the wise defence of unstable virtue, and general rules the refuge of feeble authority. (Newman 1890: 188-189)

180

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

These are among the finest statements in Newman’s work. Healthy traditions are not insular; rather, they are able to absorb and assimilate external influences. The reason is surety in their inner principles, which are then brought into relationship with ideas from outside the tradition. Outside influences do not corrupt a healthy tradition but rather assist it in its development: that an idea more readily coalesces with these ideas than with those does not show that it has been unduly influenced, that is, corrupted by them, but that it has an antecedent affinity with them. (Newman 1890: 187)

A dynamic tradition finds in the wider social environment a rich source of nutrition, from which it draws strength in the process of perfecting its internal principles. A development, then, in Newman’s account, is a change in the body of the tradition that can be seen as following logically from the fundamental principles of the tradition, even if it is not brought about by causes internal to the tradition; it is a change consistent with those principles, and a fuller expression and articulation of them. Newman’s reference to the respective deeds of the morally vigorous saint and the morally shaky rule-follower leads me into my discussion of trust in the Indian tradition. There is in India a long tradition of valuing truthfulness, along with the correlative values of trustworthiness and sincerity. What then should we make of that pivotal event in the Mahåbhårata, Yudhiß™hira’s lie that brought the downfall of Droña? Was this a development of the tradition of truthfulness, a sign of its health and flexibility? Or was it a corruption, an indication that the tradition was in decay? If virtues have histories, can we then speak of these events as a form of moral progress, meaning by that phrase that the conception of what the virtue of truthfulness consists in has undergone a development? B.K. Matilal has argued that Kr¢ßña’s role in this incident is as a sort of moral expert, someone who is able to bring about a paradigm shift in the conception of dharma: Sometimes it is possible for a leader to transcend or breach the rigid code of conduct valued in the society, with the sole idea of creating a new paradigm that will also be acknowledged and esteemed within that order. Our Kr¢ßña might be looked upon as a leader of that sort. It may be that he created new paradigms for showing limitations of such a generally accepted moral code of truth-telling and promise-keeping. (Matilal 2002a: 105)

There is a clear echo, in this remark, of Newman’s theory of development. Kr¢ßña, the moral expert, brings about a dramatic change in the tradition, but one that is ‘acknowledged and esteemed’ by that tradition itself; or, in other words, capable of being rendered intelligible according to the internal critical principles of the tradition as progressive, a change in the direction of a perfection of the inner principle of the tradition.

III. How to Produce, Construct and Legitimate a Tradition

181

3. The Role of Deception in the Mahåbhårata.1 The Mahåbhårata is an epic tale of illusion, deceit, and the manipulation of myth, and nowhere are these themes more visible than in the story of Droña in the Droñaparvan. The morality of deception is investigated simultaneously on four levels, through the respective powers of four myth-makers. There is, first of all and most obviously, Yudhiß™hira himself, whose false words induce Droña to lay down his weapons. In the figure of Yudhiß™hira the story explores the morality of myth-making between individuals, the virtues of trust and truthfulness, and the shame involved in their violation. Kr¢ßña, Bhœma and Arjuna embody different ethical voices –Kr¢ßña for a straight deception, Bhœma for a ‘crooked’ deception that pretends not to be one, and Arjuna for no deception at all. Yudhiß™hira might have preserved his moral integrity and saved himself from shame, had he followed the council of either Kr¢ßña or Arjuna, but in opting instead for Bhœma’s slippery double deceit (hiding both the truth and the fact of deception), he displayed the moral weakness we associate with his name. Myth-making at another level is woven into the fabric of the story, the great battle between Påñ∂ava and Kaurava. The weapons of battle, in fact, really are myths and illusions, summoned up by mantras to envelop the enemy in a cloud of deception. Droña is, in this sense, a great myth-maker, but greater still is his son A†vatthåman at whose invocation the deadly nåråyaña itself is deployed. In this great battle for truth, the power of the weapon is a measure of the legitimacy of the illusion that sustains it (it is much too simplistic to say that the Kauravas use only illusion, and the Påñ∂avas only truth). The third myth-maker is our story-teller himself, Vyåsa, who, extraordinarily, has himself enter as a character within his own story. On the road, Vyåsa meets first A†vatthåman and then Arjuna, and to each one retells the tale of the mighty battle in the form of cosmological allegory. But here we see the story-teller in the very act of practising his craft, giving A†vatthåman and Arjuna subtly different versions of the story, manufacturing in them different understandings of the moral foundations of what has taken place. More than that –all this is done in front of us, the readers, for he is ‘showing off’ his myth-making skills with the help of the literary device that has himself enter the plot. He is revealing to us what this story is all about– the clash of illusions, the fight of illusion by illusion, the elusiveness of truth. Nobody emerges unscathed: not Arjuna, who is perhaps the most resilient in his opposition to the use of all forms of illusion; nor Kr¢ßña, who promotes a tactical morality, who makes illusions self-destruct; nor yet Yudhiß™hira, through whom the morality of rules is exposed as a sham. The battle takes on cosmogonic dimensions when, in Vyåsa’s 1 This discussion of deception in the Mahåbhårata revised and adapts an earlier treatment, published in Chong Kim Chong, Yuli Liu 2004.

182

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

retelling, a fourth myth-maker is introduced in the shape of Rudra, the lord of creation, who has the power to determine each individual’s relationship with the world ‘from the outside’ as it were –i.e. by creating the world they inhabit. To one individual alone the creator gives the power to refashion the relationship that has been created, a power, as one might put it, to overpower. This is Nåråyaña who is twinned with Nara who is equated with Arjuna, who therefore alone can defeat Aßvatthåman’s mighty arsenal of illusion. God has fashioned a correspondence of word and world, but man can refashion it; God has created a moral order, but man has to recreate it. Truth and morals are god-given, but God has also given to man a greater gift, the power to re-work truth and morals when the times require it, to repair God’s work,2 to refine his first approximation. The world-creator, the myth-maker and the story-teller have this in common –they give to their created characters the power to retell the story and so to revise their own relationship with the world. 4. Droña: the chronicle of a death foretold Droña’s murder has been foreseen already. Much earlier in the Mahåbhårata, on the eve of the mighty battle between Påñ∂ava and Kaurava, warring factions of a divided family, Arjuna succumbs to the force of moral scruple. He is unwilling to take up arms against Droña, who was the teacher of the Påñ∂ava brothers in the arts of weaponry, if not in the crafts of moral reason: ‘How can I fight back at Bhœßma with my arrows in battle, or at Droña, Madhusüdana? Both deserve my homage, enemy-slayer! It were better that without slaying my gurus I went begging instead for alms in this land than that I by slaying my covetous gurus indulge in the joys that are dipped in their blood. And we do not know what is better for us: that we defeat them or they defeat us; Dhr¢taråß™ra’s men are positioned before us, after killing whom we have nothing to live for. My nature afflicted with the vice of despair, my mind confused over what is the Law (dharma), I ask, what is better? Pray tell me for sure, pray guide me, your student who asks for your help! There is nothing I see that might dispel this sorrow that desiccates my senses, if on earth I were to obtain without rivals a kingdom, nay even the reign of the Gods!’ Having spoken thus to Hr¢ßœke†a, enemy-burner Gu∂åke†a said to Govinda, ‘I will not fight!’ and fell silent. (Bhagavadgœtå 2.4-9 [trans. van Buitenen])

In its entirety, the Bhagavadgœtå is a prelude to Droña’s murder, an anticipation of the moral quandary his existence will cause for the Påñ∂ava brothers Yudhiß™hira, Arjuna and Bhœma. Kr¢ßña, the sidekick süta, gives council, and we shall have reason to see that his persuasiveness secures only the acquiescence but not the moral acceptance of Arjuna. When Droña is killed, Arjuna will regard him2

See the discussion of vedism in Smith 1989.

III. How to Produce, Construct and Legitimate a Tradition

183

self and the whole Påñ∂ava army as condemned to hell by that act. And is the assessment of the poet Vyåsa very different when, having had Droña very conspicuously ascend to heaven, he endows his son A†vatthåman with a nåråyaña, the supreme weapon of divine retribution? If, in the end, Vyåsa permits the Påñ∂ava brothers to remain alive, purveying the mass carnage on the battlefield of Kurukßetra, it seems to be only in order that he have someone there able to give voice to the folly of their ways. The Droñaparvan is the chronicle of a death foretold. It begins with Duryodhana, eldest of one hundred brothers, the one hundred sons of the blind king Dhr¢taråß™ra who commands the Kaurava army. Duryodhana, fearful of losing the battle, hits upon an ingenious but disingenuous subterfuge. He directs Droña to secure the capture of Yudhiß™hira, whom he can then command to tell the Påñ∂avas to surrender. For Yudhiß™hira, he reasons, is known for his impeccable truthfulness, and the Påñ∂avas will certainly believe what he says if by him told. Duryodhana, in other words, connives to exploit Yudhiß™hira’s virtue, to bend it against him and into his own service. The success of his scheme requires that Droña take Yudhiß™hira alive, and this Droña vows to do. Twice early on Droña encounters Yudhiß™hira, and twice apparently tries with all his considerable might to kill him, fortunately for his own vow without success. Once Arjuna encounters Droña, and Droña flees while Arjuna appeals for him to stay awhile and talk. We can only speculate what Arjuna might have said, and whether he might have been able to avert with words and worldly reason what was instead to come. Droña flees, but with his expertise in the theory and practice of weaponry, and with the Kaurava army at his command, he begins to overwhelm the Påñ∂ava troops. Fear then enters the hearts of the Påñ∂avas, and Arjuna, who alone among them possesses the skill to defeat his teacher, still refuses to fight. Kr¢ßña chooses this moment again to speak to Arjuna, this time with an astonishingly base suggestion: He cannot in any way be defeated by force in battle. Casting aside virtue, O Påñ∂avas, resort to a method fit for victory, so that Droña might not kill everyone in battle. I think that he will not fight if [his son] A†vatthåman were killed. Let some man say that he has been slain in battle. (Mahåbhårata, 7.164.67-69)

Although it is not included in the stemma of the critical edition, many manuscripts –indeed all except the Ka†mœrœ– insert after Mahåbhårata, 7.164.67ab the line: ‘With weaponry cast down, he can be killed in battle by mortal men’ (see 1305*, p. 953). The ambiguity and moral puzzlement that accompanies Kr¢ßña’s advice is evident even within the manuscript base itself! If we follow the Ka†mœrœ recension, as do the editors of the critical edition, then Kr¢ßña’s recommendation is only that Droña be disarmed, not that he himself be killed. This is a position which permits Kr¢ßña a little more moral leeway than any of the

184

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

other recensions, Northern or Southern, which force Kr¢ßña to assume full complicity in the murder of Droña, casting him as entirely conscious of the mortal consequences of his recommendation. Be that as it may, the varied response from the three present Påñ∂ava brothers to Kr¢ßña’s devious proposal reveals a great deal about their respective moral voices. Arjuna immediately and unequivocally dissents. Yudhiß™hira hesitates and falteringly condones. Bhœma embraces the idea, and with enthusiasm makes it his own. He kills an elephant belonging to the Påñ∂ava army, whose name also happens to be A†vatthåman, and then announces to Yudhiß™hira that A†vatthåman has been slain. Our narrator Vyåsa tells us that Bhœma, ‘keeping in his mind the fact that it was an elephant by the name of ‘Aßvatthåman’ that had been killed, spoke falsely’ (Mahåbhårata, 7.164.73). Droña falters, but is not duped. Indeed, he is provoked into going on a rampage, massacring large parts of the Påñ∂ava force and strewing the earth with their remains. Now, curiously, the noble sages of old are made to enter the field, wherein they entreat Droña to desist from his ‘exceedingly cruel acts’, which, they reason, are both unjust and unbecoming of a bråhmaña. Droña begins seriously to question the rectitude of his behaviour, and to wonder whether it is perhaps true what Bhœma said. He begins to worry, we might say, whether he is rightly avenging a deceit or wrongly perpetrating an atrocity while in denial about the truth concerning his son. In this state of moral perplexity he turns to Yudhiß™hira, whom he believes would never tell a lie. It would appear that Kr¢ßña also thinks the same, for he now inveighs Yudhiß™hira as he had previously done Arjuna: If Droña fights in anger for even half a day, I say truly that your army will meet with destruction. To protect us from Droña, a falsehood (anr¢ta) is better than the truth (satya). A falsehood uttered for the sake of a life is untouched by falsehood. (Mahåbhårata, 7.164.98-99)

Bhœma joins in, telling Yudhiß™hira about the dead elephant A†vatthåman, and Yudhiß™hira crumbles: Sinking in fear of untruth but addicted to the victory, Yudhiß™hira indistinctly said ‘king [he] is dead, or the elephant’. (Mahåbhårata, 7.164.106cd [avaktavyam abravid råjan hataΔ ku∞jara ityuta])

Those improper words 3 of Yudhiß™hira do the trick: Droña collapses in grief and, laying down his weapons, assumes the posture of a yogœ. But in a dramatic metaphor, Vyåsa conveys to us Yudhiß™hira’s 3 Taking the clause to be one of direct speech, as is usual in Sanskrit, the name ‘A†vatthåman’ is not used. Might this then be a rare instance of indirect speech in Sanskrit, the actual words uttered by Yudhiß™hira being ‘A†vatthåman is dead’, adding the ‘elephant’ in murmured apposition? The impropriety of the statement is not necessarily a matter of its being false; even if true, it is intentionally deceptive.

III. How to Produce, Construct and Legitimate a Tradition

185

fall from grace. His chariot, which had up until now floated a few inches above the earth’s surface, abruptly crashes to the ground! Yudhiß™hira will never again be taken at his word. Nor will things go as planned for the Påñ∂avas. For it will turn out that the real A†vatthåman’s vengeance is mightier than anything the Påñ∂ava conspirators had foreseen. Only Arjuna’s last-minute intervention will save them from complete annihilation. And we shall have cause to wonder whether or not it was Kr¢ßña who created this mess, provoking A†vatthåman into avenging his father, discrediting Yudhiß™hira, and putting Arjuna in an impossible moral position from which he would emerge wishing only for death. In the final analysis, is it not Kr¢ßña upon whose head falls the moral responsibility for this result? How, indeed, is Kr¢ßña’s advice here consistent with his own council on the eve of the mighty battle, when Arjuna was wracked with doubt about the morality of the war and his own part in it? Do we recognise here the ‘deontological’ Kr¢ßña, who lectured on the virtues of detached action in the Bhagavadgœtå? There, he seemed to offer Arjuna quite different advice, albeit still exhorting him to fight. The advice he gave there is that all action motivated by desire for the fruits or results of one’s actions leads only to further desires and so, ultimately to the propagation of suffering. For one can never satisfy all one’s desires, and the more one acts with the hope of getting rewards, the more one is liable to disappointment and frustration: When a man thinks about sense objects, an interest in them develops. From this interest grows desire, from desire anger; from anger rises delusion (sa∫moha), from delusion loss of memory, from loss of memory the death of the spirit (buddhi), and from the death of the spirit one perishes. (Bhagavadgœtå, 2.62-63)

If the doctrine of karman is right, then it seemed that those who make morality regulative of their activities will achieve integrated and stable selves, in the sense of having well-coordinated projects, plans and interests. Morality would not undermine our interests, but would rather be the necessary condition for there to be a harmony among them, of the type that pursuit of our overall good requires. But Kr¢ßña argues that none of this matters; for any goal-directed activity, if motivated by the desire to achieve one’s ends, will generate only further discontent and lead eventually to the disintegration of the self as the locus of a unified set of purposes. How then ought one motivate oneself to act, if not from a desire for the results of the action? What Kr¢ßña claims in the Bhagavadgœtå is that the only virtuous action is action free from desire for any result (nißkåmakarma): Your entitlement is only to the action, not ever at all to its fruits. Be not motivated by the fruits of acts, but also do not purposely seek to avoid acting. (Bhagavadgœtå, 2.47)

186

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

If action based on desire for its results leads only to a disintegration of the soul, and ‘non-action’, way of the ascetic and renouncer, is also ruled out, then what is left? Kr¢ßña, I think, transforms the older idea that deliberative action is related to self-constitution. For instead of saying that it is the pursuit of a coherent set of projects that leads to an integrated and stable self, Kr¢ßña says that this is a function achieved by performing the action itself, freed from all consideration of its karmic results. Deliberative action alone is what holds things together, and makes a person ‘connected’ (yukta): The connected man, renouncing the fruits of his acts, reaches the peace of the ultimate foundation, while the disconnected man, who acts on his desires because he is interested in fruits, is fettered by the consequences (karman). (Bhagavadgœtå, 5.12)

And it is by doing the action that is ‘natural’ or ‘proper’ to oneself, that one achieves this: “Each man attains perfection by devoting himself to his own task: listen how the man who shoulders his task finds this perfection”. (Bhagavadgœtå, 18.45)In summary, the attitude of Kr¢ßña in the Bhagavadgœtå seems to be this. We have to act, for action is what holds a person together as an integrated agent. But desire for the results of some plan of action cannot be a motivation for acting, since such motivations actually undermine rather than reinforce the agent’s integrity. Kr¢ßña’s advice to Arjuna is that he must act instead according to his particular duties and obligations, but remain detached from any self-interest in the results of his actions. To act in a way that is true to one’s self is what it is to act well, and the consequences, as prescribed by the principle of karman, can then only be good. But to be motivated to act by those consequences would be to lose sight of who one is and will lead eventually to the fragmentation of oneself as the source of a coherent set of goals and plans. Can we find any underlying pattern in these two interventions of Kr¢ßña, one at the very beginning of the battle and the other near to its end? Appearances notwithstanding, there is indeed a consistency in Kr¢ßña’s character. His role in the great epic, it seems, is to oversee the unfolding of a chain of events that is destined to be, and he intervenes whenever human beings threaten to throw things off-course, whether it be because of their moral weakness or indeed their moral strength. The battle was meant to take place, but Arjuna’s moral queasiness and willingness simply to surrender put it in jeopardy right at the outset; the battle was also meant to be won by the Påñ∂avas, and again this ‘proper’ outcome was thrown into doubt by Droña’s colossal strength of will. If human beings orient themselves in moral space with the compass of duty and rule, Kr¢ßña represents orientation by the pole-star, seeing to it that the final destination is reached even if the path taken must sometimes meander and backtrack.

III. How to Produce, Construct and Legitimate a Tradition

187

5. Competing moral voices Let us now complete the story. Seeing Droña assume the defenceless pose of a yogœ, Dhr¢ß™adyumna, who Bhœma has earlier deputised to protect Yudhiß™hira, and who is Bhœma’s lieutenant in matters both military and moral, attacks, fights and kills Droña. Adding insult to injury, Dhr¢ß™adyumna severs the head of Droña and parades it by its locks, much to the disapproval of the Påñ∂ava troops and of Arjuna in particular, who, we are later informed, had cried out to Dhr¢ß™adyumna not to kill Droña but to bring him alive. Droña ascends to the realm of Brahmå, but his ascent is witnessed only by Sa∞jaya, Arjuna, Yudhiß™hira and Kr¢ßña(and possibly also A†vatthåman); nobody else knows. Following Droña’s murder, indeed, (moral) panic sets in, causing disarray within the Kaurava army. Seeing this, and apparently unaware of his father’s death, A†vatthåman asks Duryodhana why the Kaurava troops are fleeing, and upon hearing the entire story, delivers his own moral assessment of the event: I have heard of the ignoble, wicked deed perpetrated by that impersonator of virtue, and I have heard of that very cruel ‘son of virtue’ [dharmaputra, i.e. Yudhiß™hira]. For those engaged in warfare, either victory or defeat is certain –one of these two must occur. There [in warfare], killing will be praised. Such killing of a fighter in battle as is in accordance with the rules is not worthy of grief, for it is thus seen by the sages. No doubt, my father has gone to the world of heroes. Reaching such an end, that best of men ought not be mourned. But the trustworthy and rightfully engaged [Droña] was seen by the whole army to have his hair grasped, and that cuts me to the core. From love or hate or contempt or pride or even immaturity, people perform unjust deeds, as again from disrespect.4 Dr¢ß™adyumna will see its very dreadful consequences; and also, having acted with the greatest ignobility, the falsespeaking Påñ∂ava. The earth will today drink the blood of that ‘lord of virtue’ who, by deceit (chadmanå), made the teacher surrender his weapons. (Mahåbhårata, 7.166.19-27)

A†vatthåman’s moral evaluation is very interesting. Blame does not attach as such to Dhr¢ß™adyumna for killing Droña, for there is nothing immoral about killing one’s enemies in battle. Blame does attach to Dhr¢ß™adyumna for humiliating Droña after death, and it attaches too to Yudhiß™hira, for the deceit that made Droña’s death unfair. A†vatthåman holds Yudhiß™hira morally responsible for the murder. Any attempt to defend Yudhiß™hira with the argument that Droña’s death was an unintended outcome of his actual intention, which was only to make Droña desist from battle –an appeal to the ‘doctrine of double effect’– is categorically rejected here. Precisely because Yudhiß™hira resorted to deceit, he is morally responsible for the ‘double effect’ his misdeed produced. A†vatthåman will not per4

See Månavadharma†åstra, 8.107.

188

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

mit Yudhiß™hira to shirk responsibility, either through the connivance of the muttered word ‘elephant’ which clothed Yudhiß™hira in false virtue, or because someone else did the actual killing. Yudhiß™hira is responsible because Droña trusted him, and by trusting him, put himself in Yudhiß™hira’s care (he allowed Yudhiß™hira’s world to become his own). Yudhiß™hira failed in the duty of care his very trustworthiness created. Yudhiß™hira did not protect Droña, whom he himself had made dependent on Yudhiß™hira for his protection. We will see that Arjuna’s moral assessment of the episode is a little different. He is soon to be called upon again, for it is now revealed that A†vatthåman has in his possession a divine weapon known as the nåråyaña, a weapon A†vatthåman received from his father Droña, who was himself given it by the lord of all creation, Nåråyaña, in return for his piety. The nåråyaña is a weapon so powerful that it can slay the unslayable, and its chief feature is that it will turn upon its owner if used unfairly, injudiciously, or repeatedly. That too is its symbolic function –it is a weapon that represents the reversal of fortunes, the inversion of fate; as a weapon of illusion, the illusion it stands for is the mirror image. Sure enough, the employment of the nåråyaña soon results in an inversion of fortunes for the two warring factions: now it is the Påñ∂ava troops who flee from the battle in a state of disarray and defeat, now it is Yudhiß™hira who inquires, as A†vatthåman had done before, as to the cause of his army’s (moral) panic. Replying, Arjuna delivers up the sternest of rebukes in a speech that is his definitive moral assessment of the event: The teacher was told a deceit (mitthyå) by you, your honour, as a means to gaining the kingdom. Done by one who knows what is right (dharmaj∞a), this is a very great wrong. ‘The Påñ∂ava is endowed with all virtues, and he is also my pupil; he will not speak falsely’, such was the opinion he formed of you. You told our teacher that ‘he is dead, the elephant’; this being but a falsehood wearing the truth as an armour-skin (satyaka∞cuka∫ nåma praviß™ena tato ’nr¢taΔ). Laying down his weapons, he became dispassionate and insensible […] For even though I cried out mightily as one who loves his teacher, unheeding of his duties, the pupil killed the master. (Mahåbhårata, 7.167.33-41) Ignobly and with light heart, we performed that act of treachery against the learned teacher in order to gain the kingdom. My teacher believed that for my love of him I would abandon my sons, brothers, father, wife and indeed life itself. But I neglected him even though he was being killed, because of my desire for the kingdom. Therefore, o mighty king, with head hung down I go to hell. For having caused the death of the unarmed, wise and sagely bråhmaña teacher, it is better to die than to live. (Mahåbhårata, 7.167.47-50)

Arjuna’s verdict is clear: a crime has been committed, the murder of an innocent, pious and unarmed man. The motive was base selfinterest, the method was underhand, and the opportunity came when Droña disarmed himself. This passage stands in curious relationship to

III. How to Produce, Construct and Legitimate a Tradition

189

the apparent moral of the Bhagavadgœtå. Is Arjuna saying that he now regrets taking Kr¢ßña’s advice, allowing himself to be drawn into an unworthy fight simply because he is a kßatriya and that’s what kßatriyas do? Or is he saying something else, that Kr¢ßña had taught that one should act without regard for the fruits of one’s action, and this cardinal principle of moral motivation has been violated? But it was Kr¢ßña himself who had urged them to deceive Droña. Notice too that Arjuna differs from A†vatthåman in his moral assessment of the event. While A†vatthåman had identified the deceit but not the killing of Droña as the source of moral injustice, Arjuna locates the injustice in the killing itself. Yudhiß™hira is certainly held up for criticism, not because of his fall from truthfulness, but because of his complicity in Droña’s death. Kr¢ßña had recommended deceit as the only way to make Droña desist from his frenzied rampage, but he had not said that Droña should be killed (not, at least, in the critical edition –this defence of Kr¢ßña’s moral conduct is unavailable in the less forgiving recensions of the text). The suggestion was a skilful way to bring an end to the carnage without resort to violence. A lie is justified when it non-violently prevents violence –it is a paradigm of non-violent activism. Arjuna shares the commitment to non-violent struggle, but he is more of a pacifist. Like Gandhi, he is willing to die for his beliefs but not kill for them, and unlike Kr¢ßña he is not willing to cheat with other moral rules even to save himself. If Kr¢ßña is the Dionysian moral hero of the tale, Arjuna is its apollonian epic fall-guy. It is now Bhœma’s turn to provide a moral assessment of Droña’s death, which in his case takes the form of an apologetics. His principal argument is that Droña’s behaviour was morally culpable, not because he went on a rampage but because in doing so he violated his caste, behaving more like a kßatriya than a bråhmaña. The Påñ∂avas, on the other hand, were just doing what kßatriyas do. Yudhiß™hira, furthermore, had not actually told a lie. Rather, he had fought one illusion by means of another: You cut us to the core with these words, o destroyer of foes, that are like acid being poured into the wounds of injured men. You break my heart, which has been pricked by those thorn-words. An ethical man, you profoundly misunderstand the unethical, for you do not praise either yourself or us, though we be praise-worthy. (Mahåbhårata, 7.168.14cd-16) Departing from his own duty (svadharma) and resting in the duty of a warrior (kßatradharma), this doer of evil deeds kills us with weapons not of human origin. Calling himself a bråhmaña, he summoned an illusion (måyå) of an unendurable kind, and by an illusion (måyayå) has he today been killed. Arjuna, what is improper in this? (Mahåbhårata, 7.168.24-25)

The fight of illusion with illusion is, indeed, the leitmotif of this book, and Bhœma is correct so to redescribe the event. What remains unclear is whether it is permissible to use any illusion to fight anoth-

190

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

er, or whether there are limits to the morality of deceit. Bhœma’s moral voice one of social role and ethical duty, ‘my station and its duties’ in the language of Bradley. If Yudhiß™hira embodies one strand in the closely woven ethics of the Bhagavadgœtå, and Arjuna another, then Bhœma manifests yet a third, the voice of caste, hierarchy and social order. 6. On truth and lies in the moral sense Yudhiß™hira deceived Droña but did not lie. His infamous muttered utterance of the word ‘elephant’ in apposition with ‘A†vatthåman is dead’ preserved truth at the expense of trust. In resorting to this linguistic manoeuvre, Yudhiß™hira placed himself in the unlikely company of the Catholic casuists of medieval England who, faced with St. Augustine’s absolute prohibition on lying in De Mendacio and Contra Mendacium, but forced to speak under conditions of persecution, took their lead from St. Thomas Aquinas’s remark that ‘it is licit to hide the truth prudently by some sort of dissimulation’, and introduced the new doctrines of equivocation and mental restriction, whereby one made what one said strictly true but intentionally deceptive. Yudhiß™hira, however, was a special case –his need to resort to casuistry resulted from his personal reputation for absolute truthfulness rather than any universal prohibition on lying. Indeed, in the Hindu epic and religious literature, a general endorsement of the virtue of truthfulness is balanced against a recognition that there can be circumstances in which it is permissible and even obligatory to lie. Kr¢ßña’s recommendation of a straightforward lie in an exceptional circumstance is, in fact, more squarely within the prevailing moral framework than Yudhiß™hira’s austere conception of what it is to follow a moral rule. An acknowledgement that truthfulness is among the principal virtues is already to be found in the Chåndogyopanißad: When a man is hungry, thirsty, and without pleasures –that is his sacrificial consecration; and when he eats, drinks, and enjoys pleasures –by that he performs the preparatory rites; when he laughs, feasts, and has sex –by that he sings the chants and performs the recitations; austerity (tapas), generosity (dåna), integrity (årjava), non-injury (ahi∫så), and truthfulness (satyavacana) –these are his sacrificial gifts. (Chåndogyopanißad, 13.7.1-4)

Truthfulness is situated here within a framework of co-dependent virtues, themselves related to a conception of human flourishing through a ritual homology. The passage already raises the question of the relationship between truthfulness and other supporting or commeasurable virtues, and with that the possibility that truthfulness might come into conflict with other virtues of equal centrality. The idea that one only injures oneself by lying is hinted at in the Pra†nopanißad:

III. How to Produce, Construct and Legitimate a Tradition

191

Then Suke†a Bharadvåja asked him: “Hirañyanåbha, a prince of Kosala, once came to me, Lord, and asked this question: ‘Do you know the person consisting of sixteen parts?’ I told the prince: ‘I don’t know him. If I had known him, how could I have not told you. Up to his very roots, surely, a man withers when he tells a lie (anr¢ta). That’s why I can’t tell you a lie.’ He got on to his chariot silently and went away. So I ask you: who is that person?” (Pra†nopanißad, 6.1)

In the Chåndogyopanißad (4.4.1-5), we were told the story of Satyakåma. Satyakåma (lit. ‘one whose desire is for truth’) confesses to his teacher the obscurity of his origins, and the very frankness of his admission permits his teacher to conclude that he is indeed a bråhmaña, for only a bråhmaña would speak truly without regard to the humiliation. It might seem that Satyakåma could have avoided injury to himself (in the form of humiliation) by telling a lie. The morale of the story, though, is that speaking the truth trumps the preservation of oneself from embarrassment or other such forms of emotional harm. Later thinkers do not regard it as obligatory to tell the truth when there is a question of self-preservation from physical harm, or of doing and preventing harm to others. Manu succinctly brings out the logical relationship in such cases of conflicting obligation: A man should tell the truth and speak with kindness (priya); he should not tell the truth unkindly nor utter lies out of kindness. This is a constant duty. (Månavadharma†åstra, 4.138)

If we take ‘kindness’ here to include the virtues other than truthfulness, Manu’s clever formula implies that the ‘constant duty’ (sanåtanadharma) is to joint satisfaction of all the virtues, there being no categorical obligation to one when it is in conflict with another. One is not obliged to tell the truth regardless, when doing so involves one in an ‘unkindness’, nor is one obliged to lie even when doing so would be ‘kind’. In a situation where the only choice is between speaking truly but unkindly or speaking untruly but kindly (the so-called ‘paternalistic’ deceit), an appeal to the standing virtues will not be able to guide one’s action. Yudhiß™hira found himself in exactly such a situation, having to choose between a truth that would lose the battle finally for the Påñ∂avas and a lie that would decisively win it. Arjuna recommended the first course of action, Kr¢ßña the second, and Bhœma a suspiciously casuistic fudge. The problem with the casuistic resolution is that it is hard to see how there can be any moral difference between deception by assertion and deception by conversational implicature; and if that is right, then, ironically, it is more honest simply to lie. In the twelfth book of the Mahåbhårata, the ‡åntiparvan, Bhœßma is on his death-bed, and at Yudhiß™hira’s request enters into a lengthy moral discourse. If Yudhiß™hira’s deeds and misdeeds in the course of

192

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

the battle are not explicitly referred to, this is nevertheless the place where a moral assessment of them is attempted. Of particular significance, then, are the sections where Bhœßma discusses the moral status of truth and truthfulness (Mahåbhårata, 12.110; 12.140; 12.156). Bhœßma stresses more than once that it is not always immoral to lie, and that one has to use one’s reason and intelligence in each particular case to discriminate between the moral and the immoral: Therefore, O son-of-Kuntœ, one who is wise and self-restrained should dwell in this world resorting to his intellect (buddhi) in the discrimination of the moral and the immoral (dharmådharmani†caya). (Mahåbhårata, 12.139.94)

We have just been told the story of a bråhmaña Vi†vamitra who chooses to eat dog-meat rather than to starve, the morale being that the preservation of one’s own life justifies erstwhile violations of religious duty. Yudhiß™hira’s response is extremely telling: he says that if an act so detestable that it resembles a lie is permissible, then moral duty (dharma) itself is made loose, and there is no act from which one should desist (Mahåbhårata, 12.140.1). In raising the spectre of moral chaos, Yudhiß™hira reveals the fear that motivates the inflexible rulefollower and displays again his moral cowardice. Unable to trust his moral judgement, he prefers instead a blind allegiance to the moral law. Bhœßma proceeds to give Yudhiß™hira a sharp lesson in the necessity for kings to have sound practical reason! A king should not depend on a morality derived from one faction alone, but must use his intellect (buddhi) to draw wisdom from a variety of sources and examples. For the moral sometimes assumes the outward form of the immoral; the sin involved in killing one who ought not to have been killed is on a par with the sin of not killing someone who ought to be killed (i.e. one cannot simply follow the rule ‘do not kill’ [Mahåbhårata, 12.140.26ab]). Yudhiß™hira then asks Bhœßma if there is any rule at all which permits of no exception, to which Bhœßma responds that the only rule is to worship learned and pious bråhmañas as if they are gods. A cruel turn! –Up to this point, Yudhiß™hira might have hoped to find some comfort in Bhœßma’s words for his decision to deceive Droña, but now it is almost as if Bhœßma plays a trick on him. The rulefollower’s desperate need is for a rule to follow –then let it be ‘don’t kill revered bråhmañas’ rather than ‘don’t tell lies’. Yudhiß™hira’s rulefollowing morality leads down a blind alley. How then ought one’s moral reason guide one when what is at stake is whether to tell the truth or tell a lie? More than once in the Mahåbhårata are inserted verses from the dharma†åstra giving a list of exemptions to the general prohibition on lying (e.g., following Mahåbhårata, 7.164.99 [in some mss.]; 12.110.18ab). Typically, five exemptions are noted in the dharmasütras, although there is no great consistency in the list:

III. How to Produce, Construct and Legitimate a Tradition

193

A man may tell a lie at a marriage; during a sexual encounter; when his life is at stake; when there is a risk of losing all his property; and for the sake of a Brahmin. These five types of lies, they say, do not entail loss of caste. (Vasiß™hadharmasütra, 16.36 [trans. Olivelle]) According to some, telling a lie at a marriage, during sex, in jest, or in grief is not a sin. (Gautamadharmasütra, 23.29 [trans. Olivelle]) When speaking truly would lead to the death of a †üdra, vai†ya, kßatriya or bråhmaña, one should tell a lie, for that is better than the truth. (Månavadharma†åstra, 8.104)

Apparently, then, it is the generally accepted norm that a lie carries no moral sanction when it falls into one or more of the following categories: — — — — —

the protective lie (defence of self or others, esp. kin and bråhmaña), the dutiful lie (benefiting someone to whom one owes respect, esp. teachers), the marital lie, the jocose lie, the amorous lie.

A ‘pernicious’ lie, by contrast, is a lie that benefits the liar and harms someone else. It is curious indeed that Kr¢ßña should appeal to the rough-cut norms of conventional morality in his attempt to persuade Yudhiß™hira, if indeed his designated role is that of moral innovator. Kr¢ßña’s argument, in effect, is that Yudhiß™hira’s lie falls into the category of the protective lie. Arjuna dissents from this evaluation: for him, the lie is pernicious, deriving solely from a desire to reacquire the kingdom. Kr¢ßña and Arjuna have differing moral assessments, then, of Yudhiß™hira’s atypical speech-act. How are we to decide which of them is right? On what grounds can we decide if Yudhiß™hira was, after all, displaying a virtuous disposition, the disposition to trustworthiness in speech, where that denotes a reliability to say the ‘right’ thing, the thing the particular circumstance calls to be said? In one section of the ‡åntiparvan, the book of the Mahåbhårata where some explicit form of moral assessment of the epic tale is attempted, this question is addressed head on.5 Yudhiß™hira asks Bhœßma our question, and coming from his lips the question has an unmistakable poignancy: Yudhiß™hira said: Bhårata, how does one live who desires to reside in morality (dharma)? Wise bull of Bhårata, do reply to this inquiry! Concealing the world, truth (satya) and falsity (anr¢ta) occur together. Of these two, O King, which should one practice who is settled on morality? What is truth? What is falsity? And what is the morality con5 John Brockington brings my attention to a closely parallel passage at Mahåbhårata, 8.49.21-55.

194

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

stant for all (sanåtana)? At which times ought one speak the truth, and at which times ought one to speak falsely? (Mahåbhårata, 12.110.1-3) Bhœßma said: Speaking the truth is a good (sådhu), and there is nothing higher than truth. I will tell you, Bhårata, something that is very difficult in the common world to understand. When the truth would be false, and also when the false would be true, one should not speak truly; one should speak falsely. For this reason, a fool is in a state of confusion when not resting on the truth. Discriminating between truth and falsity, then, is the one who understands morality. (Mahåbhårata, 12.110.4-6) A person might obtain great merit though he be ignoble, stupid and cruel; as for instance Valåka, who slew the blind monster [by mistake –he was simply out to kill]. How extraordinary that a fool, desiring to be moral without understanding morality, might obtain to great sin; [yet another] instance is the owl on the bank of the Gaõgå. Your question here is similar: it is extremely hard to say what is moral. Although very hard to find a reply, this will be attempted now by means of the use of reason (tarka). Morality (dharma) has been described as for the sake of the glory or dignity (prabhåva) of creatures, from which morality is determined to consist in non-violence (ahi∫så). Morality is, it is said, that which upholds and maintains (dhårañå), creatures being supported by morality. From that, morality is determined to consist in maintenance. Some people say that morality has its basis in the scriptures, but other people deny this. We do not condemn this, for not all morality is based in scriptural prescription. (Mahåbhårata, 12.110.7-12) One ought never speak with those who, wanting another’s money, seek to please. This is certainly the right thing to do. If one is able to get free by remaining silent, one should not make a sound. If one has to speak, or if not to speak would invite distrust, then it is better to speak falsely than the truth –this is the considered opinion. (Mahåbhårata, 12.110.13-15ab.)

Bhœßma displays greater sensitivity to what B.K. Matilal has aptly termed the ‘elusiveness’ and ‘ambiguity’ of dharma (Matilal 2002b) than is always apparent in dharma†åstra writers like Manu (e.g., ‘truthfulness is better than silence’ [Månavadharma†åstra, 2.83]). Sometimes it is wrong to tell the truth; sometimes, indeed, it is right to lie. When? Bhœßma’s answer is clear: when one conceals the truth from a person who has no right to the truth. To give somebody something of value (a truth, a piece of wealth) to which they have no rightful entitlement is not to display any virtue; indeed, it is to fall victim to a vice. A person of sound moral reason is able to discriminate and decide for themselves in each particular case whether the person who makes a demand has a right to what is being demanded, the basis for that judgement involving an assessment of the person’s reasons for wanting it. One might indeed incur a great moral wrong in blindly following some such moral precept as ‘do not lie’, as the example to which Bhœßma alludes reveals. This is the case, mentioned in the Karnaparvan, of a man who reveals the whereabouts of an innocent fugitive to a would-be assassin. Bhœßma’s argument defends lies of concealment, when the person from whom one is concealing the truth does not have a right to the truth. On the

III. How to Produce, Construct and Legitimate a Tradition

195

other hand, his argument gives no succour to pernicious lies, lies that manipulate others when they have not asked for the truth, usually for one’s own gain. Now a new subtlety in the Droña story begins to unfold. The lie that deceived Droña –was it a lie of concealment or a lie of manipulation? Did Droña himself request the information, or was he ‘fed’ that information with the intention of contorting his view of the world? We see now a clear difference between the sentence ‘A†vatthåman is dead’ as uttered by Bhœma and the same sentence as uttered by Yudhiß™hira. In the mouth of Bhœma, the sentence is a manipulative lie, intended to produce in Droña a false belief he had until then no reason even to suspect. But the seeds of doubt were laid, and by the time Droña turns to Yudhiß™hira, he is demanding the information from him. Sense can now be made of the puzzling intervention by the great sages of the past, who suddenly appear and inveigh Droña to desist. The cumulative effect is to reinforce in Droña’s mind the necessity of demanding the information from Yudhiß™hira. And now the question is no longer whether to manipulate Droña by feeding him lies, but rather the issue is whether Droña has a right to the truth which he demands. The ethics of concealment have replaced the ethics of manipulation, and we have entered the territory of the honest lie. Is the attempt to classify Droña with the bandit and the inquisitor successful? One reason to think that he does not, after all, have a right to the truth about his son from Yudhiß™hira is that he is ‘trading upon’ Yudhiß™hira’s virtue, exploiting his reputation for truthfulness for his own ends, and surely virtue is not so easily abused. Droña, we might say, is trying to be a ‘free-rider’ on Yudhiß™hira’s goodness. Yudhiß™hira simply beats him at his own game (thus ‘One who uses illusion [måyå] should be met with illusion; one who is good should be answered with goodness’ [Mahåbhårata, 12.110.26]). That is, indeed, one great lesson of the Droñaparvan –the need to fight illusion with illusion, the requirement not to play straight against a crooked opponent. In that sense, Kr¢ßña is right; but Arjuna is right too in his moral assessment, for when one plays dirty against the dirty, there is no hope of emerging with one’s honour, dignity and self-respect intact. Arjuna is right that the Påñ∂ava warriors can feel only regret for what they have done, but wrong to imagine that choosing the beggar’s life would be, ethically, better (Bhagavadgœtå, 2.2). If Yudhiß™hira was later to feel regret for what he had done, perhaps it was the kind of regret any ethically sensitive person would feel when they have had to deceive another human being for the sake of the greater good, rather than the kind of regret that is mixed with shame, when one regrets doing something one knows one should not have. Our good poet Vyåsa naturally cannot leave it at that. Yudhiß™hira, we are reminded, uttered the words being ‘desirous of victory’, and pathetically muttered ‘elephant’ under his breath. That

196

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

was no casuistic attempt to outwit the inquisitor, but a feeble concession to vanity. And if it is true that Yudhiß™hira’s motive was victory, then the lie was simply pernicious. True to form, Vyåsa again muddies the dark waters of moral reflection. 7. The virtues of truth The Mahåbhårata is a sustained moral reflection on the value of truth and truthfulness. Is truth an intrinsic good or simply an instrumental one? If intrinsically good, does that imply that there is nothing more to be said in its defence? Or is there some way to explain why something is of intrinsic value without reducing that value to others? Bernard Williams has an extremely insightful discussion of the notion of an intrinsic value, and specifically of the idea that truthfulness is an intrinsic value (Williams 2002). He claims that for something to be an intrinsic good it must be stable under reflection, which requires that […] the agent has some materials in terms of which he can understand this value in relation to other values he holds, and this implies, in turn, that the intrinsic good, or rather the agent’s relation to it, has an inner structure in terms of which it can be related to other goods. (Williams 2002: 92)

He continues: For us to get clear about trustworthiness as an intrinsic good, we need to answer two kinds of question. First, we have to decide what disposition or set of dispositions trustworthiness is; as we might also say, what it needs to be. […] Second, we have to see what those other values may be that surround trustworthiness, values that provide the structure in terms of which it can be reflectively understood […]. That [structure] has been differently understood in differing cultural circumstances. Everywhere, trustworthiness and its more particular applications such as that which concerns us, sincerity, have a broadly similar content –we know what we are talking about– and everywhere, it has to be related, psychologically, socially, and ethically, to some wider range of values. What those values are, however, varies from time to time and culture to culture, and the various versions cannot be discovered by general reflection […]. Sincerity has a history, and it is the deposit of this history that we encounter in thinking about the virtues of truth in our own life. This is why at a certain point philosophy needs to make way for history, or, as I prefer to say, to involve itself in it. (Williams 2002: 92-93)

We are interested in the history of sincerity in the Indian tradition. It is extremely interesting, therefore, to find Yudhiß™hira asking precisely the two questions Williams says need to be asked, and to hear Bhœßma answering by situating the value in question (here truth) within a framework of values and emotions that help to make sense of it as something of worth:

III. How to Produce, Construct and Legitimate a Tradition

197

Yudhiß™hira said: When it comes to morality, the gods, the fathers and the sages all commend truth. I want to learn about truth; tell me about it, O Grandfather. What is the indicating mark of truth, O King, and how is it to be secured? What might truth obtain, and how? Tell me this. (Mahåbhårata, 12.156.1-2) Bhœßma said: Bhårata, mixing the moral duties of the four castes is not recommended, [yet] the truth is unchanged among all the castes. For the good, truth is always morally right, truth is the morality constant for all (sanåtana). One ought submit oneself to truth alone, for truth is the highest path. Moral duty is truth, as is austerity (tapas) and mental discipline (yoga); brahman is truth, constant for all. Truth, it is said, is a high ritual. On truth, everything stands. Having spoken thus of the customary forms of truth, I will now describe in sequence its indicating marks. And you must also hear about how truth is secured. Bhårata, among all people, truth is of thirteen kinds. Without doubt, truth is impartiality indeed, as well as self-control; it is freedom-from-envy, toleration, modesty, patience and freedom-from-spite; it is renunciation, contemplation, nobility, steadiness, perpetual calmness and non-violence –these, O King, are the thirteen aspects of truth. Truth is thus indeed imperishable, eternal and unchanging. Not in conflict with any moral duty, it is secured by means of mental discipline (yoga). (Mahåbhårata, 12.156.3-10) […] These thirteen forms are the several indicating marks of the one truth; they partake of, and they speak of, the truth, Bhårata. One cannot speak enough of the qualities of truth, and that is why the gods, the fathers and the sages commend truth. There is no higher morality than truth, nor a greater sin than falsehood. Truth is the foundation of morality; therefore, one should not suppress truth. From truth comes generosity, rituals with offerings, rituals with fire, the Vedas, and indeed everything else determinative of morality. Truth has been held in the balance against a thousand a†vamedha horse sacrifices, and truth indeed outweighed them. (Mahåbhårata, 12.156.22-26)

We have been accustomed, at least since Kant, to see the value of truth as grounded in a sense of respect for the autonomy of others; to deceive is to manipulate, and that is a way to treat the other as a means to one’s own advancement and not as an end in themselves. A more archaic conception situates truth in a framework of ethical emotion: to be true to one’s word is a matter of honour, and with dishonour comes public shame and private remorse. Bhœßma here gives voice to a complex third structure of values and sentiments that ground a sense of the intrinsic value of truth. A different list in the Bhagavadgœtå situates truthfulness within an even wider matrix of virtues: Fearlessness, inner purity, fortitude in the yoking of knowledge, liberality, self-control, sacrifice, vedic study (svådhyåya), austerity, uprightness, nonharming (ahi∫så), truthfulness, peacableness, reliinquishment (tyåga), serenity, loyalty, compassion for creatures, lack of greed, gentleness, modesty, reliability (acåpala), vigour (tejas), patience, fortitude, purity, friendliness, and lack of too much pride comprise the divine complement of virtues of him who is born to it, Bhårata. Deceit, pride, too

198

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

much self-esteem, irascibility, harshness, and ignorance are of him who is born to the demonic complement, Bhårata. (Bhagavadgœtå, 16.1-4)

Not all the virtues in this wider matrix are virtues specifically involved in the maintenance of a practice of truth. Bernard Williams has classified the virtues of truth into two general categories –virtues of sincerity and virtues of accuracy. Sincerity is a disposition towards trustworthiness in speech, a reliability to do what one has said one will and say what one believes. Accuracy is a disposition towards honesty in belief, the promotion of forms of conduct that ensure that the beliefs one acquires are consistently correlated with the way things really are. A third virtue of truth –the virtue of receptivity– needs recognition. This is the disposition to allow a truth, once believed, to have its full weight within the structure of one’s mental space, the disposition to resist the insulation of belief. Receptivity and accuracy are the virtuous counterforms of concealment, sincerity the counterform of manipulation. Let us consider in more detail how each of the thirteen virtues listed by Bhœßma comes to be a form of a virtue of truth. One outcome of this exercise will be to achieve a clearer assessment of the respective moral strengths and weaknesses of Yudhiß™hira, Bhœma and Arjuna, Kr¢ßña and A†vatthåman. Helpfully, the text supplies definitions of the truth-related virtues: (1) Impartiality (samatå). Impartiality is sameness as between what is good for oneself and what is bad for one’s enemy, and is grounded in the elimination of both desire and aversion, love and hate. (Mahåbhårata, 12.156.11) (2) Self-control (dama). Self-control is constantly not to envy another, steadiness/fortitude and depth, freedom from fear and anger. It is acquired by knowledge. (Mahåbhårata, 12.156.12) (3) Freedom from envy (amåtsarya). Freedom from envy was called by the wise generosity and dedication to one’s duty. One becomes free from envy by constantly remaining with truth. (Mahåbhårata, 12.156.13) (4) Toleration (kßamå). In the matter of toleration and intolerance, a good man tolerates entirely the agreeable as well as the disagreeable. It is acquired by a good man, possessing the truth.(Mahåbhårata, 12.156.14) (5) Modesty (hrœ). Out of modesty, someone whose mind is tranquil does much that is good and does not boast. It is acquired by moral duty. (Mahåbhårata, 12.156.15) (6) Patience (titikßå). Patience, also called forbearance (kßånti) is that by which one tolerates for the sake of moral duty. It is acquired by steadiness/fortitude, and has as its purpose the encouragement of ordinary people. (Mahåbhårata, 12.156.16)

III. How to Produce, Construct and Legitimate a Tradition

199

(7) Renunciation (tyåga). Renunciation is the renunciation of both love and material possession. For one who has cast off both attraction and aversion is there renunciation, and not for any other. (Mahåbhårata, 12.156.17) (8) Nobility (åryatå). He who exerts himself to do good deeds for creatures is said to be noble. It consists in freedom from passion, and is without a form [of its own]. (Mahåbhårata, 12.156.18) (9) Steadiness (dhr¢ti). Steadiness is that by which one remains unperturbed in pleasure as in pain. The wise man who wishes for his own good fortune should always cultivate it. One should always have toleration and then truth. The clever man who casts off joy and dread and anger acquires steadiness. (Mahåbhårata, 12.156.19-20) (10) Calmness (sthira) [no definition]. (11) Non-injury (adroha). Not injuring in action, thought or speech is a constant duty of a good man. (Mahåbhårata, 12.156.21) (12) Kindness or freedom from spite (anugraha or anasüyatå). Kindness too is a constant duty of a true and good person. (Mahåbhårata, 12.156.21). (13) Generosity (dåna). (Mahåbhårata, 12.156.21) So too is generosity [but the original list has contemplation –dhyåna– instead]. Why is it these virtues that render intelligible the practice of truth within an Indian context? The underlying theme is that of the mind as unsettled by emotions like fear, anger, greed or even attachment and attraction. A mind thus wracked by strong emotion will place the satisfaction of desire above the determination of truth. The practice of truth makes sense only within a framework of virtues that puts great weight on steadiness of mind. The steady mind is the one that will be objective and impartial, unbiased by its own needs. In this Indian context, the practice of truth is strongly associated with the cultivation of a mind free from directive passions. Once more it is receptivity which emerges as the cardinal virtue of truth, ahead of both accuracy and sincerity. When, then, might it be consistent with a practice of truth to conceal one’s beliefs from others? When the disclosure would threaten the very stability of mind upon which the practice of truth rests. If this provides occasional justification for lies of concealment, if offers no justification for the manipulative lie. For to deceive someone by manipulation is to distort their relationship with the world, and that will always threaten to undermine the calmness and steadiness of mind upon which their retention of the practice of truth depends. To manipulate another with a lie is, in that sense, to harm their well-being. Concealment is a weapon of resistance, and is consistent with the maxim of not harming (“to reserve to

200

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

a Mans Selfe, a faire Retreat”, as Bacon put it). Manipulation is a weapon of aggression, and is not. Gandhi characterised the relationship between the maxim of not harming (ahi∫så) and the practice of truth (satyagraha) as a means-ends relationship, and this captures some of what Williams meant in speaking of an “inner structure”; but we should not think of the relationship in excessively instrumental terms, for what is in question is not how the practice is made, but how it is made sense of, and that is what preserves each of the virtues involved as itself an intrinsic good. Making sense of revisionås in the practice in accordance with the immersed critical norms of the tradition is also what permits one to say of such a revision that it is a genuine development in the tradition. Bhœma, we can clearly see, lacked all the virtues on our list and saw no harm in lying. Arjuna possessed them all, and could not bring himself to lie or to condone the lie. Yudhiß™hira’s possession of the virtues was present but shaky; his practice of truth was not properly grounded in a solid cultivation of the virtues of truth, but remained at the level of a rule to be followed. He did not really see that in deceiving Droña he was not only doing him harm, but was doing him exactly the sort of harm a proper understanding of the value of truth would never permit. By corrupting Droña’s relationship with the world, he deprived Droña of the ability to sustain the very calmness of mind that would have brought the carnage to an end: he deprived Droña of the capacity to be human. Had he been Kr¢ßña, he might have attained of a higher-level understanding of his action, one that sees it in terms of a necessary realignment of Droña’s attitude, an attitude that has become misshapen. Yudhiß™hira, however, does not see Droña as someone in the grip of a delusion from which he must be deceitfully freed. Yudhiß™hira sees Droña as a force that must be stopped, even at the expense of Droña’s own humanity. He lacked the ‘reflective understanding’ of the practice of truth that would have permitted him either to endorse the lie as Kr¢ßña did, or to follow Arjuna and reject it. He knew that it was good to be truthful, but he had no insight into the framework of correlative virtues that go to makes sense of truthfulness as a good in itself. Kr¢ßña managed to impart some of that understanding to Arjuna, but had less success with Yudhiß™hira, and none with Bhœma. This inability to comprehend the point of a virtuous practice was the source of Yudhiß™hira’s great moral failing, and it would in the end cost the Påñ∂avas dear. 8. Conclusion The idea that morality has a history is a difficult one, for morality is more commonly thought to consist in universally applicable and timeless principles. In our analysis of truthfulness in the Indian tradition, however, we have seen how there is a progression and a development in the virtue of sincerity. Kr¢ßña’s role is that of a ‘moral

III. How to Produce, Construct and Legitimate a Tradition

201

expert’ who recommends that the tradition change: the underlying idea of sincerity as a matter of not harming others in speech is better fulfilled in a complex disposition not to undermine their relationship with the truth than in an adherence to any simple rule on lying. This change is a genuine development, and not a corruption, of the tradition, to the extent that it brings out in a richer form the underlying principle of the tradition, in such a way that other participants to that tradition can makes sense of it, or ‘acknowledge’ it, as such. The extent to which they are able to do so is what the complex moral deliberation in the Mahåbhårata, given voice in its respective players, seeks to establish. In the Mahåbhårata we see directly a dynamic moral tradition employing its immersed critical principles in a process of genuine development.

BIBLIOGRAPHY Bacon, F. (1897), Essayes or Councels, Civill and Morall, J.M. Dent & Co., London. Buitenen van, J.A.B. (1981), The Bhagavadgœtå in the Mahåbhårata: text and translation, University of Chicago Press, Chicago. Chong Kim Chong, Yuli Liu (eds.) (2004), Conceptions of Virtue East and West, Marshall Cavendish Academic, Singapore. Dharmasütras: The Law Codes Åpastamba, Gautama, Baudhåyana, and Vasiß™ha, annotated text and translation by Patrick Olivelle, Motilal Banarsidass, Delhi 2000. Hacker, P. (1965), Dharma im Hinduismus, in «Zeitschrift fr Missionswissenschaft und Religionswissenschaft», 49, pp. 93-196. Halbfass, W. (1988), India and Europe, State of New York University Press, Albany. Matilal, B.K. (2002a), Kr¢ßña: in defense of a devious divinity, in Id., Epics and Ethics, Oxford University Press, Delhi. Matilal, B.K. (2002b), Elusiveness and ambiguity in dharma-ethics, in Id., Epics and Ethics, Oxford University Press, Delhi. Nussbaum, M., Sen, A. (1989), Internal criticism and Indian rational traditions, in Krauz, M. (ed.), Relativism, Notre Dame. Newman, J.H. (1890) [1845], An Essay on the Development of Christian Doctrine, Longmans Green, London. O’Hear, A. (1998), Tradition and traditionalism, in Craig, E. (ed.), Routledge Encyclopedia of Philosophy, Routledge, London. Smith, B.K. (1989), Reflections on Resemblance, Ritual and Religion, Oxford University Press, New York. Williams, B. (2002), Truth & Truthfulness, Princeton University Press, Princeton.

FRANCIS X. CLOONEY, S.J. From Person to Person: A Study of Tradition in the Gurupara∫paråsåra of Vedånta De†ika’s ‡rœmat Rahasyatrayasåra The ‡rœmat Rahasyatrayasåra (‘The Auspicious Essence of the Three Mysteries’) by Vedånta De†ika (or Veõka™anåta, 1268-1369) is a kind of summa theologiae that proceeds largely by exegeting the three holy mysteries or mantras (tirumantra, dvayamantra, and carama†loka [Bhagavadgœtå, 18.66]) cherished by the south Indian ‡rœvaißñava tradition. The Rahasyatrayasåra is a comprehensive presentation of ‡rœvaißñava exegesis, philosophy, theology, practice and religious sociology, all woven into a single treatise expressive of what it means to believe and practice the truths and values inscribed in the mantras as read by ‡rœvaißñavas. The Gurupara∫paråsåra is an additional, introductory chapter prefixed to the Rahasyatrayasåra, of no more than several thousand words. In this essay I examine the ‡rœvaißñava understanding of tradition by a close reading of the Gurupara∫paråsåra, attending to the content and style, as well as the sources cited in the text and its commentarial heritage. The Gurupara∫paråsåra offers an intracommunity rationale for the ‡rœvaißñava choice of founding the tradition in the person of the åcårya, succinctly arguing that the student/reader should mindfully recollect the lineage of ‡rœvaißñava åcåryas whenever beginning to study the mantras and related religious topics. As an articulate presentation of tradition by an important Indian intellectual, it is also a resource for understanding tradition more widely in India. For the most part, I read the Gurupara∫paråsåra on its own, even if it can and ultimately should be read also in relation to the Rahasyatrayasåra and to others among De†ika’s works, such as the Sa∫pradåyapari†uddhi. I will, however, be reading it in light of the numerous commentaries on it, a value I will say more about later in this essay.

204

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Before interpreting the text in detail, a preliminary comment is in order on the notion of tradition. I am working with a theological notion of ‘tradition’ rather than a strictly philological or sociological delimitation of ‘tradition’ in an Indian context. From a theological perspective, numerous factors and features must be taken into account from the start: tradition as a commonplace fact of the continuation of the past into the present, as a heritage is transmitted in a series of present moments; the ways in which religious communities have both tradition and traditions, and also a normative determination of the past honored not merely as ‘tradition’, but as ‘the Tradition’; tradition as inscribed in written and oral sources; tradition as the content that is passed down, and as the process of teaching perpetuated over generations; tradition as founded in privileged scriptures but also specified by later teachings considered authoritative; the recognition of those permitted to speak for a tradition, teachers holding various degrees of authority; exclusory rules enunciated in order to determine what is not traditional. For the sake of a theological analysis that is broad and sensitive to the rich meaning ‘tradition’ may have in the life of a religious community, I have found particularly helpful Yves Congar’s Tradition and Traditions (Congar 1997). This monumental study is dedicated largely to sorting out various meanings of ‘tradition’ considered from various angles. Congar admits that the matter is complex enough to defy simple definition –hence the great length of his study– but several passages serve as starting points for a reflection on the ‡rœvaißñava context as well: Tradition means, in itself, a transmission from person to person. It thus implies a living subject. From the point of view of its content, tradition in this most primitive and general sense requires merely a deposit of some sort. This deposit or content can include writings, as well as words, actions, rules of conduct and institutions […]. (Congar 1997: 296)

In a smaller accompanying volume, Congar cites a concise definition of M. Dufrenne: “Tradition, in the true sense of the word, implies a spontaneous assimilation of the past in understanding the present, without a break in the continuity of a society’s life, and without considering the past as outmoded” (Congar 1964: 8). He goes on to distinguish content (‘passive tradition’) from the action of transmission (‘active tradition’) by divine and human persons 1 “[…] written texts, in the total content of tradition, only imperfectly satisfy the special character of tradition, which is required to be a transmission from person to person, and thus presupposes a living subject […]. Hence, if, taking tradition in its widest meaning, we include the writings, we can also take it in a narrower, less extensive sense […] objective tradition is defined, in contradistinction to Scripture, as that which is transmitted otherwise than by writing, at least originally” (Congar 1964: 298).

III. How to Produce, Construct and Legitimate a Tradition

205

due to whom texts are formed and given content, and also the written from what is oral.1 In each case, the more dynamic category is primary –and most pertinent in interpreting the ‡rœvaißñava understanding of tradition. The study of tradition is necessarily the study of the community (ecclesia) in which key figures –leaders, teachers– undertaken to pass down what God has given: [Divine] assistance is given to the ecclesia as such, as the adequate subject for the gifts of the covenant and more particularly among these, of tradition. It is distributed in the ecclesia according to the place and role which the different members have to take. Ordinary members have their share in this assistance within the unity of the whole to live by and according to the covenant; the members who occupy a leading place profit from this assistance to exercise their right and duty as leaders, with an authority which implies the right to command and be obeyed. These are the charisms of the magisterium, classically defined by its charge to guard faithfully and to stare or explain authentically the deposit entrusted to the ecclesia as the Bride of Christ. (Congar 1997: 302)

The Gurupara∫paråsåra is consonant with Congar’s approach, offering a rich notion of tradition that is centered on persons. De†ika does not use technical terminology to describe tradition, but only a series of terms that give a feel for his understanding of tradition, in an ordinary manner: para∫parå (lineage); sa∫pradåya (community, tradition); va∫†a (lineage); upade†a (teaching); aru¬i ceytal (graciously effecting); ko™u (to give); kë¬ (to hear); prakå†a (glorifying); avißkåra (making manifest); pravartanam (putting into play). The terms are unsurprising: teaching, hearing, giving and receiving, etc. Some, like ‘glorifying’ and ‘making manifest’, suggest that tradition is a kind of revealing –uncovering– as the åcåryas glorify and make known sacred truths. The terms indicate an active, person-based sense of tradition, as the transmission of the efficacious knowledge of the tradition –on tattva (the real), hita (the valuable, beneficial), purußårtha (the ultimate human goal)– from one generation to the next, in the work of particular teachers. The various terms suggest an active process in which the tradition is passed down, by persons, teachers, who are responsible for this reliable handing down. Tradition is never abstract, because it subsists in a series of particular, chosen acts of handing down, in the lineage of gurus (gurupara∫parå), the personal and personalized linking of one generation to the next by teachers who choose students who in turn also become teachers. Yet even the most pertinent terms –guru (‘spiritual guide’), åcårya (‘master’), and de†ika (‘instructor’)– seem not to be used in technical senses, and the commentators seem not to take them as such. Therefore, except in the citation

206

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

of passages from the Gurupara∫paråsåra itself, where I preserve De†ika’s own terms, I generally use the neutral term ‘teacher’. The Gurupara∫paråsåra organizes its reflection on tradition in a way that accentuates a mix of argument with rhetorical and stylistic strategies: a) b) c) d) e) f) g)

Opening †loka. Two Tamil verses on the tradition. That having a teacher is essential for liberation. The Lord as first teacher and at work in the lineage of teachers. A recollection of the succession of actual teachers. Devotion for the teacher is essential. Verses: one Tamil, two Sanskrit, three Tamil.

These sections may be divided into four parts: an opening †loka (a); a thesis (c) and its developed reaffirmation (f); the account of the Lord’s involvement in the world as the justification underlying the reverence for teachers (d); supportive verses (b, g); and, in the middle of all this, an actual recollection of actual teachers, by name (e). In the following pages, we shall explore the meaning of tradition in each section, reserving the verses to the end. Now let us turn to a more detailed analysis of the text. 1. The Thesis of the Gurupara∫paråsåra The thesis of the Gurupara∫paråsåra is previewed –uttered, enacted– in the opening †loka: ‘We declare our reverence for our gurus and in turn for their gurus; here, at the beginning, we invoke the couple ruling the worlds’ (pp. 1/1).2 To honor the teachers is not a separate practice, apart from reverence for the divine couple, since they are all part of one plan. The thesis of the Gurupara∫paråsåra is then succinctly stated: As they say, ‘Kßatrabandhu, the worst of sinners, and Puñ∂arœka, the merit-doer, both were liberated due to having an åcårya. Therefore, have an åcårya’. This affirms that having an åcårya is for every one the cause for liberation. (pp. 17-18/2)3

As all the commentators agree, the point of this passage is both the assertion of necessity, and the comprehensiveness of that assertion, for there is no one who does not require a teacher. Merit does 2 Throughout, all references to the Gurupara∫paråsåra will be marked according to the Viraraghavachariar edition, and the translation by Rajagopalan, to which I am indebted, although all translations are my own. Thus, the numbers 1/1 refers respectively to Vedånta De†ika, ‡rœmad Rahasya Trayasåra (ed. 1980, 2 voll.) and to Vedånta De†ika, ‡rœmad Rahasyatrayasåra (trans. Rajagopala Ayyangar 1956). 3 Kßatrabandhu is referred to in the Vißñudharmapuråña, and Puñ∂arœka in the Padmottarapuråña.

III. How to Produce, Construct and Legitimate a Tradition

207

not gain liberation, nor does sin make it impossible. Having a teacher is decisive, for everyone. De†ika nowhere in the Rahasyatrayasåra fully explains this claim. Rather, it is itself an instance of traditional thinking, an assertion that is accepted by those who are members of the community, and likely rejected by those who thus show and define themselves to be outsiders. Just as the opening †loka, as a prayer, made clear that actually practicing reverence matters, the thesis paragraph is itself followed by a practical application: ‘The preceding has been uttered to indicate that those desirous of liberation should meditate on the line of åcåryas up to the supreme Lord’ (pp. 19/2). The truth one learns about the universal need for a teacher must always then be instantiated in a regular practice of recollecting those teachers. 2. Location in the Tradition by Citation Rather than arguing directly the need for tradition, De†ika embeds his thesis in tradition in several ways. He repeatedly cites respected texts selected from a multi-layered tradition composed of both Sanskrit and Tamil texts, including the following: Mahåbhårata; Vißñupuråña; Bhågavatapuråña; Jåyåkhyåsa∫hitå; Vißñudharma; ‡eßasa∫hitå; Gautamadharmasütra; Ŭavañ™år (Yåmuna), Stotraratna; Periyåÿvår, Tirumoÿi; Maturakavi, Kaññi Nuñ Ciruttåmpu (by paraphrase); Chåndogyopanißad, Ka™hopanißad, and Muñ∂akopanißad (by paraphrase); implicit references to stories found in the Dharmottarapuråña and the Padmottarapuråña; plus unidentified smr¢tis. This is a list by which De†ika can display the tradition itself as a woven array of con-texts, all richly, mutually referential when understood properly. The list itself is not surprising, since it includes rather standard sources that a ‡rœvaißñava might be expected to use. One can still note, however, that the range of texts, Tamil as well as Sanskrit, in practice reinforces the point that both language traditions matter. Even without any elaborate defense, he is drawing the language traditions together and making a statement about their cohesiveness. There is likewise no notable separation between †ruti and smr¢ti, or between scripture and later tradition. Here, as with respect to the relationship between divine action and human cooperation, boundaries are flexible. As we shall see below, there is a theological underpinning, pertaining to God’s action, that authorizes this and even wider arrays of diversity. 3. Divine Improvisation and the Authority of Tradition The preceding claims on the necessity and warrants for tradition, particularly in light of apparent difficulties, are given a theological justification in the core section of the Gurupara∫paråsåra dedicated

208

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

to divine action. The reliance on tradition is justified by the melding of divine sanction with an undeniably fragile lineage of teachers since, in De†ika’s view, the Lord is at work at every point in the tradition. There is only a provisional boundary between human and divine action, as the Lord works directly and through teachers. Nor are the teachers assessed as deficient versions of God, since the point is indeed that there is no substantive difference between the divine and human action, in this case. This key theological section of the Gurupara∫paråsåra (pp. 20-28/3-5) begins by asserting that the Lord alone is the primal teacher: The Lord of all is the highest åcårya for the whole world. As it says, ‘The one who possesses all excellences, the åcårya, father, guru […]’ (Mahåbhårata, Sabhåparva 41.21); ‘Nåråyaña, guru of all worlds, is also my guru’ (Vißñupuråña, 5.1.14); ‘You are my kinsman, and you are my guru’ (Gandharœ, Mahåbhårata); and ‘You alone are my relative, and you alone are the guru […]’ and ‘You are the guru, you are the goal of the worlds […]’ (Stotraratna, 60).

The Lord is, moreover, a teacher who, though perfect, is willing to persist in teaching whenever necessary: He is the one who, in the beginning, gave the Vedas to Brahmå. When they were stolen he gave them again. He put into play the †åstras by his own mouth, and so too by the mouths of Brahmå’s sons, Sanatkumåra and others. As it says, ‘Of themselves they came to knowledge, and they were established in right action and the cessation of action’ (Mahåbhårata, ‡åntiparva 349.71). He thus put into play what is beneficial.

Again, the point is made that the Lord chooses to work through human teachers: In addition to all this, lest the spiritual tradition suffer decline, he established many great r¢ßis, Nårada, Parå†ara, ‡uka, ‡aunaka, and others. He entered Vyåsa and others and put into play the Mahåbhårata, Brahmasütras, and other teachings. As it says, ‘Know that Vyåsa, Kr¢ßña Dvaipåyana, is Lord Nåråyaña himself. Who else on earth, O Maitreya, could be the author of the Mahåbhårata?’ (Vißñupuråña, 3.4.5), and, ‘After thus praising the great r¢ßi, hands joined, Bhœßma spoke […]’ (Mahåbhårata, Ådiparva 114.40).

His descents are thus interactive, alongside the activity of human teachers: By descents such as the swan, the fish, Hayagrœva the horse, NaraNåråyaña, and the åcårya of the Bhagavadgœtå, he himself stood forth and glorified the true, the beneficial, and the human goal. He established men abundant in knowledge, ‡rœ Bhœßma, and others, and made deep-rooted the meanings he had graciously effected. He thus took the

III. How to Produce, Construct and Legitimate a Tradition

209

lordly teaching he had graciously made in the beginning –as it says, ‘Nåråyaña himself is the speaker of the whole Pa∞caråtra’ (Mahåbhårata, ‡åntiparva 359.68)– and made it again manifest at various opportunities. As it says, ‘By brahmins, kßatriyas, vai†yas, and †üdras, all adorned with proper marks and steadfast in their proper duties, he was honored and served; in order to establish the Såttvata rule, he was sung by Sa∫karßaña at the end of the Dvåpara Age and the beginning of the Kali Age’ (Mahåbhårata, Bhœßmaparva 66.39.40).

In what seems to be a culminating, second series of descents in the form of human poets, the Lord sings through the åÿvårs: Then he made ten new descents in the forms of åÿvårs such as Paråõku†a, Parakåla and others. As it says, ‘In the Kali Age, the unfailing one enters into various persons already born, and does what he desires’ (Vißñudharmapuråña, 108.50). Just as clouds take up the water of the ocean and then pour it down as rain for the sustenance of all, he thus synthesized and expressed the most essential portion of the meaning of the Vedas in a language for which all were competent.

A key reason why the Lord comes repeatedly in the form of teachers is to engage in argument and refute heterodox teachers, those who reason without respect for tradition: Lest due to heterodox teachers –evident and concealed– there be any obstacle to this good path he had put into play, he descended in the revered land of Agastya by the teaching of many teachers. As it says, ‘The god, Nåråyaña himself, having made a mortal form, lifts up the drowning worlds, due to his compassion and by his hand, the †åstras’ (Jåyåkhyasa∫hitå), and so too, as it says, ‘Wearing a kavi robe, as a Brahma-guru the Lord came […]’ (Periyåÿvår, Tirumoÿi, 5.2.8).

South India is thus conceived as a land blessed with a divine insistence and inventiveness in making the teaching that enables liberation effective in every age: All this the great r¢ßi has made clear: ‘During the Kaliyuga, highly fortunate devotees of Nåråyaña, will be born in large numbers and in various places in the land of the Dravi∂as, where flow the rivers Tåmraparñœ, Vaikai, Pålå, Kåviri, and the Mahånådi running to the West’ (Bhågavatapuråña, 11.5.38-39).

The overall effect of this discourse on divine activity and improvisation, contained in our section of the Gurupara∫paråsåra (pp. 2028/3-5), is to emphasize and undergird theologically the importance of the teachers, in and through whom the Lord has worked, and to value their presence as not inferior to earlier revelatory interventions. The text and the content of texts are subordinate to the person of the teacher; the Lord is present, working in and through the teacher. In a sense, the tradition is a tradition of divine persons.

210

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

4. Naming the Persons of the Tradition At the beginning of the Gurupara∫paråsåra, readers were instructed to remember their lineage of teachers. Near its end, as we shall see, the importance of the right relationship of teachers and their students, in whom the tradition subsists, is made clear yet again. In the middle, we find the enunciation of specific lineages, the naming of names; enjoining remembrance as essential to tradition, De†ika actually engages in an act of remembrance. He gives names and faces to a ‡rœvaißñava tradition hitherto praised in general terms. The lineage is too lengthy to include here, but we can note that it combines three objects of remembrance. First, there is (partial) familial lineage, then a lineage of teachers, and third, a remembrance of compositions. In the beginning we find the mention of a first, influential teacher: ‘Among these åcåryas, Nåtamuni, son of Œ†varamuni, graciously composed the †åstra known as the Nyåyatattva, and also the Yogarahasya’ (pp. 28/5). Next, almost as a justificatory aside, we find an interesting explanation for how the authority of Nåtamuni is linked to that of ‡a™ak≠pa¡ (Nammåÿvår), the most important of the åÿvårs: ‘Nammåÿvår was his åcårya, because he was manifest to [Nåtamuni] in a yogic state due to the succession in the tradition beginning with Maturakavi, and by the words of Tiruvåymoÿi’ (pp. 29/5). These nearly parenthetical remarks serve to justify the continuity of the lineage back beyond the teachers –see again the section above on the improvisation by the Lord– and links Nåtamuni to the greatest of the åÿvårs. It highlights the improvisational nature of the lineage that mirrors the divine improvisation already noticed above: ‡a™ak≠pa¡ (Nammåÿvår) had no human teacher, but lived in meditation; awakened from meditation by Maturakavi, he taught simply by reciting to Maturakavi all his (over 1200) verses. Nåtamuni met neither of these åÿvårs, but by meditation and recitation of the few verses to which he had access, he induced ‡a™ak≠pa¡ to appear to him and recite his verses. It was a help that he could learn from Maturakavi through his Kaññi Nuñ Ciruttåmpu (on which we shall see more, below), even if there seems to have been no comprehensive memory of Tiruvåymoÿi and other works in the Divyaprabandham kept alive in the family of Maturakavi. Rather, Nåtamuni had to rediscover Tiruvåymoÿi more or less on his own initiative. Lineage, founded in the divine and human fidelity to tradition, is nonetheless not simply a monument to human virtue; divine intervention is required, even before Nåtamuni, to provide continuities where circumstances would only have fostered a forgetting of tradition. The recitation of the tradition then turns to family and learning lineages, for the tradition subsists in persons who are teachers or in the family of teachers:

III. How to Produce, Construct and Legitimate a Tradition

211

Nåtamuni’s son was Œ†vara Bha™™a. To Œ†vara Bha™™a was born Ŭavañ™år. The works of Ŭavañ™år are eight: Ågamapråmåñyam, Purußanirñaya, Åtmasiddhi, Œ†varasiddhi, Sa∫vitsiddhi, ‡rœgœtårthasårasa∫graha, the Stotraratnam, and the CatuΔ†lokœ. (pp. 30/5)

Strikingly without comment, De†ika ends his recollection of family lineage with an abrupt reference to two females: Co™™ai Nampi was the son of Ŭavañ™år. Co™™ai Nampi’s son was E¡¡åccå¡. E¡¡åccå¡ had four sons, one of whom was Pi¬¬aiyappar. Pi¬¬aiyappar’s son was T≠ÿappar. T≠ÿappar had two daughters. (pp. 30/5)

We cannot tell if there is a point being made here: does the fact of two daughters, and no sons, indicate the end of this particular familial tradition? There is emphasis on persons as proper ‡rœvaißñava families, but more importantly as teachers and students, some of whom also write commentaries and treatises. Indeed, the fact of natural family connections is quickly superseded by more pointed memories of who was chosen as a student by which teacher and chose to learn from that teacher. Tradition is always about the past, but it exists primarily in the free choices of its present practitioners. After a further listing of other teachers and disciples, De†ika focuses finally on Råmånuja (Emperumå¡år, ‘our lord’) and his learning at the feet of several teachers: Those taking refuge at the feet of Periyanampi were six: Emperumå¡år [Råmånuja], Malaiku¡iya Ni¡„år, Årya ‡rœ Ca™ak≠pa Tåsar, Añiyaraõkatta Muta¡år, Tiruvåykulam U™aiyå¡ Pa™™ar, Tirukkacci Nampi. Emperumå¡år studied the meaning of the mysteries at the feet of Tirukk≠™™iyür Nampi; he heard Tiruvåymoÿi at the feet of Tirumålai Añ™å¡; after reciting Tiruvåymoÿi at the feet of Ŭavañ™år Åÿvår, from him he heard the hymns of praise and the rest, and also the gracious good commentaries. At the feet of Tirumålai Nampi he studied the ‡rœmat Råmåyaña. Emperumå¡år’s writings are nine: ‡rœbhåßya, Dœpa, Såra, Vedårthasa∫graha, Gœtåbhåßya, Ci„iyagadya, Periyagadya, Vaikuñ™agadya, and the Nityam. (pp. 31-32/6)

Råmånuja is most prominent not because he is a pioneer or innovator, but rather because in him the collective wisdom of the older tradition, Tamil as well as Sanskrit, is gathered; and this in turn is due in part to his painstaking study of particular texts with particular teachers. The concluding words of the lineage again highlight the possibility and importance of local application, as if to make sure no reader is omitted: ‘The names of the chief disciples of Råmånuja may be learned from the respective tradition of each (reader, student)’ (pp. 32/6). This passing statement too is actually an act of faith in the reliability and vitality of tradition; however glorious the past may seem

212

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

to be, the tradition does in fact continue, and continues reliably among contemporary teachers, including those of each reader. 5. Honoring the Teacher Near the end of the Gurupara∫paråsåra we find reemphasis and implementation, with reference to repeated acts of reverence glorifying the teacher and care in protecting the sacred mysteries, the mantras: It has been said, ‘The wise should glorify their gurus and watch over the mantra with care. By not glorifying [the guru] or by exposing [that is, not watching over the mantra], wealth and age decline’. ‘One’s glorifying the guru and another’s omission to do so’ occur due to varying degrees of devotion to the guru. (pp. 32-33/6)

The reason for glorifying the teacher is clear: For the one whose devotion is focused on the guru just as on the Lord, everything he desires will become evident, as is well known from the Ka™hopanißad4 and Jåbålopanißad [Chåndogyopanißad, 4.4] texts and from the story of Sanjaya. (pp. 33-34/6)

Similarly clear is a warning about the dangers of disrespect and negligence: ‘When it says that those lacking this [devotion] will not gain a wealth of knowledge, this must be observed in regard to students’ varying degrees of knowledge’ (pp. 34/). Conversely, the teacher is instructed on the duty to be careful in teaching prospective disciples: It is well known from the story of Raikva, and other such stories, that when an åcårya does not immediately shed light on spiritual truths, even for those disciples who are rich in qualities, he will nonetheless not have his reputation disturbed. (pp. 34-35/7)

In turn, If an åcårya assumes, ‘What I have here is of quality’,5 and teaches [precipitously], then his reputation will be disturbed. As it says, ‘the disciple’s sin is the guru’s’. This may be exemplified with reference to Four-Faced Brahmå. Without reflecting on what would happen, he taught Indra, and so forgot his own knowledge of Brahman; the Lord of all made him understand it again only in reliance on Nårada, his own disciple. (pp. 35-36/7)

The conclusion again emphasizes the obligation to one’s teacher, 4 Rajagopalan, the translator, suggests that the reference is to Ka™hopanißad, 1.2.9: ‘This understanding [of self] is not [gained] by reasoning but only when spoken by another’. 5 That is, the teacher rushes to the judgment that he has a worthwhile student.

III. How to Produce, Construct and Legitimate a Tradition

213

It follows that in all circumstances a person should glorify his guru by overflowing devotion for the guru. So too, the mantra should be watched over and not be revealed to fickle persons not possessing the fullness of qualities prescribed for a disciple; otherwise, its great value, like that of a ruby casket containing precious jewels, will suffer, and so too the guru’s reputation will be disturbed. (pp. 37/7)

De†ika concludes by reaffirming right practice: ‘It is settled then with the force of an injunction that whenever a person meditates on these mysteries, he should meditate on the lineage of åcåryas’ (pp. 38/7).6 The study of the Rahasyatrayasåra is presumably to be in accord with this injunction; reading it is to be accompanied by a remembrance of the tradition out of which its wisdom has grown, which has preserved its holy mantras over the generations.7 5. Tradition’s Poetic Flourishing In addition to a clear presentation of arguments with supporting warrants, De†ika’s Gurupara∫paråsåra is also a skilled rhetorical and literary composition that shows tradition as a kind of experience and flourishing richer and more vivid than conceptual representations of the importance of åcåryas and tradition. This is evident particularly in a feature characteristic of the whole Rahasyatrayasåra, De†ika’s mixing of prose with introductory and concluding verses in Sanskrit and Tamil. He is a skilled and learned writer, quite adept at expressing himself clearly and persuasively in prose. Yet he chooses also to write rather elaborately in poetic form (Tamil and Sanskrit), and it is in these verses that we see how a tradition as it were exults in its tradition. In verse form, De†ika expresses in fresh terms, perhaps most intensely and fully, meanings otherwise stated in prose. In order to fill out De†ika’s understanding of tradition, let us examine just several of these verses. The opening Tamil verse of the Gurupara∫paråsåra tells us of the importance of the ‡rœvaißñava saints known as the åÿvårs; for right from the start, it is persons, particularly the master teachers who are these poets, in whom the tradition subsists most delightfully and gloriously. The åÿvårs invite the listener’s pleasure, since they make known in clear Tamil the meaning of the hitherto obscure Veda: Poykai Muni, Bhatattår, Pëyåÿvår, Kurukë†a¡ born at the cool Tåmraparñœ, Vißñucitta¡, 6 He adds that remembrance is a remedy even for dealing with outsiders, who are a perennial threat to those cherishing insider status: ‘This meditation on the åcåryas is also expiation for having conversed with those that are forbidden; ‘having conversed, let him meditate mentally on those doing virtue’ ’ (pp. 38/7-8). 7 The entire Gurupara∫paråsåra prefigures the chapters on the qualities and duties of the teacher (chap. 30) and student (chap. 31) at the end of the Rahasyatrayasåra.

214

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

pure Kula†ëkhara¡, and our Påñanåtha¡, Toñ™ara™ippo™i, and the light born in Maÿicai, and the prince of Maõkai with the sword and spear,8 making the hidden Veda shine forth in all the world: if we recite with clarity the beautiful garlands of Tamil they sung with delight, we understand clearly the real meanings of the unclear, hidden Vedas. (pp. 9/1)

While this first verse seems to state poetically what is said in prose consequently, it is also due homage to the Tamil poets in their own language, by which they have taught so illustriously. The point is intensified in the second Tamil verse, wherein De†ika indicates the greatness of devotion to the åÿvårs, by alluding to a brief song of ten verses by Maturakavi, the Kaññi Nuñ Ciruttåmpu. This ‘minor’ åÿvår’s devotion in that work was almost directly entirely to his teacher ‡a™ak≠pa¡, as De†ika’s verse tells us by comparing responses to Kr¢ßña, the divine cowherd who does all for his devotees, with responses to the teachers, who instruct their disciples on the right path: The cowherd descended for the sake of those who love him, to be their bliss, their refuge, all they desire to gain, their unfailing many relations, to change their desires, to be himself their gain, to remove their deeds, flood forth compassion, make reality understood, and make them himself; but even so, instead of his feet, sorrowless Maturakavi has grasped the feet of the one who rendered in Tamil the difficult secrets, and thus [Maturakavi] has revealed the ancient path to those trusting in the good paths. (pp. 14/1-2)

Students do well to trust their teachers completely, dispensing even with the notion that the superior goal is to attain God; this is still true, even after one readily acknowledges, as does Maturakavi, that God is the great benefactor whose ways of helping the devotee (by being their bliss, refuge, etc.) are well known and easy to enumerate, as he does in the verses of the Kaññi Nuñ Ciruttåmpu. Even so, devotion to a wise teacher is preferable, for experience of God lives on in the tradition. Verses near the end of the Gurupara∫paråsåra praise Nåtamuni and his successors in the ‡rœvaißñava community, again bringing to life and exemplifying the earlier stated insistence that one should glorify the teacher. These teachers are defenders against external and internal enemies: 8 Interestingly, ten åÿvårs are mentioned here, Maturakavi and (sadly) Åñ™åÿ, the sole woman among them, being omitted; De†ika seems to have done this in order to have just ten saints, to match the idea of the Lord’s ‘ten new descents’ mentioned above.

III. How to Produce, Construct and Legitimate a Tradition

215

May Nåtamuni and the others find their enjoyment right here, so they can block the arrow of desire harming me; their lives are varied, all pure, they are praised in the three Vedas they convey to us the divine river of the compassion of the Couple9 with whom refuge is to be taken in faith; they are untutored in the paths of envy, error, and deception, they are our teachers. (pp. 40/8)

In them, Hayagrœva –Nåråyaña as the horse to taught again the Veda Brahmå had lost– takes delight: With heartfelt pleasure, in the lotus of the hearts of our gurus in succession Hayagrœva is pleased to take his royal seat, as the pride of adversaries is cast down by the ocean roar of his neighing– our gurus are possessed of many good qualities; fastened above the mansion of the directions, shaken by gusts of wind and casting off like tufts the host of various positions, is their banner of victory, the good path made evident. (pp. 41/8-9)

The first of these two verses implicitly spotlights the wrong knowledge threatening individuals and a community spoiled by ignorance and vice; there are always those who are attached and ignorant, and they are likely to be misled. The second of the verses similarly glorifies and warns. Below we shall return to this verse in order to see how commentators read it, but here we can simply notice the mix of pleasure, triumph, and the defeat of enemies it captures: devotion, glorification, divine combat against adversaries, exultation in victory over those holding views not in keeping with the tradition. The ‡rœvaißñava community itself thrives on the mix of emotions De†ika expresses here, and presumably he hopes that his verse will arouse these same emotions in contemporary members of the tradition. The final three verses of the Gurupara∫paråsåra praise the primary teachers, Nåtamuni, Ŭavañ™år, and Råmånuja; to highlight their scholarly prowess is also to draw attention to the chief danger from outside the community: clever reasoning uninformed by tradition. The first exemplifies the style and tone of all three: Råmånuja adorns the world with his renown, the sage whose fame is an adornment to the world, a peerless elephant among those reason-givers who wreak great havoc along the right path of the forest books, he is the lord slicing the plantain of their arguments: 9

Nåråyaña and Lakßmœ.

216

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

we have reflected on the sweet excellences of his words and no longer even think of evil deeds. (pp. 42/9)

The verse is both exultant and fierce, thankful and aggressive; it demonstrates the imaginative power of the tradition, the images and narrative symbols accompanying its ideas and doctrines. Again, the point is to inspire readers to a greater imaginative identification with the tradition, as they learn to travel the narrow path of virtue and truth trod by those living in accord with the tradition. 6. Commentary as the Future of a Tradition To fill out our understanding of tradition in the Gurupara∫paråsåra, we must finally consider the fact and function of commentaries on the Gurupara∫paråsåra, and how these affect both our reading and the context in which the Gurupara∫paråsåra endures. Along with the Rahasyatrayasåra, this is a text that has been revered and studied with care in later tradition. Reading commentaries, especially multiple commentaries on the same text, is a daunting process,10 yet it is also exceedingly worthwhile. Some initial aids are clear: commentaries make clear how words and concepts are understood according to a tradition; they contextualize each sentence by unpacking implicit, wider meanings; they make explicit rejected alternatives or options only obliquely expressed; occasionally, doctrinal points are rehearsed and further elaborated in the commentaries; they identify and contextualize citations, and provide additional citations; later commentators cite earlier commentators and draw our attention to continuities even in the later tradition; their introductions remind readers of the reverential nature of the reading act shared by the commentators and contemporary reader; they enrich a text by elaborating, in their generation, meanings found there, and by weaving a broader web with further citations drawn from other, related sources; occasionally, they introduce new ideas not evident in the text commented on, and thus reveal new issues alive at the time of their writing. Commentaries thus enable readers to locate a text like the Gurupara∫paråsåra in a conversation rooted in the past, but also extended in further acts of commentarial reading, up to the 10 I have been consulting the following commentaries that are partly or in entirety at my disposal: The Såradœpikå of Sribhasya Srinivasacarya (cc. 1-5); the Såråsvådinœ of Vedanta Ramanuja (1-12), Gopala Desikar (13-32); the Såraprakå†ikå of Srinivasa (ed. by S. Venkata Ranganatha Mahadesika and Raghunatha Tatayarya Dasar, complete); the Vivarañœ of Srisaila Srinivasarya GPP, 1); the Pramåñasa∫graha (GPP, 1); the Såraprakåßikåsa∫graha of Sriparakala Samyamindra Mahadesika (GPP, 1); Sårabodhinœ of Sri Rangasankopayatindra (complete); the Sårårthabodhinœ of Gopalacarya (complete); a mañipravå¬a commentary (no specific title) by Tiruvahindrapuram Cetlur Narasimhacary Svami (complete); the Sårasa∫graha of Tuppul Venkatacharyadasan (1-5); the Sårartharatnåva¬œ (apparently also by Tuppul Venkatacharyadasan; 1-5); the Sa∫sk®tånuvåda by K.V. Nilameghacarya (complete); an English translation by M. Rajagopala Ayyangar (complete); and the Såravistara of Uttamur Viraraghavachariar (complete).

III. How to Produce, Construct and Legitimate a Tradition

217

present moment. They show us the conversation that actually developed around a great text among those who actually remembered it, one particular conversation out of all the possible ones that might have taken place. The tradition of a religious classic such as the Rahasyatrayasåra and its Gurupara∫paråsåra extends not only into its past, but also into its commentarial future, where the tradition of teaching the classic is remembered and written down. The lineage of teachers, even if primarily oral, achieves written form in the commentators, and by reading with them our apprehension of the tradition is enhanced. Commentaries are thus of great value. If we wish to say more about them and assess their value with some specificity, however, a problem arises. They are, indeed, traditional, and commentators write with a sense of allegiance to the commented text. As it were, they selflessly avoid the spotlight of separate attention, the kind of individualization that begs for separate attention and study. While the selflessness of commentary –‘we add nothing, we simply indicate what the master said’– makes perfect and powerful sense at the moment of reading, when all attention is focused on the primary text that is the object of reading, it also means that subsequent acts of writing (such as this essay) also naturally tend to discuss the primary text and not the commentaries, or perhaps only of themes treated in both text and commentary. In this situation, the commentaries keep getting lost sight of. To retrieve a sense of the commentarial contribution as marking the ongoing life and future of a tradition, here we attend to just one verse from the Gurupara∫paråsåra, and how commentators read it. It is a verse we have already seen above: With heartfelt pleasure, in the lotus of the hearts of our gurus in succession Hayagrœva is pleased to take his royal seat, as the pride of adversaries is cast down by the ocean roar of his neighing– our gurus are possessed of many good qualities; fastened above the mansion of the directions, shaken by gusts of wind and casting off like tufts the host of various positions, is their banner of victory, the path of the good made evident.

To understand it in light of the commentarial conversation, I examined its analysis in several of the commentaries available to me: the Dœpikå, Såråsvådinœ, Prakå†ikai, the Cëtlür (by Tiruvahindrapuram Cetlur Narasimhacary Svami), the Vistara of Uttamur Viraraghavachariar, and the Sårasa∫graha of Venkatacharyadasan. All perform the services mentioned above, and also help us to read more deeply. Two insights are standard, beginning with the Dœpikå, which I take to be the oldest of the commentaries available to me. First, the commentators uniformly note that the verse indicates a

218

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

twofold defeat of error by the teachers: the adversaries are humbled (‘the pride of adversaries is cast down’), but also as their various positions are overturned (‘the host of various positions’ is shaken off). This is a double act, since while the erroneous teachings are linked closely to the persons who teach erroneously, both the teachings and the teachers must be put to flight. The right teaching of the tradition, communicated by the right persons who are the teachers, naturally overwhelms the corresponding wrong teachings generated by unreliable thinkers. In this context, the commentators discuss in detail the image of the victorious banner snapping in the wind, as it indicates both the way in which the fluff –other views– are cast away, precisely due to the sheer glory manifest in the victorious teachers. Second, the first line –‘With heartfelt pleasure, in the lotus of the hearts of our gurus in succession’– indicates to the commentators the enjoyment of the Lord when he chooses to be in the hearts of the åcåryas specially (for he is already in all hearts, always); his delight is especially to be within those who teach properly. Tradition is indeed a serious business, but it is manifest particularly in the gladness of the hearts of those who teach and those who learn –a gladness in which God chooses to dwell. On the level of method, a key aspect of the commentators’ method –in addition to their conscientious exposition of each of De†ika’s words– is their expansion of context by additional quotations. Even in this, of course, they follow De†ika, who was himself following the custom of earlier commentators. Nonetheless it is a process that continues to be richly illustrative over successive generations. To illustrate this process, I draw attention only to Narasimhacary Svami’s seven citations in the Cëtlür comment on this verse. Two emphasize the importance of the Lord’s entering the heart, ‘with heartfelt pleasure, in the lotus of the hearts of our gurus in succession’: […] The tall one, dark one, Lord of dark hue has entered me, never to leave, he is deep inside me. (Periya Tiruvantå™i, verse 68) O bridegroom Lord, you gave up your accustomed rest in the cool ocean, and came running, to make your home in the ocean of my heart. (Periyåÿvår, Tirumoÿi, 5.4.9)

By his presence, the Lord is able to transform realities from bad to good; when he enters the hearts of the åcåryas, he cries out in delight, and his cry –the neighing of the equine Hayagrœva– echoes in their teaching that overwhelms the noise of adversaries. In turn, this transformative power is like that of the child Kr¢ßña who turned the poison in Pütanå’s breast into nectar: ‘Her breast-milk, mixed with her poison, became of good taste for the world’s teacher’ (Hariva∫†a, 53).11 11 Locating the citation in chap. 53 of the Hariva∫†a is in accord with the editor’s identification, at the parallel citation of the same text at Tiruvåymoÿi, 1.5.9, where the verse

III. How to Produce, Construct and Legitimate a Tradition

219

In his fourth and fifth quotations Narasimhacary Svami recalls that De†ika uses the image of a victorious banner also in two other works: ‘This is the victory banner of the Bhåßya composed by the lord of ascetics […]’ (Tattva™œkåbhåßya, introductory †loka 7);12 ‘The banner of victory, renowned in its excellence for subduing vices […]’.13 The final citations address why, if the teachers of old were so great, there is a need for yet more writing in De†ika’s time. The ‘gusts of wind’ (‘shaken by gusts of wind and casting off like tufts the host of various positions’) are glossed in light of verses such as the following, which explains why the truths of tradition require the fresh air breathed in a new treatise such as De†ika’s Nyåyasiddhå∞janam: ‘This work is begun in order to purify the truth that has been inundated by the dance of various positions, some true, some false, some mixed’ (Nyåyasiddhå∞janam, introductory †loka 2).14 That ‘the path of the good’ is not a new one made superfluous by the old, is indicated by the seventh and final quotation: Let it be that the system established by the ruler of ascetics [Råmånuja] is recent. So what? Let it be that others are old by comparison. So what? Tell me! Let the systems of Taõka, Drami∂a, Guhadeva, and the rest be free from any affliction such as would arise from obscurations of their systems! (Yatiråjasaptati, verse 57)15

That a work –such as Råmånuja’s– is new, relative to older and revered works of the tradition, does not count against it: a new treatise has a perfection of its own that does not impugn the value of the tradition that precedes it. Even if the old masters are ‘without affliction’, new compositions are still in order. Citation is a perfect commentarial strategy, since it allows the commentator to demonstrate versatility while yet not intruding his own voice into the conversation.16 The commentator speaks, but in revered words spoken long before. The broader effect of citation is to bring the wider tradition to bear in particular ways on Vedånta De†ika’s verse, itself now intensified by the fact of a wider text woven around the Gurupara∫paråsåra. By doing so, of course, Narasimhacary Svami also brings to the fore all the further memories related to the cited texts, how they were taught and commented on, etc. In addition, cited complementary verses often play a role elsewhere. is cited and elaborated by the Bhagavat Vißayam commentators. But I have not been able to find the verse in available editions of the Hariva∫†a, nor in parallel accounts in the Vißñupuråña or in the Bhågavatapuråña. 12 See Vedånta De†ika, Tattwateeka (Tattva™œkå): 2. 13 Narasimhacary Svami says this verse is from the CatuΔ†lokœbhåßya, but I have been unable to find it in either the CatuΔ†lokœ or its Bhåßya. 14 See Vedånta De†ika, Nyåyåsiddhå∞jana: 2. 15 See the translation in Raghavan, Lakshmi Kumari, Narasimhachary 1995: 344-345. 16 Here, as throughout, I use the male ‘he’, since there are almost no female commentators in the ‡rœvaißñava tradition.

220

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

For instance, the Hariva∫†a verse regarding how Kr¢ßña transforms Pütanå’s poison into sweet milk is cited at Tiruvåymoÿi, 1.5.9 by the commentators there: ‘We will not perish: to kill the giant deceitful demoness, evil boaster, the pure little child sucked out her poison milk like ambrosia […]’. There, the Hariva∫†a verse is woven into the Tiruvåymoÿi verse’s meaning in its own context, along with additional texts also cited there. Learned readers are expected to remember those other contexts and references when reading De†ika’s verse, and Narasimhacary Svami’s comments and citations. In the end, though, there is ultimately a single comprehensive context, wherein all texts and all teaching coalesce. One might call it ‘the Tradition’. For a final example of the role of commentary in the ongoing life, the future, of a tradition, I wish to draw notice to the Sårasa∫graha of Tuppul Venkatacaryadasan. As we have just seen, commentators such as Narasimhacary Svami meticulously cite older commentaries; but Venkatacaryadasan’s work is striking because it is almost entirely a weave of paragraph-length passages from older commentaries. The overall effect of this commentary is the rendering of a personal voice still almost entirely woven from the voices of the tradition. The commentators provide us with a visible context –a larger text often literally and legibly composed of multiple texts printed together– that circumscribes the original text or underlies it, even at the bottom of the page. Thus, in reference to the single verse we are considering, Venkatacaryadasan cites 17 different passages from the commentaries: Dœpikå (3 times), Såråsvådinœ (2 times), Prakå†ikå (5 times), Sårasa∫graha (3 times), Vivarañœ (4 times). Here, by way of paraphrase, are the beginnings of several of the commentarial passages he cites: Dœpikå: Since he has gained what he prayed for, De†ika now expresses his enjoyment […] Prakå†ikå: By unassailable proofs he has defeated very powerful adversaries, and has refuted their works by his own works […] Sårasa∫graha: Here he expresses his prayer for devotion to the guru in his heart; for by itself his heart is unable to establish his own positions or refute those of others. Or perhaps the verse is simply an act of praise […] Sårasa∫graha: Perhaps the verses show how the Lord causes gladness in the hearts of the good […] Vivarañœ: The neighing is the succession of the sounds a horse makes, a roar. Or his point is that by this, as if by neighing, the pride of the adversaries is struck down […] Dœpikå: De†ika does not speak jealously when he says that the Lord came into the åcåryas’ hearts to experience his own pleasure; his point is rather that like himself, the Lord experiences this object of enjoyment, as a pleasure for himself too […] Såråsvådinœ: Like the roar of the waves of the sea, neighing that comes in unobstructed, successive roars […]

III. How to Produce, Construct and Legitimate a Tradition

221

Såråsvådinœ: ‘fastened’ –fastened entirely; because they pervade all places, so here, by way of an exaggerated identification, he mentions as banners of victory the Nyåyatattva, Siddhitraya, ‡rœbhåßya, and other such works […].

The effect of this collection of citations from older commentaries is a richly textured explication of the verse, and a very vivid instance of how a ‡rœvaißñava intellectual chooses to speak in the voice of learned commentators who have gone before him. He becomes, in a sense, the voice of tradition: speak in the voice of those who have gone before, let your personal creativity be dedicated to a masterful voicing of their words. The commentarial conversation articulates how tradition is enacted, lived into its future; we can also presume that commentators always require an intelligent community of likeminded persons who in each generation hear and support new teachers, allowing the tradition to speak on. 7. In summary In the preceding pages, we have considered only a single, albeit good, instance of the ‡rœvaißñava view of tradition, as expounded in one key text by one key teacher. While the Gurupara∫paråsåra is complex and subtle, its major points are simple and clear: tradition is key to the life of a religiously defined intellectual community; tradition has to do first of all with the lineage of persons who are its agents, those who pass down the teaching from one generation to the next; each student of the tradition must continually remember and mindfully appropriate that lineage; God is accessible in a special way through that tradition of teachers. They are the living embodiment and conscious agency of the tradition enacting itself from generation to generation. Similarly, they are models for right thinking, over against those who merely reason, or merely learn from books. Key to the survival of tradition is the work of these learned and religiously mature persons who wisely transmit the whole tradition to well chosen students. The tradition survives even during moments of danger, loss, and decline, not due simply to text or creed, but because God works in that wise transmission. De†ika has thus nicely combined a high theory of the teacher –the åcårya, guru, is the cause of liberation– with a realistic sense of the fragility of tradition and a necessary faith in divine providence. Caring for tradition requires that readers of the Gurupara∫paråsåra likewise be persuaded to take tradition seriously, to enact the remembering it enjoins. The exhortation toward the end of the Gurupara∫paråsåra –glorify the teacher, watch over the content of the tradition– ventures a right balance between the openness of teaching and the protective, cherishing role of its most precious elements. The commentaries attest to the fruitfulness of De†ika’s perspective.

222

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

But still, one may ask, why does religious vitality depend on tradition, and why does tradition depend on authoritative teachers? In part, the answer is that there is humanly observable wisdom in such claims, and that alternatives such as ignoring tradition or reducing tradition to its content are unwise. But it is still a grand claim to identify the teacher as the cause of liberation. The point is perhaps simply that in fact ‡rœvaißñavas reverence åcåryas, and hence do not raise difficult questions such as, ‘Why teachers?’. That is, the argument in defense and praise of tradition is itself couched in traditional terms. Much of De†ika’s argument also appeals to authoritative texts, traditional sources that are likely to sway those inside the tradition; and he himself is an authoritative teacher of that tradition. The point argued is already assumed to be true, because authoritative teachers, who are not to be questioned, have made it. The arguments for tradition works for insiders to a tradition, and De†ika seems not to feel obliged to persuade those outside the tradition who, however they may agree with the notion of tradition as a lineage of wise persons, are not likely to defer instinctively to Nåtamuni, Ŭavañ™år, and Råmånuja. De†ika would not be surprised by the skepticism of outsiders. Tradition is always in danger since individuals are inclined to think on their own about religious issues; if so, outsiders are almost by definition unable to assess this matter properly, since their assessments will be based on their own thinking and on the books they read. There is more to be done, in any case, before this consideration of tradition as understood in the Gurupara∫paråsåra, and by extension in the ‡rœvaißñava tradition, can be counted as adequate. First, we need to review our conclusions in light of the whole of the Rahasyatrayasåra. As should not be surprising, De†ika’s theology of tradition is not for its own sake. Rather, the Gurupara∫paråsåra and the related chapters 30-31 of the Rahasyatrayasåra, are themselves at the service of the two major core elements of the Rahasyatrayasåra. First, there is the exegesis of the tirumantra, dvayamantra, and carama†loka. This reading comprises a full half of the text, since those mysteries must be understood properly, with insight, and according to the tradition. Second, understanding the mantras properly leads to an enactment of the truths and values of the mantras in the act of surrender (prapatti) that is the topic of chapters 7-22. A study of the full Rahasyatrayasåra will thus also demonstrate what it means to live tradition as idealized in the Gurupara∫paråsåra. Second, comparisons are needed with other views of tradition, first of all in the ‡rœvaißñava tradition. One might read, for instance, the Å„åyirappa™i Gurupara∫paråprabhåvam of Pi¡paÿakiya Perumå¬ Jiyar,17 a text that deals with the lives and importance of the åÿvårs, 17

See the edition of Krishnaswami Ayyangar, 1975.

III. How to Produce, Construct and Legitimate a Tradition

223

Nåtamuni, Ŭavañ™år, and Råmånuja, and the first generations of åcåryas after Råmånuja (but not up to the time of De†ika). It will also be important to make comparisons with other Indian views of tradition, a larger comparison now made possible by this volume itself. Third, there is a need for still wider comparisons with other theologies of tradition, such as the Catholic Christian understanding alluded to with reference to the work of Congar at the beginning of this essay. One might also very fruitfully engage more recent writings, such as David Brown’s two volumes on tradition (Brown 1999; 2000). Brown able finds his way through a rich set of theological, aesthetic, and historical resources; his stress on the creativity and necessary imaginative component of tradition provides an apt counterpart to De†ika’s view of tradition. For Brown, as for De†ika and the ‡rœvaißñavas, religiously significant truths and goods live in the sources and original revelation, but also in the subsequent strata of tradition in which individuals –ordinary believers, leaders, teachers, God– engage in acts of imaginative participation in the sources and thus, by living continuities, keep bringing tradition to life anew (Brown 1999: 1). It makes sense, too, to keep this vital and person-oriented understanding of tradition at the fore even in interreligious considerations. Tradition is not a burden blocking adaptation to new religious situations, but rather, when properly understood, the vehicle of a required new imagining and living that are faithful both to sources and the current interactions in the divine and human community. It may just be, however, that the teachers we reverently remember will be drawn from a much wider array of local communities than has ever been the case before.

BIBLIOGRAPHY Sources Pi¡paÿakiya Perumå¬ Jiyar, Å„åyirappa™i Gurupara∫paråprabhåvam, edited by S. Krishnaswami Ayyangar, Trichi, 1975 [1968]. Raghavan, S.S., Lakshmi Kumari, M.S., Narasimhachary, M. (ed.) (1995), Sri Vedånta De†ika’s Stotras, Sripad Trust, Chennai. Vedånta De†ika, ‡rœmad Rahasya Trayasåra with Såra Vistara of Sri Uttamur T. Viraraghavacharya, Ubhaya Vedanta Grantha Malai, Chennai 1980, 2 voll. Vedånta De†ika, ‡rœmad Rahasyatrayasåra, translated by M.R. Rajagopala Ayyangar, Agnihotram Ramanuja Thathachariar, Kumbakonam 1956. Vedånta De†ika, Nyåyåsiddhå∞jana with Two Old Commentaries, Ubhaya Vedanta Grantha Malai, Chennai 1976.

224

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Vedånta De†ika, Tattwateeka (Tattva™œkå), edited by Uttamur Viraraghavacharya, Ubhaya Vedanta Grantha Malai, Chennai 1974. Studies Brown, D. (2000), Discipleship and Imagination: Christian Tradition and Truth, Oxford University Press, New York. Brown, D. (1999), Tradition and Imagination: Revelation and Change, Oxford University Press, New York. Congar, Y. (1997), Tradition and Traditions: the Biblical, Historical, and Theological Evidence for Catholic Teaching on Tradition, Simon and Schuster, Needham Heights (MA). Congar, Y. (1964), Tradition and the Life of the Church, Burns and Oates, London.

CHRISTOPHER MINKOWSKI What Makes a Work ‘Traditional’? On the Success of Nœlakañ™ha’s Mahåbhårata Commentary 1

Preamble: The Deluge and After. One of the first stories to attract the attention of European readers of Sanskrit literature was the account of a deluge that floods the world. In an article on comparative mythology that he wrote in 1782, Sir William Jones drew attention to a version of the story that is found in the Bhågavatapuråña (Jones 1788: 230ff).2 In 1829, the comparative philologist Franz Bopp used a version from the Mahåbhårata, and published the Sanskrit text with a German translation (Bopp 1829a; 1829b).3 In the Mahåbhårata version, Manu is warned by a fish of an impending deluge. The fish instructs Manu to build a boat and to go aboard with the seven sages and a collection of seeds. This Manu does. The flood comes and the boat is floated on the waters. When Manu summons him, the fish reappears and tows the boat, which enables it to weather high seas and buffeting winds. Eventually the boat is brought to dry land on a high peak of the Himålaya. Here the story ends with the revelation that the fish who has saved Manu is the deity.4 1 A version of this paper was presented at the March, 2001 meeting of the Association for Asian Studies in Chicago, as part of a panel on Sanskrit Knowledge-Systems on the Eve of Colonialism. My thanks to Sheldon Pollock, Robert Goldman, Gary Tubb, and Yigal Bronner, (all of whom were participants in the panel), for their comments on an earlier draft of this paper, and also William Pinch, Farina Mir, and James Fitzgerald. 2 At that time Jones knew of the passage only through a Persian intermediary (Jones 1788: 249; Bopp 1829b: ix). The Sanskrit text is found at Bhågavatapuråña, 8.24. 3 Bopp notes with pride that his is the first translation of the Mahåbhårata (henceforth MBh) version of the story into a European language. The story is found at MBh 3.185 in the Critical Edition (henceforth CE), or at MBh 3.187 in the ‘Bombay’ or ‘vulgate’ text. Bopp produced a series of texts and translations of sections of the MBh in the years before the first full publication of the MBh appeared in Calcutta. 4 The deity identifies himself as the god Prajåpati or Brahmå (MBh 3.185.48 [CE]).

226

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

The interest of the story lay in its comparative possibilities, with the Biblical account of Noah and the Flood. For Jones, the Indian flood story could confirm both the historical truth of the Biblical account, and also the theory which underlay his comparative project, that the people and ancient civilizations of the world shared a common ancestry dating from the Flood (Trautmann 1997: 28-61). By the time of Bopp, the comparative project had become somewhat different, and more philological. Deferring larger conclusions, Bopp was especially interested in establishing which version of the Manu story was the oldest, and in getting the primary text and its translation right. Imagine, therefore, the reaction of Bopp and other early Indologists when they encountered the Sanskrit commentary on the Mahåbhårata’s flood episode by the seventeenth century scholar Nœlakañ™ha Caturdhara. For in reading Nœlakañ™ha they were quickly immersed in the technical vocabulary of advaitavedånta. While both Jones and Bopp wanted to find something allegorical in the flood story, Nœlakañ™ha’s sort of allegorizing was certainly not it. According to Nœlakañ™ha, the story’s problematic is the ontological possibility of jœvanmukti, that is, the possibility of continuing embodied life after spiritual enlightenment.5 In the story, Manu is the mistaken egoism (aha∫kåra); the fish is the jœva, the delimited form of the metabrahman, which is ensouled in a succession of gross bodies.6 The boat that Manu builds is his last human embodiment.7 Tying the boat to the mountain is the ending of self-ignorance (avidyå), and the jœva vanishes just as the fish does, since the egoism subsides upon re-identification with brahman.8 Life as an individual can go on for a while even after enlightenment, the story teaches us, just as a spinning wheel continues to turn for a while on its own momentum.9 5 “Previously the tree of created existence, whose root is self-ignorance (avidyå), arose on the earth of consciousness. When self-ignorance is destroyed, it dissolves into the earth, such has been the teaching. Once self-ignorance has been destroyed, then, there should be no apparent cycling of the wheel of births and deaths (sa∫såra). And yet the cycle appears to continue for Mårkañ∂eya and other enlightened sages, whose self-ignorance has been destroyed, just as it does for (unenlightened) people like us. The king, wondering about embodied life after enlightenment (jœvanmukti), spoke to Mårkañ∂eya, so the narrator tells us” (Bhåratabhåvadœpa [henceforth Dœpa] on vs. 1 [pürvatra cidbhümau avidyåmülo jagadvr¢kßa utthitaΔ Ù so ’vidyåkßaye bhümav eva lœyata iti dar†itam Ù tatra sakr¢davidyåkßaye sa∫sårabhåna∫ na syåt Ù d®ßyate ca kßœñåvidyånåm api mårkañ∂eyådœnå∫ sa∫sårabhånam asmadådivad iti jœvanmuktau sandihåno råjå mårkañdeyam uvåcety åha Ù). In the following, all translations of Sanskrit and German are my own unless otherwise indicated. The following synopsis is only an approximation of Nœlakañ™ha’s argument, which presupposes a density of advaitin textuality which there is insufficient room to explicate further here. 6 vs. 1 (manoΔ manute iti abhimånåtmako ’haμkåro manuΔ Ù); vs. 2 (parabrahmaña eva rüpånantara∫ matsyåkhyo jœvaΔ). See Muir 1858, 1: 201ff. Muir’s publication includes the text and a translation of some of Nœlakañ™ha’s remarks. 7 vs. 2 (caramadehanåvy årü∂ha∫ Ù). 8 vs. 2 (jœvamatsyo ’dar†ana∫ pråpnoti Ù vilœne hy aha∫kåre jœvatva∫ na†yati Ù). 9 vs. 2 (naß™e py avidyåkhye kårañe sa∫sårabhånalakßaña∫ kårya∫ cakrabhramivat ka∫cit kålam anuvartata ity adhyåyatåtparya∫ Ù).

III. How to Produce, Construct and Legitimate a Tradition

227

Nœlakañ™ha then sustains this interpretation in the comments he provides for the verses. The seven sages are the life-breaths and the sense-powers.10 The seeds Manu stores on board are the unseen past karmas.11 The waters are the interpenetrating network of births and deaths (sa∫såra).12 Bopp did not like Nœlakañ™ha’s allegorical interpretation.13 Making the assumption that Nœlakañ™ha’s views were representative of commentators generally, he deplored the lack of historical method he found in the “Scholiasts, who uncritically interpret everything in the biases of their sect and time, and who treat language and myths in an arbitrary fashion”.14 The British Indologist, John Muir, writing in 1858, seems to have been even more appalled. “It is scarcely necessary to remark”, he wrote, “that the narrator of the legend himself appears to have had no such idea of making it the vehicle of any Vedantic allegory such as is here propounded” (Muir 1858, 1: 203). Nœlakañ™ha, Innovation, and Indological Reception Nœlakañ™ha’s commentary first became known to Indologists from their readings of episodes of the Mahåbhårata such as the flood story. A judgment of Nœlakañ™ha’s value as a commentator was formed that was not very complimentary. Yet Indologists continued to use the commentary, and it was not as though they had no choice. There are at least half a dozen other extant commentaries on the great Sanskrit epic, but to this day, Nœlakañ™ha’s has remained the only commentary in regular use. Indologists have continued to depend on the commentary, and have continued to run up against interpretations that they do not like. What explains this peculiar situation? Nœlakañ™ha’s commentary sits in what one might call an Indological blind spot: present in the field of vision, never the object in focus. As the citations above indivs. 31 (saptarßibhiΔ pråñendriyaiΔ Ù). vs. 32 (bœjåni karmåñi Ù). 12 vs. 34 (åpaΔ sa∫såraΔ Ù). 13 “According to the allegorical explanation that Nœlakañ™ha gives in the introduction to this episode, the fish should be Life, Manu the ego, the pot and the other large receptacles that many different bodies, the seeds the past actions, etc.” (Bopp 1829b: xxi [“Nach der allegorischen Erklärung welche Nœlakañ™has in der Einleitung zu dieser Episode gibt, wäre die Fisch das Leben, Manus die Ichheit (ahankara), das Gefäss und die anderen grossen Wasserbehälter eben so viel verschiedene Körper, die Samen die frheren Handlungen u.s.w.”]). 14 Bopp recognized that a reading of the MBh text that sought to conjoin its meaning with that of later ‘Vißñuite’ versions of the story would disallow the sort of chronologizing that he was attempting. “Wäre dies der Fall, so wrden wir in den alt-indischen Gott mehr von dem anderen unterscheiden können, wir wären immer blos den gewaltsamen Deutungen später Scholiasten hingegeben, die ohne Kritik alles in der Befangenheit ihrer Sekte und ihrer Zeit erklären, und Sprache und Mythen gleich willkurlich behandeln” (Bopp 1829b: xx). 10 11

228

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

cate, some early Indologists found Nœlakañ™ha’s general approach to commentary improper, especially in its lack of historical sense. What was left unexamined, however, were the historical implications of this criticism of Nœlakañ™ha’s faulty historicism. What exactly was the nature of Nœlakañ™ha’s ‘sect and time’? In what ways might they have given rise to a particular vision of the meaning of the epic? Could this vision have been something new in its day, even unusual? Could that have been the reason the commentary succeeded well enough to be disseminated and copied, its circulation bringing it into the hands of the Indologists? Here we encounter two distinct but related problems, linked to our understanding of what it means to be traditional in the Sanskrit literary tradition. The problems can be divided temporally: first, the response to and uses of Nœlakañ™ha’s commentary by modern Indologists in the nineteenth and twentieth centuries; and second, the commentary’s success in the seventeenth and eighteenth centuries. In the first part of this paper I will argue that Nœlakañ™ha’s commentary has enjoyed its peculiar status among modern Indological readers in part because of its misdiagnosis as a typically ‘traditional’ commentarial work, and I will offer an explanation of how that misdiagnosis came to occur. In the second part I will argue that Nœlakañ™ha’s commentary was already successful earlier, even in his own lifetime in the late seventeenth century. Among the reasons for this success was the fact that the commentary was anything but ordinary or conservative; it was, rather, an innovation, and valued as such. This will return me, at the close of the paper, to our Indological blind spot, and how the study of the early modern period might serve as a corrective. 1. Nœlakañ™ha and Modern Scholarship To this day Nœlakañ™ha maintains a paradoxical position in Indological studies. He is the only Sanskrit commentator on the complete Mahåbhårata who is regularly made use of by the epic’s academic readers. At the same time those readers complain. We begin our consideration of the problem of Nœlakañ™ha’s commentary by surveying the history of its modern reception in both modes. 1.1. Uses and Abuses Nœlakañ™ha’s commentary is called the Bhåratabhåvadœpa, (‘Illuminating the Inner Meaning of the Mahåbhårata’, henceforth referred to simply as the Dœpa). It appeared in conjunction with published editions of the Mahåbhårata from early on. The first full publication of the epic text, the ‘Calcutta’ edition of the Asiatic Society in 1834-39, included no commentaries, but generally followed

III. How to Produce, Construct and Legitimate a Tradition

229

Nœlakañ™ha’s text (Sukthankar 1933, liii and passim; Brockington 1998: 58). The second full publication, that of Gañapata Kr¢ßñåji in Bombay in 1862-63, presented Nœlakañ™ha’s commentary along with the epic. This edition was enlarged and reprinted in 1885-88 (Khå∂ilkar 1862-1863). Even the publication of the ‘Southern Recension’ contained in its notes excerpts from the Dœpa, and from no other commentary (Krishnacharya, Vyasacharya 1906-1914). The Chitrashala Press edition of 1929-1936, published in Poona, also included the commentary of Nœlakañ™ha, as the editor said, because previous editions had gone out of print and the text was still in demand.15 The text from these editions of Nœlakañ™ha’s commentary is kept in print by booksellers in Delhi today. Other partial editions have also appeared (Emeneau 1935: 60-72; von Stietencron 1992: 742ff.). Although there are numerous other commentaries on the Mahåbhårata that are extant in manuscripts, none of these works has been fully published.16 Two parvans of the commentary of Arjunami†ra, a commentator of the late fifteenth century, were published 80-85 years ago, in a run so small as to be effectively unavailable today (Bakre 1915; 1920).17 The Indological use of Nœlakañ™ha’s Dœpa has taken several forms. The Dœpa is used as a guide to difficult-to-read words and sentences. It serves as a trove of realia, of variant readings, of references to other commentators, of data about vernacular languages, or of other archival features.18 The Dœpa is the commentary that informs translations of the epic into English.19 And it is taken by modern scholars to stand for the so-called ‘traditional’ viewpoint on the text. In this regard Nœlakañ™ha is sometimes referred to as a ‘medieval’ commentator, and his commentary is taken to transmit understandings of the 15 Ki∫java∂ekara 1929-1936, 1: i (sa∫prati tu tasyåpy atidurlabhatvåt punaΔ sa∫skarañam åva†yakam ity åkalayyådisabhåparvañoΔ prathamakhañ∂am ida∫ prasiddhœkriyate ’småbhiΔ). 16 For a list of MBh commentators see Sukthankar 1930; Sukthankar 1935-36; Holtzmann 1892-95, 3: 67-100; Gode 1944; Raghavan 1941. 17 Four parvans of the commentary of Devabodha have been published: on the Ådiparvan (Dandekar 1941); the Bhœßmaparvan (Belvalkar 1947); the Sabhåparvan (Karmarkar 1949); and the Udyogaparvan (De 1944). There are also citations from several Mahåbhårata commentaries included in the notes of the various volumes of the BORI Critical Edition. 18 Thus Printz 1911. Notwithstanding his comments noted above, Sukthankar accords to Nœlakañ™ha’s commentary an “[…] immense value for the history of the received text” (Sukthankar 1933: lxv); and asserts that the study of the commentators “[…] must be now taken up more seriously, not so much for the sake of the explanations contained in the commentaries –though even the glosses of a commentator like Devabodha are extremely important– as for the readings and på™håntaras recorded in them” (Sukthankar 1935, 185). 19 Even though Ganguli rejects Nœlakañ™ha frequently, still Nœlakañ™ha’s comments give shape to many features of the translation. Ganguli himself acknowledges receiving “much aid” from the Dœpa (Holtzmann 1885-1887: 62; von Stietencron 1992, 1: 751). Van Buitenen also relied on Nœlakañ™ha regularly.

230

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

text that are continuous with ancient understandings, even sometimes continuous with the era of the epic’s composition.20 Nœlakañ™ha’s commentary is thus an unavoidable part of the Mahåbhårata for modern readers. And yet, although Nœlakañ™ha doth bestride the narrow world, one hears complaints.21 Nœlakañ™ha disappoints us in that he is often brief and sketchy. 22 Nœlakañ™ha goes on at length about points that are clear, while on the truly pesky verse he falls silent.23 Worse still, Nœlakañ™ha’s explanations can be harder to understand than the base text he is explaining.24 And Nœlakañ™ha has a penchant for the anachronistic reading (Minkowski 2004a). When Nœlakañ™ha really asserts himself as a commentator, offering more than a minimal gloss or explanation, that is, when readers come up against what Nœlakañ™ha really wants to do with his text, his views are often discounted, and not gently. Even the most grudging of his readers acknowledge that he is occasionally useful in specific passages, but when it comes to his intellectual project, he is not only set aside; he is even deplored. Rejection of Nœlakañ™ha’s interpretations among Indological readers started almost from the beginning of Indological studies. As we have seen, one finds Franz Bopp registering his dismay in 1829, and John Muir following suit several decades later.25 K.T. Telang (1850-1893) characterized Nœlakañ™ha’s comments as “marred by a persistent effort to read his own foregone conclusions into the text he comments on” (Telang 1882: 203).26 V.S. Sukthankar, the found20 A recent example of his identification as a medieval commentator is found in Bhnemann 1983: 17; 69. An older example of his use to reveal continuous understandings is described in Minkowski 2004a: 368-371. 21 In the following section on complaints I will adduce only a few to give the general tenor of the literature. The examples can be multiplied. For the nineteenth century, Holtzmann has an extensive collection of citations of other scholars’ criticisms, which I have used and supplemented (Holtzmann 1892-1895, 3: 74-100). 22 Or else the text as we have it is filled with gaps. Whole sections of the Bhœßma- and Droña- parvans are passed over in silence. The complaint that Nœlakañ™ha is ‘spotty’ (lckenhaft) is found, e.g., in Holtzmann 1892-1895, 3: 74. 23 The complaint that Nœlakañ™ha disappoints at crucial passages is found already in Bopp, who says of Nœlakañ™ha’s comments: “As experts can easily see for themselves, (the commentators) frequently explain what is easy to understand, while ignoring what is difficult; nevertheless one must be grateful to them for often providing useful information” (Bopp 1824: xxvi [“Sie erklären wie die Kenner leicht sich berzeugen werden, oft das Leichte, und lassen das Schwierige unbeachtet, doch verdanket man ihnen auch nicht selten wichtige Aufschlsse”]). Also Berlinzola 1939: 269. 24 See, e.g., Holtzmann 1892-1895: 75, who refers to his “frequently rather incomprehensible style” (“oft recht unverständlich still”). 25 See above, notes 13 and 14. 26 Elsewhere, Telang says of Nœlakañ™ha that he “[…] as usual, deserts his original, giving peculiar meanings to the words without producing any authority” (Telang 1882: 263). Ganguli is troubled by Nœlakañ™ha’s “[…] inveterate predilection for a needless subordination of the expressions of the poet to purely brahmanical ideas” (in Holtzmann 1885-1887: 62).

III. How to Produce, Construct and Legitimate a Tradition

231

ing editor of the Mahåbhårata’s critical edition project in Poona, provided a list of passages in which “Nœlakañ™ha has misunderstood the text, and given doubtful, far-fetched or fanciful interpretations”. His explanation of another stanza “is childish, to say the least” (Sukthankar 1933: lxviii).27 Among Indologists, the German scholar Adolf Holtzmann (18101870) is the only one to have made a careful evaluation of Nœlakañ™ha’s commentary as literary commentary (Holtzmann 1892-1895, 3: 74-100).28 In concluding his findings, Holtzmann decided that Nœlakañ™ha was utterly lacking in historical critical sense, and that his fondness for allegorical interpretation and his incomprehensible style were annoying, but that his commentary was of the greatest importance for understanding particular passages.29 This has become the customary judgment of Nœlakañ™ha, and one will encounter it today in most university classrooms where the Mahåbhårata is read in Sanskrit with a commentary. 1.2. Why, then, Nœlakañ™ha’s Modern Success? Regardless of the complaints, Nœlakañ™ha remains, nevertheless, the only Mahåbhårata commentator regularly used by modern scholars. Why should this be so? In the first instance, of course, it is because the Dœpa is the only commentary on the complete Mahåbhårata that is in print. The real question, then, is why that came to be so. There are two sorts of reasons to consider: those to do with the text of the epic that Nœlakañ™ha created, and those to do with the commentary in its own right. At the time that he produced his commentary, Nœlakañ™ha created a textual edition of the Mahåbhårata. In the opening verses of his introduction he describes himself as assembling manuscripts from different regions, comparing them, and settling on the best readings.30 Variant readings are then sometimes discussed in the 27 Nœlakañ™ha’s explanations are not cited. Berlinzola 1939, passim, faulted Nœlakañ™ha for many weaknesses, among them a fondness for the complicated and a tendency to the fantastic and arbitrary. 28 Holtzmann’s four volume work is most remembered for, and therefore forgotten because of, the ‘inversion theory’ that it advances. Nevertheless the work contains many valuable contributions, including the study of commentators, of which the study of Nœlakañ™ha forms a part. 29 “Historisch kritischer Sinn geht ihm ganz ab; störend ist seine Vorliebe fr allegorische Auslegung und der oft recht unverständlich Stil. Fr die Erklarung der Einzelheiten ist er von grosser Bedeutung. […] Sein Commentar ist ein sehr schätzbares, ja unentbehrliches Hilfsmittel zum richtigen Verständniss des Epos, muss aber mit Vorsicht gebraucht werden (99)” (Holtzmann 1892-1895, 3: 74). More comments in his reviews of Roy and Ganguli in Holtzmann 1884-1885; 1885-1887; 1888. 30 His own description of his textual method is found among the introductory verses to the commentary on the Ådiparvan, vs. 6 (bahün samåhr¢tya vibhinnade†yån ko†ån vini†citya ca på™ham agryam Ù pråcå∫ gurüñåm anusr¢tya våca∫ årabhyate bhåratabhåvadœpaΔ Ù).

232

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

commentary on particular verses. The text that he established is known to modern scholars as the ‘vulgate’ text of the Mahåbhårata (Sukthankar 1933: liii and passim; Brockington 1998: 58). Dissatisfaction with Nœlakañ™ha’s work extended to his constituted text of the epic, although there we find opinion more divided. V.S. Sukthankar complained that Nœlakañ™ha’s text was only a “smooth and eclectic but inferior text, of an inclusive rather than exclusive type, with an inconsiderable amount of Southern element” (Sukthankar 1933: lxvi). Nœlakañ™ha’s method was naturally repudiated by Sukthankar, who was dismayed to find that “Nœlakañ™ha’s guiding principle, on his own admission, was to make the Mahåbhårata a thesaurus of all excellences (culled no matter from what source)”. (Sukthankar 1933: lxvii)31 For those critical of Sukthankar’s Critical Edition project, Nœlakañ™ha could become the deposed king in exile. For Madeleine Biardeau, for example, Nœlakañ™ha’s “main concern was not to give a critical edition but rather to bring out as complete a collection as possible of the epic stories that were prevalent at that time and known by everybody in one form or other”. (Biardeau 1968: 121)32 1.3. Artifact of Modernity? Another possible explanation of the modern reception of Nœlakañ™ha’s commentary sees Nœlakañ™ha’s success as an artifact of modernity. On this view, Nœlakañ™ha’s prominence resulted from the introduction of print media into India in the nineteenth century, and of the editorial and publishing choices and processes instituted in the colonial context. Yet the negative opinions of Indologists just mentioned, which began early on, and which have continued throughout, would argue against such an idea; and as we shall see, the prominence of Nœlakañ™ha’s commentary was well-established before the onset of the colonial transformations of South Asia. It is worth noting, however, that the principal editions of Nœlakañ™ha’s commentary were produced by ‘native’ publishing houses, 31 Italics are Sukthankar’s. He then cites Nœlakañ™ha’s ‘naïve’ remarks at the beginning of the Sanatsujåtœya: “This commentary on the Sanatsujåtœya in the Udyogaparvan proceeds by bringing together the explanations of (past) commentators with the variant readings of verses found in modern manuscripts, following the principle of making a compendium of the valuable (guñopasa∫håra)” (udyogaparvañi sanatsujåtœye bhåßyakårådibhir vyåkhyåtån sa∫pratitanapustakeßu ca sthitån på™hån †lokå∫† ca guñopasa∫håranyåyenaikœkr¢tya vyåkhyåyate Ù). 32 Biardeau longed for a return to the study of Nœlakañ™ha’s text or at least to Nœlakañ™ha’s scholarly method as a way of arriving at an ‘authoritative’ text. Sylvain Levi, whom Biardeau cites in her essay, had suggested in his review of the first fascicules of the BORI Critical Edition, that Sukthankar abandon his project of reconstructing the ‘UrMahåbhårata’, and accept the “[…] Vulgate –autrement dit l’édition de Nœlakañ™ha, par exemple– comme point de départ, et de nous livrer au plus tôt le dépouillement, comme il est fourni par ses notes, des manuscrits décrits et classés selon l’excellente méthode qu’il a adoptee” (Levi 1929: 347).

III. How to Produce, Construct and Legitimate a Tradition

233

not by government publishers. These were publishers based in Bombay and Poona, and so a regional preference for a Mahåråß™rian pundit’s work might be implied.33 Hence one could accept a weak version of the ‘artifact of modernity’ claim, viz. that the commentary, already widespread, attained a special preeminence and authority throughout the subcontinent because of the location of the new media center in Bombay, in a Marå™hœ cultural milieu. 1.4. Why, then, the Complaints? Most relevant to our concern, however, is the content of the commentary itself. The nature of the commentary’s acceptance by Indologists has been shown in the passages cited above. The Dœpa was taken as representative of the ‘traditional’ reading of the text. In the following section, I will show why this was a misapprehension of Nœlakañ™ha’s place in the Mahåbhårata commentarial literature. Modern complaints about Nœlakañ™ha’s commentary found their source here: expectations arose that were ill-adapted to what Nœlakañ™ha’s work was for. Those expectations were inevitably disappointed. How is it that Indologists took Nœlakañ™ha’s work to be ‘traditional’? There was, first of all, the murky conception among Indologists of what being ‘traditional’ amounted to. Moreover, they would have encountered Nœlakañ™ha’s text and commentary in widespread use among ‘native’ readers. This fact, coupled with the predisposition to see Sanskritic literature as part of a timeless and changeless universe of discourse, would have given rise to a judgment of Nœlakañ™ha’s work as typical of the ‘traditional’ understanding of the text. We cannot entirely fault the early Indologists for their lack of knowledge of the vast, complicated literature of the Mahåbhårata; its temporal dynamism and its complex inter-regional and inter-textual dynamics are still far from entirely charted or understood. Nor is the predisposition to see changelessness and the absence of historical contingency surprising, given that these are claims about the literature that are built into the presentation and reception of Sanskrit texts, especially Vedåntic ones. 2. The Dœpa in the Seventeenth and Eighteenth Centuries Both aspects of the reception of Nœlakañ™ha’s epic text and commentary by Indologists depended on the wide dissemination of 33 Certainly this is the reason that the publisher of the Citrashala Press edition gives for publishing Nœlakañ™ha: “Here the commentary of Nœlakañ™ha, called the Bhåratabhåvadœpa, is included. Among the three or four commentaries (available), this one is especially relied on in Mahåråß™ra by the learned” (Ki∫java∂ekar 1929-1936, 1: i [atra ca nœlakañ™Δakr¢tå bhåratabhåvadœpåkhyå ™œkå sa∫nive†itå Ù tricaturåsu ™œkåsv iyam eva vi†eßato mahårå™reßv ådriyate sudhœbhiΔ]).

234

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

both. In this section I will show that the commentary’s success predated the modern era, began in fact even in Nœlakañ™ha’s own day. I will also argue that Nœlakañ™ha’s commentary was valued more for its departures from the comments of his predecessors than for its adherence to them. In order to make this argument, I will begin with Nœlakañ™ha and his works, and the literary context in which his Dœpa was produced. 2.1. Nœlakañ™ha Caturdhara, His Works and His Moment Nœlakañ™ha Chaudhuri, or Caturdhara as he made it in Sanskrit, was a Marå™hœ-speaking bråhmaña who flourished in the second half of the seventeenth century.34 Early in life he moved from a small town on the Godåvarœ in what is now Ahmadnagar district to Banaras, where he participated in the efflorescence in Sanskrit literature that was taking place there in that era. He wrote about fifteen Sanskrit works that we know of today, some of them commentaries, and some of them independent works. Among them were four works that belonged to a genre of his own invention, a genre that he named mantrarahasyaprakå†a, ‘The Illumination of Secrets of the (Vedic) mantras’. In these very unusual compositions, Nœlakañ™ha assembled verses selected from the R¢gveda and commented on them in such a way that, regardless of their meaning in their R¢gvedic context, they were found to disclose the narrative of the Råmåyaña in one case, of the Bhågavatapuråña in another, of the K农khañ∂a in a third, and of the Brahmasütra in a fourth.35 By far the most successful of Nœlakañ™ha’s compositions, and the work for which he is remembered, however, is his commentary on the Mahåbhårata. 2.2. Epic Commentaries Before Nœlakañ™ha How did Nœlakañ™ha’s commentary accord with the expectations of the genre of epic commentary in Sanskrit, assuming that there was such a genre? Here one immediately confronts the problem of an almost complete lack of study of the phenomenon, and a sporadic coverage even of the individual commentaries,36 but I would argue 34 For fuller bibliography about Nœlakañ™ha see Printz 1911; Gode 1942; Gode 1956; Minkowski forthcoming a, forthcoming b. 35 These are the Mantraråmåyaña, Mantrabhågavata, Mantrak农khañ∂a, and Mantra†årœraka, respectively. On the genre see Minkowski 2002; forthcoming a; forthcoming b. 36 Goldman 1984, 114-117. Also Goldman 1995, 12-13; Goldman 1992. Goldman has promised a forthcoming work on the Råmåyaña scholiasts (Goldman 1984: 115). The following is then provisional in anticipation of Goldman’s findings. Raghavan appears to have planned a compendious study of the phenomenon of Råmåyaña commentary, but the work did not come out (Raghavan 1942: 289). Given the length of the two epics, and the volume of commentarial literature, this is clearly not a project that can be completed by a single person, however industrious. Raghavan did of course produce a series of invaluable smaller studies. See Mudaliar 1968; Raghavan 1941; 1942.

III. How to Produce, Construct and Legitimate a Tradition

235

that there were both generic conventions for epic commentary and a literary context, against the background of which one can most usefully understand Nœlakañ™ha’s contribution. In the period when Nœlakañ™ha was writing, it was especially the Råmåyaña that had attracted intense commentarial attention, primarily from vaißñava religious intellectuals, for whom the text had become a central topic of exegesis, and a vehicle for the realization and articulation of theological understanding. The cumulative effect of these commentaries was a dense, layered, detailed, almost continuous reading of the text of the Vålmœki Råmåyaña, a reading that was still being expanded in Nœlakañ™ha’s day and after, as for example in the extensive and influential commentary by Någoji Bha™™a written in Banaras in the late decades of the seventeenth century, and in the Dharmåkü™aμ of Tryambaka Makhin of Marå™ha Tanjore in the eighteenth century (Pollock 1984). There appear to have been fewer attempts at commenting on the Mahåbhårata as a whole, though philosophers and theologians had commented on episodes of the epic that had become central to their intellectual projects, especially philosophical passages such as the Bhagavad-Gœtå and the Sanatsujåtœya, and the appended text, the Hariva∫†a. Two of the commentators on the total Mahåbhårata, Sarvaj∞a Nåråyaña (thirteenth century) and Vimalabodha (thirteenth century), were also commentators on the Råmåyaña (Bhatt 1965: 353). In the case of these two authors, then, there is a symmetry of treatment, but elsewhere, and perhaps more typically, an asymmetry is found. Although there is a full commentary on the Råmåyaña attributed to the vaißñava theologian Råmånuja (early twelfth century), he is credited on the other hand with only a sa∫kßepatœkå, that is, a series of brief notes on the Mahåbhårata (Aufrecht 1891-1903, 2: 101a). There are independent essays on the Mahåbhårata providing an interpretative reading from the vantage point of a particular sectarian analytic, as for example the Måhåbhåratatåtparyanirñaya (‘Determination of the Meaning of the Mahåbhårata’) by the thirteenth century theologian of dualism, Madhva, and the Bhåratatåtparyasa∫graha (‘Compendium of the Bhårata’s Meanings’) by the sixteenth century philosopher of non-dualism, Appayya Dœkßita.37 Thus one can argue that there was a preference for writing a total commentary on the Råmåyaña by comparison with a ‘summary-ofimport’ of the Mahåbhårata. This was probably a result not just of the religious importance of the Råmåyaña, but also a result of the daunting size of the Bhårata text. 37 For a publication history of Madhva’s work see Misra 1992a: iv. The Bhåratatåtparyasa∫graha was published by Vani Vilas Press in 1929. Also published at the Vani Vilas press was Appayya Dœkßita’s Råmåyañatåtparyasa∫graha. I was unable to examine copies of either in the preparation of this work. Aufrecht also attributes to Appayya Dœkßita a Råmåyañabhåratatåtparyasa∫graha (Aufrecht 1891-1903, 1: 22b). See also Bhatt 1982.

236

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

The commentaries on the total Mahåbhårata probably emerged from the vyåsagha™™a literature, which are compendia of thorny parts of the text. Early commentaries were largely written as annotations of these particularly difficult words and passages.38 Indeed, the names of many of the Mahåbhårata commentaries show this preoccupation, for example Vimalabodha’s Durbodhapadabha∞janœ, ‘the analysis of words that are hard to understand’ (Aufrecht 1891-1903, 1: 439b).39 One does see titles of this sort among the Råmåyaña commentaries,40 but on the other hand there is nothing in the Mahåbhårata commentarial literature that corresponds to the Råmåyaña commentaries of (pseudo-) Råmånuja, Govindaråja, and others, which, proceding verse by verse, attempt to incorporate the Råmåyaña into a sectarian, theological vision; at least not until Nœlakañ™ha. Before him, comments tend to be topical, local, and not oriented to articulating a vision of an overarching literary structure or the place within it of the passage in question. It is pertinent to note here that, as a phenomenon, epic commentary in Sanskrit appeared relatively late in the life of the epics, with extant commentaries datable not much earlier than the twelfth century.41 In fact, the great proportion of Sanskrit commentaries were written in the early modern period; important commentaries on the Råmåyaña appeared even in the eighteenth century (see Sastri 1942; Pollock 1984). Since there is no study of the general phenomenon, there is also no theory of the circumstances of the arising of this commentarial literature, or of its efflorescence in the early modern era. And yet the phenomenon requires some explanation. After all, the epic text’s surface meaning is usually clear enough, and does not present difficulties of the sort that would occasion a rigorous word-by-word treatment, of the type written on the sütra literature from early on, or on the Vedic literature from late in the first millennium A.D. To speak in rather broad terms about Sanskrit literary history, the date of the emergence of epic commentary would suggest that its appearance should be associated with the heyday of ‘medieval’ regional kingdoms, with the phenomenon of the arising of the ‘vernacular’ literatures, and of the feeding back into Sanskrit of the literatures of the sectarian religious movements. The vast literature of vernacular and Sanskrit retellings of the two epics and the re-exami38

On the annotative style of the earliest extant MBh commentaries see Dandekar

1941: i. 39 Vimalabodha’s commentary is also called the Kü™a†loka™œkå. Similarly the anonymous Bhåratapadaprakå†a and Kü™avyåkhyå (Aufrecht 1891-1903, 2: 100b), and the Vißama†loka™œkå of Råmaki∫kara Nyåyålaõkåra (Aufrecht 1891-1903, 2: 101a). 40 For example the Kü™atœkå of Råmacandra Tœrtha and the Vißamapadavivr¢ti of Harœta Veõkate†vara (Bhatt 1965: 351-352). 41 There is some evidence that commentaries did not exist before then. At least no subsequent commentator refers to them. Raghavan 1980: 314.

III. How to Produce, Construct and Legitimate a Tradition

237

nations and re-formulations of epic themes would seem to have given rise to an attempt to insert the Vålmœki Råmåyaña into the new cultural formation via commentary.42 That there was some sort of shift of treatment appears rather clearly if one considers the attempts of earlier Sanskrit authors at a shared, cosmopolitan, and rather secularized reading of the epic, such as one encounters in the earlier appropriations of both narratives into Sanskrit dramatic and poetic works, and in the treatments of their aesthetic sentiment by theorists such as Ånandavardhana in the ninth century (Tubb 1991). 2.3. Nœlakañ™ha Against this Background What profile does Nœlakañ™ha’s commentary strike against this background of epic commentarial genre and its (very approximate) history? While Nœlakañ™ha does have predecessors who have commented on the Mahåbhårata, and while there are a few precedents for advancing a sectarian reading of the epic in a summarized form, there is little precedent for the sort of overarching goals he has set for his commentary, and little precedent for his methods. Nœlakañ™ha offered a reading of the epic based in an advaitin viewpoint, frequently depending on allegorical readings. He did this in a way that placed a high value on novelty and ingenuity. We have already seen the example of Nœlakañ™ha’s allegorizing the Mahåbhårata’s flood episode in non-dualist terms. The passage shows that Nœlakañ™ha is capable of rendering an advaitin reading of passages that do not seem at first glance amenable to such readings. As I shall show, Nœlakañ™ha was aware that he was doing something unprecedented, both in his content and in his approach. It should be noted, however, that this is not the only sort of work that Nœlakañ™ha did in the Dœpa. There is the textual edition he created, first of all, and the discussion of variant readings. Furthermore, at the close range of individual verses, Nœlakañ™ha’s commentary often operates as other commentaries do. One finds simple glosses and simple explications of individual terms and sentences. At close range, furthermore, these comments do not appear to pursue a single-minded exegetical agenda. In providing such readings, Nœlakañ™ha relied on the work of his predecessor commentators.43 Sometimes Nœlakañ™ha will refer to the earlier commentators whom he is using; often he does not.44 In this sense, Nœlakañ™ha represents a commentarial approach that is continuous with what had been done before. 42 Pollock includes a historicizing of Råmåyañas and their impacts in approximately this period (Pollock 1993). 43 Even if more often, as we have seen, he does not. Holtzmann referring to Telang, who shows that Nœlakañ™ha follows Arjunami†ra, and in at least one place in the Dœpa on the Gœtå reproduces the commentary of ‡aõkaråcårya (Holtzmann 1892-1895: 81; Telang 1882: 181). 44 See Sukthankar 1933: lxv-lxvi; Holtzmann 1892-1895: 80-81. Even in the statements of his general program, Nœlakañ™ha acknowledges a debt to his predecessors. On

238

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Yet when one looks at Nœlakañ™ha in a larger frame, the nondualist features of the commentary become clearer. As we have seen, in the flood story Nœlakañ™ha articulates some governing principles for the interpretation of the whole episode, then implements these principles in the interpretation of selected terms and phrases. This method is used in many places in the epic. It is the chief way that Nœlakañ™ha carries out his advaitin agenda, while blending it with theistic themes. For the period in which he lived, this was something new. There is little evidence for a specifically advaitin commentary on either epic before the Dœpa. Certainly there had been ‘Vedånticizing’ of the Råmåyaña in retellings, most notably in the Adhyåtma Råmåyaña of the early sixteenth century, the Sanskrit retelling of the epic which was based on a fairly doctrinaire version of non-dualist vedånta, and which then fed into subsequent versions of the Råmåyaña, especially the hugely popular sixteenth century Hindœ retelling of Tulsœdås (Raghavan 1988: 57-71). As for the Mahåbhårata, Appayya Dœkßita had written the Bhåratatåtparyasa∫graha from a theistic non-dualist viewpoint, but the latter text is not intended to be a commentary .45 In short, I can find no forerunner in the genre Nœlakañ™ha has chosen, with all the imaginative effort necessary for articulating and sustaining an advaitin reading through all the particularities of the narrative. 2.4. Innovation and Ingenuity by Intention Not only did Nœlakañ™ha do something unprecedented; he was aware that did so, and intended to do so. This can be shown by Nœlakañ™ha’s own statements of purpose for his Dœpa, articulated for the first time in the seventh verse of the introduction to his commentary on the Ådiparvan:46 Let other commentaries, which resemble the sun and moon, cause the gems of external meanings to shine. (This) lamp (of a commentary of mine) is capable of lighting up the hoard of internal, hidden meanings in the palace that is this Bhårata epic. ™œkåntaråñœnduraviprabhåni båhyårtharatnåni cakåsayantu Ù antarnigü∂hårthacayaprakå†e dœpaΔ kßamo bhåratamandire ’smin ÙÙ Arjunami†ra reusing Devabodha in a wholesale way, Sukthankar 1935: 198-200. Free borrowing of the glosses of predecessors is a common practice for Sanskrit commentators. 45 See above note 37. 46 Among Indologists, only Holtzmann has taken note of what Nœlakañ™ha declared as his intention in producing the Dœpa. “Wie aus dem siebenten seiner Eingangsverse zu ersehen ist, thut sich Nœlakañ™ha gerade auf diese Ergrndung des inneren ‘geheimnen Sinnes’ des Textes, anderen Commentaren gegenber, die sich mehr an die Aeusserlichkeiten halten, etwas zu Gute” (Holtzmann 1892-1895, 3: 100). Mañ∂ana Mi†ra’s preface to the 1988 reprint of the Dœpa also identifies the seventh verse as articulating what is distinctive: vastuto bhåratabhåvadœpåkhye ’tra vyåkhyåne padasvarüpaparicayakåriñœ paddhatir naivåsti svœkr¢tå, bhåvaprakå†anasyaiva prådhånyam parilakßyate (1988: 16).

III. How to Produce, Construct and Legitimate a Tradition

239

The image sustained in the verse is of the Bhårata as a mansion. On its outside are individual gems of meaning that the sun and moon can illuminate, just as other commentaries can illuminate obvious points. Nœlakañ™ha’s commentary, however, shines on the far more valuable hoard of gems lying within, where the light of sun and moon cannot reach, and makes them shine. This is a more difficult, yet more valuable thing to accomplish. Due honor is paid to other commentators, who are suns and moons of commentary, but Nœlakañ™ha is proposing to do something different –to reveal the hidden inner sense of the text as other commentators have not previously done.47 Other verses of the preamble also suggest this purpose. Many of the verses of invocation can be read on at least two levels, and juxtapose the great characters and sages of the epics with Nœlakañ™ha’s own teachers and with the founders of the dar†anas. In the ninth verse, a fully double-meaning verse, the most difficult of all to read, Nœlakañ™ha tells us that he relied wholly on his guru while writing the commentary at the same time that he relied wholly on Råma while living through Råma’s story.48 The intention of this multivalent verse with its allusion to the Råma story is, perhaps among other things, to point out the difficulties of writing a commentary on the Mahåbhårata by comparison to the Råmåyaña, for the reasons discussed above. There is a great deal more embedded in these opening verses, but in short, Nœlakañ™ha’s plan to make use of ingenious and artful reading in order to understand the epic as a text of systematic philosophy, with advaita as its culmination, is already announced. Nœlakañ™ha then sets out his intellectual plan more explicitly in the commentary on the first verse of the epic, which serves as a systematic statement of the purpose and rationale of the work as a whole. There 47 Here it is relevant to return to Nœlakañ™ha’s commentary on the Flood described above, where, at the end of an allegorizing comment in vs. 33 (of the vulgate text, =31 CE) Nœlakañ™ha closes with the statement: kathåpakße sugamam eva, or, roughly, that the interpretation of the preceding from the point of view of its meaning as a story is simple and needs no explanation. This term pakßa- is an element of the technical interpretative apparatus of commentaries on the †leßakåvya or multi-leveled poetry exemplified in note 48. The implication is that Nœlakañ™ha does not seek to replace the surface meaning of the text with his ‘Vedånticized’ interpretation, but only to add, or reveal, another level. 48 “I could not rely on the array of lexicons and grammatical analyses in explicating the long verses; nor could I rely on abridgements for the deep verses, nor could I explicate knotty verses or subtle ones; nor were broken verses, nor the obscure ones, nor the rare ones analyzed; nor were the displeasing ones, nor the dispiriting ones, nor the frightening ones explained by me, except in that I relied on my guru Lakßmañårya” (uttåneßv iha koßavigrahabalam padyeßu naivå†rita∫, ga∫bhœreßu na setavo na vihitåΔ ku™å na naspho™itåΔ Ù na chinnå na tama†carånanatatir bhaktå, na nåhlåditå no dœnå na vibhœßañå† ca vihitåΔ †rœlakßmañåryå†ritaiΔ Ù). The same lines, divided and read differently, tell us that in the drawn out marches he relied on no treasury or military force; that he built no bridge when inspired to do so by the sharp-faced one (i.e. Sampåti), that the row of heads of the demon (Råvaña) was not cut off, nor were the faces of devotees gladdened, nor was Vibhœßaña exalted and put in place, except by his relying on Råma. Thanks for Yigal Bronner for help with this verse, whose meaning is only approximated here. The double and perhaps triple negatives are difficult to render. No doubt it was the guru’s name, as Robert Goldman pointed out to me, that prompted this literary conceit.

240

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

are two features of note for our purposes here. The ‘Vedånticizing’ project is directly stated, and a peculiar and characteristic ‘Vedicizing’ is resorted to in order to argue out the heart of the rationale. First of all, Nœlakañ™ha makes his ‘Vedånticizing’ intentions clear through an articulation of the epic’s anubandhacatuß™aya or ‘topical tetrad’. That is, he provides a précis of the topic and purpose of the text, of the relevance of the topic to the purpose, and of the text’s intended audience.49 In brief the Mahåbhårata is about the (ultimate) unreality of the world, the truly existent nature of brahman, and the soul’s non-difference with It. The purpose of the work is to bring about an absolute cessation of misfortune through the cessation of ignorance and thereby the blocking, in past, present, and future, of the ‘created’ universe. The intended audience is anyone who wishes to gain this cessation, and the relevance of the text is that it makes known what needs to be known.50 It is hard to imagine a more programmatically Vedåntist literary agenda. Here Nœlakañ™ha re-formulates the understanding of the Mahåbhårata’s subject matter, its canonical status in Sanskrit literature, and what methods are appropriate for its understanding. As programmatic a Vedåntist as he might appear from his anubandhacatuß™aya, Nœlakañ™ha is nevertheless traveling far from the standard path when it comes to his mode of demonstration, and this too is announced in the opening comment, in the deployment of a characteristic ‘Vedicizing’. In the longest section of the commentary on the epic’s first verse, Nœlakañ™ha anticipates an objection: why study the Mahåbhårata at all, and why comment on it, since it is only man-made (paurußeya) and men are capable of error and deceit?51 Nœlakañ™ha’s reply incorporates the statements one would expect, about the quasi-Vedic status of works by omniscient authors such as Manu and Vyåsa, and about how the smr¢tis serve as ancillary literature to complete the Vedas’ meaning.52 The whole discussion is directly dependent on the arguments on this old topic marshalled by the Mœmå∫så philosophical-exegetical school.53 49 Such a statement was standard for later Vedåntic works. On anubandha-catuß™aya see, e.g., Jhalkikar 1978: 28, s.v. anubandha. Also note the Vedåntasåra of Sadånanda Yogœndra (fifteenth century), khañ∂as 3, 4. The four terms are: vißaya, prayojana, sa∫bandha, and adhikårin. 50 eva∫ ca jœvåvidyåkalpitatvåj jagato mithyåtva∫ brahmaña† ca tatra sattåsphürtipradatvena jœvasya tadabhinnatva∫ ceti vißayo dar†itaΔ Ù avidyånivr¢ttau tatkr¢tasya prapa∞casya traikålikabådhåd åtyantiky anarthanivr¢ttiΔ prayojanam | arthåt tatkåmo ’dhikårœ Ù granthasyoktavißayasya ca j∞åpyaj∞åpakabhåvaΔ sa∫bandha iti ca dar†itam Ù (see Sharma 2001: 95-96). 51 nanv etan na pa™hanœya∫ na vyåkhyeya∫ ca paurußeyatvåt purußeßu ca bhramavipralambhakatvayoΔ sa∫bhavåd iti cet Ù. 52 na sarvaj∞eßu vedaikapramåñeßu manvådißu tadasa∫bhavåt Ù pürvatantre ca smr¢tyådidharme pråmåñyasya siddhatvåt |. 53 In the smr¢tyadhikaraña of the Mœmåmsåsütra (1.3.1-2). With one notable exception, Nœlakañ™ha follows the standard examples found in that adhikaraña’s discussion of how smr¢ti is based on †ruti. Such examples as aß™akåh kartavyåΔ, prapåΔ kartavyåΔ and others are found already in ‡abara’s commentary.

III. How to Produce, Construct and Legitimate a Tradition

241

However, Nœlakañ™ha makes his argument primarily by introducing a discussion of two R¢gvedic verses. The ninth and tenth verses of R¢gveda, 1.161, a hymn to the R¢bhus, are usually taken to narrate the R¢bhus’ feat of fashioning wooden Soma cups for the gods.54 Nœlakañ™ha instead reads these two verses in conjunction with passages from the smr¢tis prescribing the wearing of the ürdhvapuñ∂ra, that is, the mark worn on the forehead by Hindu sectarians as part of their religious practices (sådhanå).55 Nœlakañ™ha says that the two verses lay down rules for practice under the pretext (vyåja) of offering praise to the R¢bhus.56 We should read and comment on the Mahåbhårata because all works by Manu, Vyåsa and other omniscient sages are Vedic at base, whether we unintelligent people can see that or not.57 We see here Nœlakañ™ha’s willingness to read the Veda to reveal a pluralist vision of religious behavior, extending even to smårta and tåntrika practice.58 As the way of justifying the usefulness of writing a commentary on the Mahåbhårata, therefore, Nœlakañ™ha turns to an unprecedented sort of reading of R¢gvedic verses. This mode of interpreting the R¢gveda is found elsewhere in the Dœpa, and gave rise to another genre of Nœlakañ™ha’s work (Minkowski forthcoming b). 2.5. Reception Well before the advent of the colonial transformation of India, Nœlakañ™ha’s commentary was a great success, as judged by this external measure, that, beginning even in Nœlakañ™ha’s own lifetime, it was widely disseminated and widely used. Intellectuals writing in Banaras in Nœlakañ™ha’s era were working at the center of a well connected network of communication. Nœlakañ™ha benefited from his location at this nexus, as can be deduced from the situation of surviving manuscripts, which outnumber those of other commentators on the Mahåbhårata by far. 54 In the ninth verse each of the three R¢bhus offers an explanation of the origin of the wood, and in the tenth verse each of the three R¢bhus accomplishes a necessary task in making the cups. See the notes on these vss. ad loc in Geldner 1951. Såyaña’s reading is somewhat different, but much more resembles Geldner’s than it does Nœlakañ™ha’s. 55 Thus verse 10, apparently about the R¢bhus’ doings with a lame cow, its flesh, and its dung, is shown by Nœlakañ™ha to be about the sectarian choice between red pigment, yellow pigment, and ash, respectively: evam ihåpi våsudevopanißad-bråhmañatantrapuråñopabr¢∫hanånusåråd ürdhvapuñ∂rårtha∫ †lakßñåμ mr¢dam jalena mi†rayed iti vyåkhyeyam Ù tathå kålågnirudrabr¢hajjåbålådyupanißad-bråhmañatantrapuråñopabr¢∫hañånusåråt tripuñ∂ra∫ kartu∫ nimrucaΔ †akr¢d apåbharad iti puñ∂rårthe tattattantre vidhœyate Ù. The forehead mark is not included as a standard example in the Mœmå∫sakas’ discussions in the smr¢tyadhikaraña. For a compilation of smr¢ti sources that prescribe the tilaka or ürdhvapuñ∂ra, see Kane 1968-1977, 2.1: 672-675. 56 atra r¢bhvåkhyadevatåstutivyåjena †reyaΔsådhanåni vidhœyante Ù. 57 tasmåd asmadådibhir alpaj∞air dr¢ß™amüla∫ adr¢ß™amüla∫ vå manuvyåsådisarvaj∞aprañœta∫ smr¢tijåta∫ vedavad vedavidå∫ vacanam iti nyåyenågamayitavya∫ pa™hanœya∫ ceti sarvam anavadyam Ù. 58 eva∫ satyavådinå∫ tœrthåni yaj∞ådayas tantramårgeña süryådyanyatamopåstiΔ kevalavaidikatå ceti devatåbhåvapråptisådhanånœty mantradvayatåtparya∫ siddha∫ Ù.

242

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

For example, one can consider the information in Aufrecht’s Catalogus Catalogorum, which provides a union list of the Sanskrit manuscripts that had been catalogued anywhere in the world as of 1902 (Aufrecht 1891-1903). The catalogues on which this ‘catalogue of catalogues’ was based were created in the nineteenth century, some earlier, and included reports from British officials who toured provincial regions of India and examined private manuscript collections. In Aufrecht’s listing of the Mahåbhårata commentaries in his first volume, there are three manuscripts of Yaj∞anåråyaña’s commentary, four of Vimalabodha’s, five of Nåråyaña Sarvaj∞a’s, and about a dozen of Arjunami†ra’s. On the the other hand, the same page lists 82 manuscripts of Nœlakañ™ha’s commentary, and this is not a complete list, for Aufrecht puts in an ‘etc.’. The later volumes of Aufrecht show roughly the same ratio. Subsequent information confirms the impression. What evidence we have from collections that date from the period indicates that Nœlakañ™ha’s commentary had circulated widely by the early eighteenth century, especially in the North and West of the subcontinent, and in the Deccan (Gode 1951: 491-498).59 The celebrated pañ∂it Kavœndråcårya, who lived in Banaras in Nœlakañ™ha’s own day, and who had a personal library of about two thousand titles, owned a copy of Nœlakañ™ha’s commentary.60 Several other surviving copies were made in Banaras in those days, including the earliest one for which we have a date. This manuscript, copied by Nœlakañ™ha’s son in 1669 A.D., had become part of the large private collection in Banaras eventually known as the Chandra Shum Shere collection.61 Anüpasi∫ha, the Maharåja of Bikaner who commissioned a work from Nœlakañ™ha, and who famously accumulated a library of about 10,000 manuscripts, owned at least one copy, which can be dated to Sa∫vat 1749, or 1693 A.D.62 The scholar Dœkßita Mañiråma ‡ivånanda, who served Anüpa and played a central role in the creation of the Mahåråja’s library, owned his own copy.63 Indeed, Mañiråma wrote a short work that summarized the essence of Nœlakañ™ha’s commentary on the fourth and fifth books of the epic.64 59 It should be noted, however, that our knowledge for the period is incomplete. Many manuscripts from the period survive; only a fraction have been catalogued. 60 N. 1407 of Ananta Krishna Sastry 1921: 23. On Kavœndra and his list, see Gode 19431944; Raghavan 1940. 61 The collection of about 6000 manuscripts is now in the Indian Institute library in Oxford. N. 117 of Brockington 1999, was copied by Nœlakañ™ha’s son Govinda in Sa∫vat 1725 = 1669 A.D. Nos. 26 and 133 of Brockington 1999, which are also MSS of parts of the Dœpa, were probably also copied by Govinda at the same time. This is also probably the Govinda who was the owner at some time of the Berlin MS (Gode 1942: 148) see note 71. 62 Itihåsa 60 / 110 of Kunhan Raja and Sarma 1944-1948. 63 Itihåsa 52 / 102 of Kunhan Raja and Sarma 1944-1948. On Mañiråma, Pingree 1997: 98ff. 64 Virå™aparvasåroddhåra and Udyogaparvasåroddhåra, Itihåsa 89 / 139 and 90 / 140 of Kunhan Raja and Sarma 1944-1948.

III. How to Produce, Construct and Legitimate a Tradition

243

There were other princes of Råjput states of the period who owned copies of the commentary.65 Fatehsiμha Deva of Jodhpur commissioned a copy that was made in 1678.66 In Udaipur a composite manuscript of the epic that belonged to Sa∫gråma Si∫ha included many books of the epic with the Dœpa, most of them dating to 1714.67 In the collection at Kota there are two complete manuscripts, both of them dated to the eighteenth century.68 In the Khasmohor collection in Jaipur, the personal collection of the princes in the Amber dynasty, there are seven manuscripts of the Dœpa.69 Further south, in the Baroda collection, there is a manuscript of the Dœpa which is dated to 1679.70 Old manuscripts of the Dœpa turn up widely in nineteenth century collections of mostly northern manuscripts. One manuscript in Calcutta is dated 1687; one now in Berlin is dated 1691; one in Poona is dated 1695 (see Gode 1951: 491-498).71 Nœlakañ™ha’s commentary was circulated in the south as well, but one does not find the same sort of overwhelming numerical majority there.72 2.6. What Was the Appeal? I have argued that Nœlakañ™ha’s commentary was innovative by design and something extraordinary in its literary context, and that it was well-received and widely disseminated even within the lifetime 65 I will mention here only Jodhpur, Udaipur, and Kota, as centers where the princely library became the core of a government collection in the Rajasthan Oriental Research Institute, and now catalogued in the series Råjasthåna Puråtana Granthamålå (RPG). 66 N. 210/25567 of Jinavijaya 1963ff. vol. 7: (Jodhpur), dated Sa∫vat 1735 = 1678 A.D., copied for Mahåråj Fatahsi∫hadeva. 67 Nos. 733-755/445(1-21) of Jinavijaya 1963ff. vol. 12 (Udaipur). Parts 5, 15-17 are all dated Vikrama Sa∫vat 1771. Another manuscript of the collection, of the Dœpa on the ‡åntiparvan, N. 769/446 (14) is dated to V.S. 1754 or 1697 A.D. And there are numerous other undated manuscripts. 68 806-29 / 1030 and 830-53 / 1031 of Jinavijaya 1963ff. vol. 25 (Ko™a). The parts of 1030 are all dated to the 1840’s V.S. or 1780’s A.D. 69 Though they are undated: 2802, 6261-6265 of Båhura 1976. 70 Itihåsa 149 / 13813 of Nanbiyar 1950, vol. 2, consisting of the Dœpa on the Droñaparvan and the seven chapters following, dated Sa∫vat 1736. The Baroda collection includes many old manuscripts, but the collection as such is not as old. 71 Gode refers to Manuscripts in the Båõgœya Såhitya Parißat, Calcutta, (‡aka 1609 = 1687 A.D.); the Chambers collection in Berlin (Sa∫vat 1747 = 1691 A.D); and the Bhandarkar Institute, Poona (Sa∫vat 1751 = 1695 A.D.). These latter two collections were created mostly by acquiring manuscripts in the northern and central regions of India. Other early MSS are No. 15429 in Upadhyaya 1953-1965, volume 4, dated Sa∫vat 1765 or ca. 1708 A.D; and Nos. 836/6195 and 845E/6185 of Jinavijaya 1963ff., volume 2.1 (Jodhpur), both listed as belonging to the eighteenth century. 72 Worthy of note are the numerous manuscripts of the Dœpa found in South Indian scripts in South Indian manuscript collections. Some are noticed, for example, in Marulasiddhaiah 1981, Nos. E16945-63 in Kannada, Telugu, and Grantha scripts, unfortunately undated. In Madras, Nos. 1958-1962 of Rangacharya 1907, in Telugu script, and No. 26 of Rangacharya and Kuppuswami Sastri 1913, in Grantha. In Trivandrum, Nos. 379 and 380 in Sambasiva Sastri, 1937, in Malayalam and Grantha scripts, resp.

244

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

of its author. I wish to connect these two points, and argue that the principal reason for the success of the commentary lay in its innovative features. I will not claim that innovation was the only reason for its wide circulation. As we have seen, Nœlakañ™ha compiled a text of the Mahåbhårata from manuscripts drawn from different regions of South Asia. He thereby produced a trans-regional version of the epic text that lent itself to being distributed widely. The commentary that accompanied this version of the text would be disseminated widely as well. Yet this is not, I do not think, the main reason for the success of Nœlakañ™ha’s commentary in his own era, so much as the ingenuity of Nœlakañ™ha’s commentary. How would one go about demonstrating such a claim? Unfortunately we do not possess the sort of literary historical information that is available for, say, France in the same period. That is, there are no literary diaries that we know of, kept by readers and collectors, journals of what they were reading, what they liked and disliked, and why. Nevertheless it is possible to assemble some indirect evidence to support the claim. Some information emerges if we look again at the old collections where the Dœpa turns up, and consider what other works of Nœlakañ™ha were also collected. If we do so, we find that it was especially the mantrarahasyaprakå†a works, the ‘Vedicizing’ works described above, that were popular, especially the Mantraråmåyaña and the Mantrabhågavata. Kavœndra’s collection included both.73 Anüpasi∫ha’s collection had two copies of the Mantrabhågavata, one dating from 1672.74 The Khasmohor collection in Jaipur includes one Mantraråmåyaña and one Mantrabhågavata.75 In Baroda there are two copies of the Mantrabhågavata, one dating from 1678.76 As discussed above, the mantrarahasyaprakå†a works display Nœlakañ™ha’s idiosyncratic literary vision in its most pronounced form. The presence of these works in the same libraries with the Dœpa suggests that Nœlakañ™ha’s work was prized by those readers for its distinctive features, and that the Dœpa was similarly prized for Nœlakañ™ha’s distinctive contribution to reading the Mahåbhårata. As we have seen, Nœlakañ™ha’s commentary appeared at the same time as an upsurge in the number of epic commentaries, and of 73 Nos. 1296 and 1298 of Ananta Krishna Sastry 1921. The list also includes mention of another work by Nœlakañ™ha not otherwise known of, a Mantrabhårata, No. 1297. See Minkowski forthcoming b. 74 Puråña 50/152 and 51/153 of Kunhan Raja and Sarma 1944-1948. 50/152 is dated 1672 A.D. in the catalogue. 75 Khasmohor 2903 and 4761 of Bahura 1976. In Jodhpur there is an eighteenth century copy of the Mantraråmåyaña in the Oriental Research Institute collection, 187 / 9987 of Jinavijaya 1963ff. vol. 2 pt. 1. 76 Puråña 255 / 2244 and Tantra 564 / 13624 of Nanbiyar 1950, the former dated Sa∫vat 1735.

III. How to Produce, Construct and Legitimate a Tradition

245

shorter works that were ‘summaries of import’, and ‘illuminations of intent’. The commentarial efflorescence in this period would seem to be part of a general tendency in the period, in which competing, sectarian interpretations of shared, canonical texts came to be a matter of great importance. Nœlakañ™ha’s non-dualist commentary should probably be understood as competing against the sorts of views proferred by the sectarian religious views of vaißñavas such as Madhva, who had risen to great prominence in the Sanskrit sphere in the sixteenth century under the leadership of such figures as the Bengali theologians Caitanya and Jœva Gosvåmœ.77 Nœlakañ™ha’s intellectual project was continued in subsequent advaitin readings of the epic. In Bikaner, Dœkßita Mañiråma, mentioned above, made an ‘extraction of the essence’, (Såroddhåra) of Nœlakañ™ha’s commentary on two chapters of the epic.78 There was also the Mahåbhåratatåtparyaprakå†a, (‘Light on the Intent of the Mahåbhårata) of Sadånanda Vyåsa, an advaitin writing in the Punjab of the late eighteenth century (Misra 1992b). Some of the passages from Nœlakañ™ha’s Dœpa were taken verbatim into Sadånanda’s work, and it is clear that he had the work in front of him as he wrote.79 3. Conclusion In the second half of this paper I showed how Nœlakañ™ha’s text of the Mahåbhårata, together with his commentary on it, were disseminated widely and actively used in the century after their composition. I further argued that the principal reason for this success lay in the commentary’s innovative features, and in their expression in ingenious, idiosyncratic terms that were highly prized. This latter claim is of course difficult to prove, given the state of our knowledge about the Sanskrit literature and scholars of the period, but the existing evidence at least gives this claim greater plausibility than another possible explanation, viz. that Nœlakañ™ha’s commentary was valued simply for being conservative of the preceding literary tradition. Indeed, it would be safe to say that the current academic readers of the Mahåbhårata are more ‘conservative’ than Nœlakañ™ha was. This returns us to the paradoxical reception of Nœlakañ™ha’s commentary by modern Indologists. The successful publication history of the Dœpa is best explained by the fact of its pre-modern success, and the result77 See Minkowski 2004b on Nœlakañ™ha’s efforts to claim the formerly extra-canonical Bhågavatapuråña as a text assimilable to the non-dualist position. On the rise of the importance of the Bhågavatapuråña, see Majumdar 1961: 381-93. 78 I have not been able to examine this unpublished work. It did circulate, however, as there is also a nineteenth century copy of it in Jodhpur, 99 / 24809 of Jinavijaya 1963ff. vol. 6. 79 Sadånanda reuses statements of Nœlakañ™ha’s grand vision of the epic, as for example the opening sentence of the commentary to the first verse of the epic. Sadånanda also repeats Nœlakañ™ha’s ‘tetrad of topics’ nearly verbatim.

246

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

ant availability and active use that continued into the period of colonial transformation. Paradoxically, this success and availability also explains the Indological misdiagnosis of the work as ‘traditional’. The ‘traditional’ in premodern Sanskrit textual culture remained a nebulous concept for Indologists. In a trivial sense, the fact of a text’s being in Sanskrit would have made it ‘traditional’, over against the scholarly work produced by the scholar based in European languages. Slightly less trivially, the fact of its being canonically read by †åstrœs would also have made it ‘traditional’. Beyond that, as we have seen, the judgment of a commentary as ‘traditional’ could mean either that its meanings represented an unbroken continuous understanding of the meaning of the base text, or that it was anachronistic, discontinuous with that meaning. In either case, judgment by the criterion of accordance with the meaning ‘as it was’ incapacitates the ability to see the commentary’s position in the internal history of a textual community (Minkowski 2004a). Is there any chance that the arguments I have made here are likely to render Nœlakañ™ha’s interpretations of the Mahåbhårata more plausible to us as readers? Probably not. Nœlakañ™ha remains, even when rightly understood in his historical context, an idiosyncratic writer with a penchant for what I am sometimes tempted to (fondly) call nutty ideas, and nuttier ways of going about arguing them. I would not want to create an understanding of his work that smoothed out his eccentricities. Instead let us suppose that Nœlakañ™ha was successful in his day and after because he provided an innovative way of reading the Mahåbhårata, a reading that, not surprisingly, annoys Indologists today. At the same time, Nœlakañ™ha’s outlandishness has been normalized in a different sense; his penchant for the farfetched has become familiar to us as readers today. It has become the sort of thing that ‘traditional commentators’ are likely to do, and an intellectual mode that we can discover elsewhere, and be accustomed to encountering, without reference to any notion of literary history. Nœlakañ™ha’s work is in that sense a part of our habits of reading. As a way of closing, let me suggest that the history of Nœlakañ™ha’s success is probably not unique to him. Something similar might be true more generally about authors who wrote in the intellectual flowering in Sanskrit that took place in the early modern period. The summaries, categories, canonical lists, presuppositions and modes of argumentation might in many cases be constitutive of our current understanding of earlier Sanskrit intellectual traditions. The period’s salience and distinctiveness we no longer notice, but rather take for granted, not so much because of the divide between the scholars of that period and ours, but because of the unrecognized continuities.

III. How to Produce, Construct and Legitimate a Tradition

247

BIBLIOGRAPHY Ananta Krishna Sastry, R. (1921), Kavindracarya list, Central Library, Baroda. Aufrecht, T. (1891-1903), Catalogus Catalogorum: an alphabetical register of Sanskrit works and authors, Brockhaus, Leipzig, 3 voll. Bahura, G.N. (1976), Literary Heritage of the Rulers of Amber and Jaipur, Maharaja Sawai Man Singh II Museum, Jaipur. Bakre, Mahadev Shastri (ed.) (1915), Virå™aparvan, Gujarati Printing Press, Bombay. ______ (1920), Udyogaparvan, Gujarati Printing Press, Bombay. Belvalkar, S.K. (1947), Commentary of Devabodha on the Bhœßmaparvan, BORI, Poona. Berlinzola, M. (1939), Attendibilità del Commento di Nœlakañ™ha al Mahåbhårata, in «Rivista degli Studi Orientali», n. 18, pp. 268-284. Bhatt, B.N. (1982), An Analysis of the Råmåyañasårasa∫grahavivaraña of Appayya Dœkßita, in «Journal of the Oriental Institute, Baroda», n. 32, pp. 150-161. Bhatt, G.H. (1965), Råmåyaña Commentaries, in «Journal of the Oriental Institute, Baroda», n. 14, pp. 350-361. Biardeau, M. (1968), Some More Considerations About Textual Criticism, in «Puråña», n. 10, pp. 115-123. Bopp, F. (1824), Indralokågamanam- Ardschuna’s Reise zu Indra’s Himmel, Königliche Akademie der Wissenschaften, Berlin. ______ (1829a), Diluvium cum tribus aliis Mahá-Bhárati praestantissimis episodiis, Dmmler, Berlin. ______ (1829b), Sndflut nebst Drei anderen der wichtigsten episoden des Mahá-Bhárata, Dmmler, Berlin. Brinkhaus, H. (2002), The Division into Parvans and the Bhavißyaparvan of the Hariva∫†a, in Brockington, M. (ed.), Stages and Transitions: temporal and historical framworks in epic and puråñic literature, Proceedings of the second Dubrovnik International Conference on the Sanskrit Epics and Puråñas, Croatian Academy of Sciences and Arts, Zagreb. Brockhaus, H. (1862), Die 100 Parva des Mahâbhârata, in «Zeitschrift der deutschen morgenländischen Gesellschaft», n. 6, pp. 528-532. Brockington, J. (1998), The Sanskrit Epics, E.J. Brill, Leiden. ______ (1999), A descriptive catalogue of the Sanskrit and other Indian manuscripts of the Chandra Shum Shere collection in the Bodleian Library, Part. 2, Epics and Puråñas, Oxford University Press, Oxford. Bronner, Y. (1999), Poetry at its Extreme: The Theory and Practice of Bitextual Poetry, Ph.D. Dissertation, University of Chicago. Bhnemann, G. (1983), Budha-Kau†ika’s Rå∫arakßåstotra, Publications of the De Nobili Research Library, Wien.

248

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

van Buitenen, J.A.B. (1956), Vedårthasa∫graha, Deccan College, Poona. ______ (1975), The Mahåbhårata. The Book of the Assembly Hall; the Book of the Forest, University of Chicago, Chicago, vol. 2. Cardona, G. (2000), The manuscript history of the Vaiyåkarañasiddhånta-paramalaghuma∞jußå, 202d American Oriental Society Meeting, Portland. Dandekar, R.N. (1941), Commentary of Devabodha on the Ådiparvan of the Mahåbhårata, BORI, Poona. De, S.K. (1944), Commentary of Devabodha on the Udyogaparvan of the Mahåbhårata, Bharatiya Vidya Bhavan, Bombay. Dodson, M.S. (2002), Re-Presented for the Pandits: James Ballantyne, ‘Useful Knowledge’, and Sanskrit Scholarship in Benares College during the Mid-Nineteenth Century, in «Modern Asian Studies», vol. 36, n. 2, pp. 257-298. Emeneau, M. (1935), Union List of Printed Indic Texts and Translations in American Libraries, American Oriental Society, New Haven. Fitzgerald, J. (1991), India’s Fifth Veda: the Mahåbhårata’s Presentation of Itself, in Sharma, A. (ed.), Essays on the Mahåbhårata, E.J. Brill, Leiden. ______ (2000), Sanskrit pœta and †aikya / saikya: Two Terms of Iron and Steel technology in the Mahåbhårata, in «Journal of the American Oriental Society», n. 120, pp. 44-61. Geldner, K.F. (1951), Der Rig-Veda: Aus dem Sanskrit ins Deutsche bersetzt (Harvard Oriental Series 33), Harvard University Press, Cambridge. Gode, P.K. (1942), Nœlakañ™ha Caturdhara, the Commentator of the Mahåbhårata - his Geneaology and Descendants, in «Annals of the Bhandarkar Oriental Research Institute», n. 23, pp. 146-161. ______ (1943-44), The Kavindråcåryasüci –Is it a Dependable Means for the Reconstruction of Literary Chronology?, in «New Indian Antiquary», n. 6, pp. 41-42. ______ (1944), New Light on the Chronology of the Commentators of the Mahåbhårata, in «Annals of the Bhandarkar Oriental Research Institute», n. 25, pp. 103-108. ______ (1946), The Exact Date of the Advaitasudhå of Lakßmaña Pañ∂ita (A.D. 1663) and his possible identity with Lakßmañårya, the Vedånta teacher of Nœlakañ™ha Caturdhara, the Commentator of the Mahåbhårata, in «Poona Orientalist», n. 10, pp. 1-7. ______ (1951), Some contemporary Manuscripts of the Works of Nœlakañ™ha Caturdhara, the Commentator of the Mahåbhårata - Between A.D. 1687 and 1695, in «Journal of the S.M. Library», Tanjore, vol. 4, n. 1. Goldman, R.P. (1984), The Råmåyaña of Vålmœki: an epic of ancient India. Bålakåñda, Princeton University Press, Princeton, vol. 1.

III. How to Produce, Construct and Legitimate a Tradition

249

______ (1992), Translating Texts Translating Texts: Issues in the Translation of Popular Literary Texts with Multiple Commentaries, in Moore, C. (ed.), Translation East and West, East West Center and University of Hawaii, Honolulu. ______ (1995), Vålmœki Råmåyaña: Its Nature, History and Significance, East-West Center, Honolulu. Halbfass, W. (1991), Tradition and Reflection: Explorations in Indian Thought, State University of New York Press, Albany. Holtzmann, A. jr. (1884-85), Review of Roy / Ganguli’s translation of the Ådi- and Sabhå- parvans, in «Literatur-Blatt fr Orientalische Philologie», n. 2, pp. 71-73. ______ (1885-87), Review of Roy / Ganguli’s translation of the Vanaparvan, in «Literatur-Blatt fr Orientalische Philologie», n. 3, pp. 58-63. ______ (1888), Review of Roy / Ganguli’s translation of the Virå™a- and Udyogaparvans, in «Zeitschrift der deutschen morgenländischen Gesellschaft», n. 42, pp. 683-686. ______ (1892-95), Das Mahåbhårata und seine Theile, Haeseler, Kiel, 4 voll. Jhalkikar, B. (19784), Nyåyako†a, BORI, Poona. Jinavijaya, M. et al. (1963), A catalogue of Sanskrit and Prakrit manuscripts in the Oriental Research Institute, Rajasthan Puratana Granthamala (nos. 71, 77, 81-82, 85, 91, 125-127, 130-132, 136-140, 145-152, 167-169, 186), Rajasthan Oriental Research Institute, Jodhpur, 27 voll. Jones, W. (1788), On the Gods of Greece, India, and Italy. (written in 1784 and since revised), in «Asiatic Researches», n. 1, pp. 221-275. Kane, P.V. (1968-19772), History of Dharma†åstra, BORI, Poona, 5 vols. Karmarkar, R.D. (1949), Commentary of Devabodha on the Sabhåparvan of the Mahåbhårata, BORI, Poona. Khå∂ilkar, Åtmåråma (1862-1863), ‡rœmahåbhårate ådiparva [etc.], Gañapata Kr¢ßñåji, Bombay. Ki∫java∂ekara, R. (1929-1936), Mahåbhåratam with the Commentary of Nœlakañ™ha, Citrashala Press, Poona. Krishnacharya, T.R., Vyasacharya, T.R. (1906-1914), Srimanmahabharatam, a new edition mainly based on the South Indian texts, with footnotes and readings, Madhva Vilas Book Depot, Kumbhakonam. Kunhan Raja, C., Sarma, K. Madhava Krishna (1944-1948), Catalogue of the Anup Sanskrit Library, Government Press, Bikaner, 5 fascs. Levi, S. (1929), Review of Sukthankar 1933, in «Journal Asiatique», n. 215, pp. 345-348. Majumdar, B. (1961), The Bhågavata Puråña and its Influence in the Sixteenth Century, in «Journal of the Bihar Research Society», n. 47, pp. 381-393. Marulasiddhaiah, G. (1981), Descriptive Catalogue of Sanskrit Manuscripts, Oriental Research Institute, Mysore, vol. 7.

250

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Minkowski, C. (2000), Nœlakañ™ha’s Cosmographical Comments in the Bhœßmaparvan, in «Puråña», n. 42, pp. 24-40. ______ (2002), Nœlakañ™ha Caturdhara’s Mantrak农khañ∂a, in «Journal of the American Oriental Society», vol. 122, n. 2. ______ (2004a), Nœlakañ™ha’s Instruments of War: Modern, Vernacular, Barbarous, in «Indian Economic and Social History Review», n. 41, pp. 365-385. ______ (2004b), The Vedastuti and the Védisants, in Houben, J., Griffiths, A. (eds.), Proceedings of the Third International Workshop on Vedic Studies, pp. 125-42. ______ (Forthcoming a), Nœlakañ™ha Caturdhara and the Genre of Mantrarahasyaprakå†ikå, in Ikari, Y. (ed.), Proceedings of the Second International Vedic Workshop, Institute for Research in Humanities, Kyoto. ______ (Forthcoming b), Nœlakañ™ha’s Vedic Readings in the Harivaμ†a Commentary, in Petteri K. (ed.), Proceedings of the Third Dubrovnik Conference on the Sanskrit Epics and Puråñas. Mi†ra, M. (1988), Purovåk, in ‡rœmanmahåbhåratam, Nag Publishers, Delhi (reprinted from Khå∂ilkar 1862-1863). Misra, V.N. (1992a), ‡rœmahåbhåratatåtparyanirñaya of ‡rœ Ånandatœrtha (Madhvåcårya), Ratna, Varanasi. ______ (1992b), Mahåbhårata-Tåtparyaprakå†a of ‡rœ Sadånanda Vyåsa, Ratna, Varanasi. Mo™icandra, D. (1962), K农 ka Itihås, Hindi Grantha-Ratnåkar, Bombay. Mudaliar, A.L. (1968), Bibliography of the Books, Papers & Other Contributions of Dr. V. Raghavan, New Order Book Co., Ahmedabad. Muir, J. (18682) [1858], Original Sanskrit texts on the origin and history of the people of India, their religion and institutions, Trbner, London. Nanbiyar, R. (1950), An Alphabetical List of the Manuscripts in the Oriental Institute, Baroda, Gaekwad’s Oriental Series 114, Oriental Research Institute, Baroda, 2 voll. Pingree, D. (1981), Jyoti™†åstra: Astral and Mathematical Literature, Otto Harrassowitz, Wiesbaden. ______ (1997), From Astral Omens to Astrology, From Babylon to Bikaner, Serie Orientale Roma 78, Istituto Italiano per l’Africa e l’Oriente, Roma. Pollock, S. (1984), åtmåna∫ månußam manye: Dharmåkü™am on the Divinity of Råma, in «Journal of the Oriental Institute, Baroda», n. 33, pp. 231-243. ______ (1993), Råmåyaña and Political Imagination in India, in «Journal of Asian Studies», vol. 52, n. 2, pp. 261-297. ______ (2001a), New intellectuals in seventeenth-century India, in «Indian Economic and Social History Review», n. 38, pp. 3-31. ______ (2001b), The Death of Sanskrit, in «Comparative Studies in Society and History», vol. 43, n. 1, pp. 392-426.

III. How to Produce, Construct and Legitimate a Tradition

251

Printz, W. (1911), Bhåßå-Wörter in Nœlakañ™ha’s Bhåratabhåvadœpa und in anderen Sanskrit-Kommentaren, in «KZ», n. 44, pp. 69-109. Raghavan, V. (1940), Kavœndråcårya Sarasvatœ, in Bimal Churn Law (ed.), D.R. Bhandarkar Volume, Indian Research Institute, Calcutta. ______ (1941), Notes on some Mahåbhårata commentaries, in Katre, S.M., Gode, P.K. (eds.), A Volume of Studies in Indology, presented to Prof. P.V. Kane, Poona Oriental Series 75, Oriental Book Agency, Poona. ______ (1942), U∂årik®tå råmåyañavyåkhyå, Gopalakrishnamacharya Book of Commemoration, Madras. ______ (1980), The Råmåyaña in Sanskrit Literature, in Raghavan, V. (ed.), The Råmåyaña Tradition in Asia, Sahitya Akademi, Delhi. ______ (1988), Sanskrit Råmåyañas Other than Vålmœki’s: Adbhuta, Adhyåtma and Ånanda Råmåyañas, Raghavan Centre, Chennai. Raghu Vira (1936), The Virå™aparvan, BORI, Poona. Rangacharya, M. (1907), A descriptive catalogue of Sanskrit manuscripts in the Government Oriental Manuscripts Library, Government Press, Madras, vol. 4a. ______, Kuppuswami Sastri, S. (1913), Triennial Catalogue of manuscripts collected during the triennium 1910-1911 to 1912-1913 for the Government Oriental Manuscripts Library, Madras, Government Press, Madras, vol. 1, pt. 1. Renou, L. (1960), Le Destin du Véda dans L’Inde, in «Études Védiques et Påñiniéennes», n. 6, E. de Bocard, Paris. Sambasiva Sastri, K. (1937), A descriptive catalogue of Sanskrit Manuscripts in H.H. The Maharajah’s Palace Library, V.V. Press Branch, Trivandrum, vol. 2. Sastri, P.P.S. (1942), Commentators of the Råmåyaña in the Fifteenth, Sixteenth, and Seventeenth Centuries, in «Annals of the Bhandarkar Oriental Research Institute», n. 23, pp. 413-414. Sharma, R.K. (2001), Salient Features of Nœlakañ™ha’s Introduction to the Mahåbhårata, in «Puråña», n. 43, pp. 95-100. Shastri, H. (1912), Dakshini Pandits at Benares, in «Indian Antiquary», n. 41, pp. 7-13. ______ (1939), Descriptive Catalogue of the Sanskrit manuscripts in the Collections of the Royal Asiatic Society of Bengal, Royal Asiatic Society, Calcutta, vol. 8. Siromani, N., et al. (1834-1839), The Mahåbhårata: an epic poem written by the celebrated Veda Vyása Rishi, Education Committee’s Press (vol. 1), Baptist Mission Press (vol. 2), Asiatic Society of Bengal (voll. 3-4), Calcutta, 4 voll. ‡iromañi, R.N. (19992), An alphabetical list of manuscripts in the Oriental Institute, Vadodara, Oriental Institute, Vadodara, 2 voll. von Stietencron, H. et al. (1992), Epic and Puråñic Bibliography, Harrassowitz, Wiesbaden, 2 voll.

252

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Sukthankar, V.S. (1930), Arjunami†ra, in Dr. Modi Memorial Volume, Dr. Modi Memorial Committee, Bombay. ______ (1933), Prolegomena, in The Ådiparvan. For the First Time Critically Edited, BORI, Poona. ______ (1935-36), Epic Studies V: Notes on Mahåbhårata Commentators, in «Annals of the Bhandarkar Oriental Research Institute», n. 17, pp. 185-202. Telang, K.T. (1882), The Bhagavadgîtâ with the Sanatsugâtîya and the Anugîtâ, Sacred Books of the East 8, Oxford University Press, Oxford. Trautmann, T. (1997), Aryans and British India, University of California, Berkeley. Tubb, G.A. (1991), ‡åntarasa in the Mahåbhårata, in Sharma, A. (ed.), Essays on the Mahåbhårata, E.J. Brill, Leiden. Upådhyåya, B. (1983), K农 kœ Påñ∂itya Paraμparå, Vi†vavidyålaya Prakå†ana, Våråñasœ. Upadhyaya, T.P. et al. (1953-1965), A descriptive catalogue of the Sanskrit manuscripts acquired for and deposited in the Sanskrit University Library (Sarasvati Bhavana), Varanasi, during the years 17911950, Sampürñånand Sanskrit University, Benares, 9 voll. Vaidya, P.L. (1969), Hariva∫†a, Bhandarkar Oriental Research Institute, Poona, vol. 1. Weber, A. (1850), Madhusûdana-Sarasvatî’s encyclopädische Uebersicht der orthodoxen brahmanischen Literatur, in «Indische Studien», n. 1, pp. 1-24.

FRANCESCO SFERRA Constructing the Wheel of Time. Strategies for Establishing a Tradition*

The founding project of a tradition is usually expressed in its early texts, but not always explicitly. What its authors say –especially in the passages in which they introduce the tradition–, what they omit and, above all, the way in which they communicate content (apart from its being true or untrue) reveals their inner concerns, their perception of the surrounding reality and the motivations, at least in part and aside from the ones given, that led them to write. The first steps taken in establishing traditions, and particularly their founding projects, are especially interesting for historical research, which does not merely record the facts but also explores their mystification and enquires into the reasons for this. The Buddhist tantric system known as the Wheel of Time (Kålacakra), which spread through Northern India and Tibet around the beginning of the XI cent., is extremely interesting in this regard and has been studied relatively little according to the aforesaid perspective, which is the one I intend to explore here. Except for the premises necessary for my discourse (§ 1), I shall not reiterate the basic historical information on this system, for which I refer the reader to the contributions by John R. Newman.1 It is possible to show that the system’s authoritativeness was established by the early teachers of the Kålacakra and by the first Tibetan historiographers through a precise intellectual ‘operation’, * I am grateful to Raniero Gnoli, John R. Newman and Harunaga Isaacson for having read this paper before it was published and for their valuable advice and suggestions; to Federico Squarcini, the editor of this volume, for his pertinent and useful observations; and to Susan Ann White for her help in revising the English text. Within this article numbers in subscript indicate the lines of the pages/folios of printed editions and MSS referred to. 1 Newman 1987a: 70-113, 19912, 1998b, 2004; cf. also Gnoli 1994.

254

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

which involved, among other things, the definition of the Scriptures and of the qualities that their interpreter should possess, the fixing of hermeneutical criteria and the choice of the main themes. They sought to establish, within a consistent framework, a new Buddhist orthodoxy and to close the Buddhist ranks around the undisputed authority, not without an anti-secular vein, of the monastic community, probably in order to effectively meet the new socio-cultural challenges that had arisen. The following is therefore an attempt to shed light on some of the cultural forms and means through which the Kålacakra authors utilized and shared semiotic construction strategies to legitimize their lineages and doctrine. They did this in various ways: by establishing a corpus consisting of three works through which, on the one hand, they commented on celebrated works thus associating them with their own system and, on the other, founded that same system (§§ 2-4); by stating that their system began to spread at a precise and significant point in history (§ 5); by drawing on a higher level of the scriptural tradition, used as a means of interpretation and available only to a select few (§ 6); and by describing the qualities a master must possess and those necessary for an author worthy of writing commentaries (§ 7).2 1. Classic sources for studying the early history of the Kålacakra are contained in some of the oldest Sanskrit texts belonging to this tradition –particularly the Laghukålacakratantra (LKCT), the Vimalaprabhå (VP) by Puñ∂arœka, which is a commentary on it, and the introduction of the Sekodde†a™œkå (SU~) by Nåropå– and in various Tibetan historiographies, beginning with the one in rGyud sde’i zab don sgo ’byed rin chen gces pa’i lde mig ces bya ba written by Bu ston Rin chen grub (Bu ston, 1290-1364) in 1329,3 the one in Dus ’khor ™œkå chen4 by mKhas grub rje dGe legs dpal bzaõ po (mKhas grub rje, 1385-1438), written in 1434,5 and the one included in the Deb ther 2 Part of the material presented in § 4 and in § 6 was previously analyzed in Sferra 2001. 3 It is generally held that the sources used by Bu ston are very old and that some of them are based on oral traditions and perhaps date back to the very origins of the system. Bu ston narrates the history of the Kålacakra according to its two most important lineages in Tibet: 1) the Rwa lineage and 2) the’Bro lineage. The first originated from Rwa chos rab and the Newar Samanta†rœ around the end of the XI cent., the second from ’Bro †es rab grags and the Ka†mœrian Somanåtha around the middle of the XI cent. On these lineages, see Newman 1987a: 94 ff. and Orofino 1994: 17 ff. Bu ston’s account of the history of the Kålacakra in India according to the Rwa and the ’Bro lineages (fols. 28b3-31a2 = ed. Chandra, fols. 56-61) has been translated by J.R. Newman (1987a: 76-89; 19912: 66-71). 4 The final å of the word ™œkå is sometimes dropped by Tibetans in writing the title of this work. Here I adopt the correct form that occurs also in a copy of the collected works by mKhas grub rje kept in the Tucci Tibetan Fund in the Library of IsIAO (ex IsMEO); cf. De Rossi Filibeck 1994: 76. 5 The description of the Rwa and ’Bro lineages in the work by mKhas grub rje (pp. 167173) follows –sometimes verbatim– the text by Bu ston, especially where the Rwa lineage is concerned. In his translation of Bu ston’s account, J.R. Newman has translated the sentences added by mKhas grub rje between brackets (for references, see above note 3).

III. How to Produce, Construct and Legitimate a Tradition

255

sõon po (Blue Annals), written between 1476 and 1478 by ’Gos lo tså ba gz^on nu dpal (1392-1481),6 which are the most ancient. Tibetan historiographies present valuable information not given in the Sanskrit texts: they illustrate the early stages of the system’s diffusion by providing the names of the first teachers and by describing some events concerning them and their works. They record the existence of different lines of transmission of the teaching (’Bro and Rwa), differ with regard to several details, but agree, also with the Sanskrit sources, in recognizing that the scriptural nucleus of the system was composed of a short tantra (laghutantra), the LKCT, and a root-tantra (mülatantra), the Ådibuddha (or Paramådibuddha). The latter has never been translated into Tibetan and is preserved only in fragments, the longest of which is the Sekodde†a (SU, fortunately translated into Tibetan twice), which is held to be a section of the fifth chapter –or of the first chapter, according to the first Tibetan translation–7 of the work. Despite the importance attributed to the Ådibuddha, the LKCT, which summarizes its teachings, is generally considered the basic text of the system. It is a long encyclopaedic work that, according to the tradition, consists of 1,030 sragdharå stanzas;8 however, it is actually composed of 1,048 stanzas, although some are repeated.9 The tradition maintains that it was written down by Ya†as (also called ‡rœya†as or Ma∞ju†rœya†as) upon the request of the bråhmañas of Sambhala (alias ‡ambhala).10 Two other works are strictly linked to this nucleus. They are quoted as authoritative texts in Sanskrit and Tibetan sources, but have come down to us in their entirety only in the bKa’ ’gyur: the *‡rœkålacakratantrottaratantrahr¢dayanåma (Tantrottara), a supplement to the LKCT, and the *Sekaprakriyå, which deals with the initiation rites.11 Tibetan sources also concur that three exegetical works –which, significantly, are attributed to Bodhisattvas (while commentaries are usually written by pañ∂itas or siddhas)–, the Laghutantra™œkå (LT~) 6

Chapter 10, fols. 1a-41a = ed. Chandra, fols. 661-741; transl. Roerich 19762: 753-838. Cf. Orofino 1994: 14-15. 8 Cf. VP vol. 1, p. 25 . 6 9 LKCT 2.116-121 in the printed editions are an obvious interpolation (J. Newman, personal communication). These stanzas, which correspond to LKCT 4.192-197, do not appear in the Tibetan translation of the second chapter of the LKCT (cf. VP vol. 1, pp. 232-233, note). 10 For a list of titles of the extant Kålacakra literature, see Lål 1994 (Sanskrit) and Samphel 1995 (Tibetan). 11 The Tantrottara (Q vol. 1, #5, fols. 142a -158a ) is quoted here and there in other 1 3 works (e.g., in the ¯a∂aõgayoga by Anupamarakßita, ed. pp. 137-139 [for further references, cf. ed. p. 65]). The *Sekaprakriyå (Q vol. 1, #7, fols. 160a1-164a5), in fact, is wholly composed of stanzas from the LKCT: stt. 1-12 = LKCT 3.92-103; stt. 13-21 = LKCT 3.118126; st. 22 = LKCT 5.112; st. 23 = LKCT 4.119; stt. 24-37 = LKCT 5.113-126; at the beginning, between stanzas 21 and 22, and at the end, there are a few lines in prose. The commentary (vr¢tti) by Dårikapå on this text is preserved in the Tibetan translation (Q vol. 47, #2072, fols. 48b5-87b5). 7

256

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

by the Bodhisattva Vajrapåñi, the ¯a™såhasrikå by the Bodhisattva Vajragarbha and the Vimalaprabhå (VP) by the Bodhisattva Puñ∂arœka, formed a body of integrated texts, known as the ‘Bodhisattva corpus’ (byaõ chub sems dpa’i skor) or the ‘Bodhisattva commentaries’ (byaõ chub sems dpa’i ’grel pa rnams), which were fundamental to the spreading and the correct interpretation of the system. 2. However we may suppose that the concept of the corpus was superimposed later as a part of a founding project, and therefore extraneous to the three texts. In particular, the LT~ and the HTP~ were probably written at a time when the doctrines and practices of the ‘Wheel of Time’ were in the formative stage, even if the basic conception of the system was already in place when they were composed. These works –as their acceptance of the authority of the Paramådibuddha clearly shows– were the products of the initial effort to establish the tradition, although the founding project of which they were part may not have been well-defined at that point. This would appear to be confirmed by the following: 2.1 Of this trilogy, the only text directly linked to the Kålacakra is the VP, an exhaustive commentary on the LKCT and the only complete one: Vajrapåñi merely comments on the first ten and a half verses of the Cakrasa∫varatantra (CST, also called Herukåbhidhånatantra or simply Laghvabhidhåna), which correspond to a part of the first chapter, and Vajragarbha dwells on the first five pa™alas of the first part (kalpa) of the Hevajratantra (HT), which, according to him, contain the ‘concise meaning’ (piñ∂årtha)12 of the whole text. Hence, this commentary (™œkå) is also called Hevajratantrapiñ∂årtha™œkå (HTP~).13 Accordingly, in the conclusion of the LT~ Vajrapåñi defines his work as a piñ∂årthavivaraña, lit. an ‘explanation of the concise meaning’, since in it he briefly illustrates the condensed meaning [of the short tantra], which is contained in the first chapter.14 Tibetan historiographers also provide interesting information on the first phase of the system’s diffusion, including an explanation of why the LT~ and the HTP~ are incomplete, but they do not tell us why the mülatantra has not come down to us in its entirety or why it has not been translated into Tibetan –as we shall see, their silence on this point is not devoid of meaning. Nor do they tell us who Vajrapåñi, Vajragarbha and Puñ∂arœka actually were. The authors 12 With this meaning the compound piñ∂årtha recurs in the commentarial literature in Sanskrit; for instance, in the ¯a™ko™ivyåkhyå by Candrakœrti (idånœ∫ saptada†apa™alånå∫ piñ∂årtho vidhœyate; p. 5) and in the Vyåkhyåna by Ruyyaka on the Vyaktiviveka by Mahimabha™™a (ayam atra piñ∂årthaΔ […]; p. 16). 13 All passages from the HTP~ in this paper are taken from my edition of the text (Sferra 1999), to which I refer the reader for variants in the MSS and the readings of the Tibetan translation. The stanzas are quoted with the number of the section and the number of the stanza while the parts in prose are quoted with the reference to MSS. 14 iha laghutantra™œkåyå∫ sa∫kßepeña piñ∂årthaΔ praka™œkr¢to […] piñ∂årthavivaraña∫ nåma prathama™œkåparicchedaΔ (p. 159).

III. How to Produce, Construct and Legitimate a Tradition

257

themselves do not mention their own teachers in their works nor do they elucidate –as was the custom– on the historical circumstances in which they worked. The importance of this trilogy is stressed in the Tibetan historiographies, which state that the first teachers of the school transmitted their knowledge of the Bodhisattva corpus to their disciples. Cilu (Tsi lu), the first teacher in the Rwa tradition realized, as Bu ston and mKhas grub rje inform us, that this corpus was fundamental to understanding the Mantrayåna and to achieving Buddhahood in a single life.15 Piñ∂a (Piñ∂o), Kålacakrapåda the Elder, Kålacakrapåda the Younger and Cilu himself must each have studied this trilogy and, in some cases, after undertaking difficult and tiring journeys. Likewise, mKhas grub rje informs us that Kalkœ ‡rœpåla, the first master in the ’Bro tradition,16 taught all the niruttaratantras and the Bodhisattva corpus to his disciple, who later became known as Kålacakrapåda [the Elder].17 The above-mentioned historiographers themselves studied this trilogy at length. It is well-known that Bu ston and mKhas grub rje, in particular, wrote detailed commentaries on the VP, and that the VP and the other glosses of the Bodhisattvas are included in the list of works that ’Gos gz^on nu dpal claims to have learnt from his master Saõs rgyas rin chen po (1336-1424).18 The Indian tradition would also appear to confirm the significance of these texts, although indirectly, through quotations or references to individual works. The longer treatise on the sixfold yoga by Anupamarakßita (X-XI cent.) and, later, the SU~ by Nåropå (956 ca1040 ca),19 for instance, draw heavily and openly on this trilogy. The latter work summarizes the most important themes of the Kålacakra system, often using the words of the three Bodhisattvas and particularly those of Puñ∂arœka.20 From the Deb ther sõon po we know that the celebrated pañ∂ita Vanaratna (Nags kyi rin chen, 1384-1468) composed a 15 […] spyir saõs rgyas tshe gcig gis ’grub pa la sõags kyi theg pa dgos †iõ khyad par du byaõ chub sems dpa’i ’grel pas gsal bar byas pa dgos la […] (Bu ston, fol. 28b6-7 = ed. Chandra, fol. 56; cf. Newman 1987a: 78). 16 He should probably be identified with Piñ∂o (cf. Newman 1987a: 99-102). 17 The same concept appears in Bu ston: […] naõ byaõ chub sems dpa’i skor gsum la sogs pa rgyud sde maõ po thugs la bzuõ nas byon pas ’jam dpal kyi sprul pa yin par grags †iõ Ù de’i mtshan yaõ dus ’khor z^abs †es zer ro Ù (Bu ston, fol. 30a1-2 = ed. Chandra, fol. 60; cf. Newman 1987a: 86). 18 de la bdag gis kun mkhyen chen po’i dkyil mchog gi steõ nas dus kyi ’khor lo’i dbaõ yoõs su rdzogs pa Ù rgyud ’grel chen mo’i bka’ Ù yan lag drug gi khrid Ù dbaõ mdor bstan nå ro pa’i ’grel pa daõ bcas pa daõ Ù sems ’grel skor gz^an rnams kyaõ thos so Ù (Deb ther sõon po, chapter 10, fol. 13a12 2 = ed. Chandra, fol. 68; cf. Roerich 1976 : 780). 19 J. Newman has pointed out to me that there is no convincing evidence that establishes Nåropå’s birth date at 956 CE. It is probably a relatively late Tibetan ‘reconstruction’ or ‘invention’ (cf. also Newman 1998a: 315-316, note 8; 1998b: 347, note 10). 20 Nåropå in his SU~ (cf. ed. Carelli, p. 38 ff., pp. 59-60) and Anupamarakßita in his ¯a∂aõgayoga (Q vol. 47, #2102, fol. 333a and fol. 339a) quote sentences verbatim from Vajrapåñi’s and Vajragarbha’s works. J. Newman has kindly brought to my attention that in the long commentary on the HT attributed to s˚an grags bzaõ po (*Ya†obhadra) (sDe dge #1186), whom the Tibetans identify as Nåropå, there are many passages from Kålacakra texts, the Bodhisattva Corpus and the Pa∞calakßahevajra (PLH).

258

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Tibetan commentary on both the HTP~ and the LT~ and on the first two chapters of the VP,21 but I was unable to locate copies of these texts. The standing of Puñ∂arœka, Vajrapåñi and Vajragarbha is also implicitly indicated in a passage of the Åmnåyama∞jarœ, the commentary by Abhayåkaragupta (XI-XII cent.) on the ‡rœsa∫pu™atantraråja. Here, the latter quotes a criticism of the Kålacakra voiced by other Buddhists (probably Ratnåkara†ånti and Vågœ†varœkœrti)22 who, while pointing out that there were many contradictions between the basic Kålacakra texts and the [three] vehicles, and that the authors were not Bodhisattvas as they claimed, groups them, significantly, with Ya†as, and elevates them to the status of principal authors of the system. Although quotations from the works by Vajragarbha, Vajrapåñi and Puñ∂arœka were made more frequently with the passing of time, for example in the Amr¢takañikå (AK) by Ravi†rœj∞åna (XI-XII cent.) and in the Amr¢takañikoddyota (AKU) by Vibhüticandra (XII-XIII cent.),23 there is no reference to these texts as a corpus in any Sanskrit work. 2.2 Vajrapåñi’s and Vajragarbha’s link to the Kålacakra is not openly declared. 2.2.1 Neither Vajrapåñi nor Vajragarbha specifically mentions the Kålacakra as a tradition: they do not refer to the stories concerning the founding of the system, which, by contrast, are given considerable importance in the LKCT and the VP, and other later texts. In fact, they make no reference to the traditional account of the teaching of the Kålacakra that the Buddha imparted at Dhånyaka™aka, to the realm of Sambhala, where the teaching was preserved and transmitted for centuries, or to the kings of Sambhala and the myth of Kalkin. 21 […] dpal dus kyi ’khor lo’i le’u daõ po g∞is kyi ’grel b†ad Ù rdo rje s∞iõ ’grel daõ Ù phyag rdor stod ’grel g∞is kyi dka’ ’grel mdzad de Ù sems ’grel skor la bod yig byuõ ba la bzaõ Ù […] (Deb ther sõon po, chapter 10, fol. 34a1-2 = ed. Chandra, fol. 727; cf. Roerich, 19762: 824-825). 22 The passage (sDe dge #1198, fols. 198b -199a ) is transcribed and translated in 5 3 Newman 1987a: 108-109. Cf. also Newman 1987a: 107, 110. 23 In the AK there are two quotations from Vajrapåñi’s work: 1) ad Ma∞ju†rœnåmasaõgœti (MNS) 6.20cd-22ab, where a passage is quoted in an abbreviated form from the edited text (LT~ p. 1403-21) and without mentioning the source (pp. 4524-465), 2) ad MNS 8.13, where a passage is quoted with slight changes from the edited text (LT~ p. 15027-32): tad ukta∫ vajrapåñipådaiΔ ‘praj∞åcumbanenånandakßaño bhavati […] j∞ånasamaya† caturthaΔ’ (p. 657-9); and no reference to Vajragarbha’s work. In the AKU there are two references to the HTP~: 1) ukta∞ ca vajragarbhe ‘iha vairocano bhümicakre […] vajrasattvo j∞ånacakre nåyaka iti’ (ad AK 3.2 [= commentary on MNS 3.2]; p. 13014-16); the quotation occurs in the 10th pariccheda of the HTP~ ad HT 1.5.11. 2) Quotation of HTP~ 6.17 with slight changes and without mentioning the source: tad uktam — ‘pråñåpånakßayeñaiva dvåda†åõgakßayo bhavet Ù hetuphalanirodhena ko na buddho bhavißyati ÙÙ’ iti (p. 11811-12); Vajragarbha does not specify whether this is a passage from the mülatantra of the HT, which is his main source. There are many citations from the VP in the Kålacakrabhagavatsådhanavidhi by Dharmåkara†ånti (XI cent.). The LT~ is extensively quoted in the ˘åkinœjålasa∫vararahasya by Anaõgayogin (p. 35-7 = LT~ p. 12110-13; p. 39-16 = LT~ p. 1237-16; p. 71-4 = LT~ pp. 13821-1393; p. 74-6 = LT~ p. 13913-15; p. 78-19 = LT~ p. 1407-21). As H. Isaacson has kindly pointed out to me, the reference to the Hevajra™œkå in AK ad MNS 1.1 (p. 26-7) is not to Vajragarbha’s HTP~ but probably to a commentary (lost in Sanskrit) on the HT called Suvi†adasampu™a (see Q #2314 and #2321); the passage of the AK should be read ‘not

III. How to Produce, Construct and Legitimate a Tradition

259

2.2.2 The word Kålacakra never appears in the LT~, and only three times in the HTP~ as the name of the tantric deity: in stanzas 49 and 112 of the 5th pariccheda –in a long quotation from the supposed mülatantra of the HT, namely the Pa∞calakßahevajra (PLH), to which we shall return later– and in the 10th pariccheda, in the commentary on HT 1.5.9, where the twenty-four armed Heruka is identified with Kålacakra and Ådibuddha.24 The latter, in particular, appears again in the HTP~, as an attribute of the tantric deity, in stanza 173 of the 6th pariccheda, the last stanza of the section, ad HT 1.2.12-18, which in actual fact corresponds to Ma∞ju†rœnåmasaõgœti (MNS) 8.24, even though the latter is not specifically mentioned. 2.2.3 The LT~ and the HTP~ never quote from the LKCT; the only Kålacakra scripture cited therein is the Ådibuddha and in particular the SU: the few stanzas from the Ådibuddha can all be traced to the SU, with the exception of three stanzas quoted on pp. 124-125 and 147 of the edition of Vajrapåñi’s work.25 In the LT~, the quotations from the SU appear at three different points: 1) p. 126, where st. 10 is introduced with tathå ådibuddhe bhagavån åha; 2) p. 127, where we find stt. 8, 1517ab preceded by atra paramådibuddhe bhagavån åha; and 3) p. 157, where stt. 139 and 135 are preceded by tathå paramådibuddhe sekodde†e bhagavån åha. The last quotation is the only explicit reference to the SU: at this point Vajrapåñi (or someone on his behalf) felt the need to specify: ‘In the Paramådibuddha, [and more precisely] in the Sekodde†a’, which as we have already mentioned is considered a part of the former. In the HTP~ there are two references to the Ådibuddha; the first ad HT 1.5.8, where SU 86-87 are quoted, after they have been introduced with tathå cåha sa∫vr¢tyå ådibuddhe, and the second at a crucial point in the text, at the end of the first pariccheda (st. 76), where an important hermeneutic criterion, which we shall discuss briefly later, is established, namely that the deep meaning (nœtårtha) of the Cakrasa∫vara, the CatuΔpœ™haka and the HT must be understood through the words of the Ådibuddha, which contain great secrets.26 2.2.4 Neither the LT~ nor the HTP~ makes any mention of Kålacakra teachers or of other independent works or commentaries suvi†adasphu™a∫ hevajratantra™œkåyå∫ as Lal suggests but suvi†adasampu™ahevajratantra™œkåyå∫ (the MS Cambridge University Library Add 1108 reads suvißadasampu™ahevajratantra™œkåyå∫; vißada is an often found alternative orthography for vi†ada)’. The same reading of the MS kept in Cambridge occurs –but with the variant sva° instead of su°– in the MS NAK 4/21, NGMPP Mf. B24/23, fols. 2r5-2v1 (I have consulted a photo of this MS made by G. Tucci and kept in the Library of IsIAO in Rome [= folder 3.42]). 24 nåyakaΔ sahajånandaΔ †ünyatåliõgito hevajro bhagavån caturvi∫†atibåhur dviguñaΔ ßa™cakrasa∫varåt kålavi†eßeñåvasthitaΔ […] eva∫ ßa™kulåtmako herukaΔ kålacakraΔ sa evocyata ådibuddha† ceti (Kaiser Library, Kathmandu, MS 128, NGMPP Mf. C14/6, fols. 54v6-55r1, 4-5). 25 The stanza cited on p. 124, which reads karmamudrå∫ parityajya j∞ånamudrå∫ vikalpitåm Ù paramåkßarayogena mahåmudrå∫ vibhåvayet ÙÙ, is also quoted in the VP (the PAJS), with attribution to the Mülatantra, vol. 3, p. 80 (H. Isaacson, personal communication). 26 ‘The Cakrasa∫vara and the CatuΔpœ™haka must be understood through the Hevajra. The Hevajra and the CatuΔpœ™haka must be understood through the words of

260

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

of this tradition, not even the VP. By contrast, Vajrapåñi and Vajragarbha frequently quote from authoritative texts that do not belong to the Kålacakra tradition, such as the ˘åkinœvajrapa∞jara, the Guhyasamåjatantra (GST), the Måyåjåla, the Mahåmåyå, the MNS and, among non-scriptural sources, the Gurupa∞cå†ikå attributed to Åryadeva, but, in the case of the latter, without mentioning the source or the supposed author (like Puñ∂arœka, in his turn). 2.3 None of the works of the trilogy makes reference to an intrinsic relationship with the others. Vajrapåñi mentions neither Vajragarbha and the HTP~ nor Puñ∂arœka. Vajragarbha, in his turn, makes only one explicit reference to Vajrapåñi’s LT~ at the end of the fifth section of his text, where he mentions the commentary on the first ten and a half stanzas of the CST that we know to be by Vajrapåñi ([…] asyaiva kulapa™alasya punar nå∂œsa∫caro laghucakrasa∫vare sårdhada†a†lokapiñ∂årtha™œkayå j∞åtavyaΔ).27 Puñ∂arœka, for his part, never refers to Vajrapåñi and Vajragarbha as celebrated Kålacakra authors, nor do their names appear in the myth concerning the system’s foundation or the teaching lineages. In the VP (ad LKCT 5.18) there is only one specific reference to the ™œkå by Vajrapåñi, which, like the HTP~, is called ¯a™såhasrikå: […] evam aß™a†ma†åneßu devyo veditavyåΔ Ù åså∫ vispharañena karmaprasarådika∫ tantrokta∫ vajrapåñikr¢ta™œkayå ßa™såhasrikayå boddhavya∫ laghutantre Ù tenåtra na likhitam (vol. 3, p. 1326-28), and none at all to either Vajragarbha himself or his HTP~. We may suppose, nevertheless, that Puñ∂arœka was familiar with Vajragarbha’s work, since, in at least one point of the VP (ad LKCT 5.9), he paraphrases, in prose, three stanzas, which appear in the 3rd pariccheda (ad HT 1.1.7) of the HTP~, and which are taken, according to Vajragarbha, from the PLH: samåjådœni tantråñi praj∞opåyåtmakåni vai Ù yogatantråñi sarvåñi praj∞opåyåtmanåmabhiΔ ÙÙ sa∞cåro yoginœnå∫ tu yatropåyasya sa∫sthitiΔ Ù sa∫vr¢tyå yoginœtantra∫ bålånå∫ gadita∫ mayå ÙÙ yatropåyasya sa∞cåraΔ praj∞åyåΔ sa∫sthitir bhavet Ù upåyatantram evokta∫ sa∫vr¢tyå tu yathå tathå ÙÙ (stt. 27-29). the Cakrasa∫vara. The Hevajra and the short Cakrasa∫vara must be understood through the words of the CatuΔpœ™haka. But the deep meaning [of all these tantras] must be understood through the words of the Ådibuddha, which contain great secrets’ (hevajreña hi cakrasa∫varam ida∫ j∞eya∫ catuΔpœ™haka∫ hevajra∫ khalu cakrasa∫varapadair j∞eya∫ catuΔpœ™hakam Ù hevajra∫ laghucakrasa∫varam ida∫ j∞eya∫ catuΔpœ™hakair nœtårthaΔ punar ådibuddhavacanair j∞eyo mahåsa∫varaiΔ ÙÙ). The CatuΔpœ™hamahåtantra (alias Prakarañatantra) is still unpublished. I give here references to some MSS of this text: 1) NAK, MS 5-37, vi 51, NGMPP Mf. A138/10; 2) NAK, MS 5-38, vi 52, NGMPP Mf. B112/4; 3) NAK, MS 1-1078 vi, ‡aivatantra [sic!], NGMPP Mf. B26/23; 4) NAK, MS 4-20 vi, Bauddhatantra 65, NGMPP Mf. A 48/18; 5) NAK, MS 5-36, NGMPP Mf. A 138/10; 6) Cambridge University Library, MS Add. 1704 (12). I thank H. Isaacson for supplying some of these references. 27 Kaiser Library, Kathmandu, MS C 128, NGMPP Mf. C14/6, fols. 25v -26r ; NAK MS 6 1 A 1267/6, NGMPP Mf. A693/11, fol. 25r6-7; IsIAO, Rome, MS 1.20, 22/S, MT049, fol. 16r5 [photos of the same MS taken by R. Såõkr¢tyåyana and listed by him as MS XVII.2.92; cf. Såõkr¢tyåyana 1935: 36].

III. How to Produce, Construct and Legitimate a Tradition

261

Puñ∂arœka’s paraphrase is as follows: yoginœyogatantram iti yatra yoginœnå∫ sa∞cåro nåyako ni†calaΔ sa∫vr¢tyå tad yoginœtantram, yatropåyasya sa∞cåraΔ praj∞å ni†calå tad upåyatantram Ù svarüpataΔ sarvam eva praj∞opåyåtmaka∫ yogatantram (vol. 3, p. 623-25). We cannot exclude that he might have drawn on another source; however, it is worth noting that also in another part of the VP (vol. 2, pp. 233-235) Puñ∂arœka may have based himself on the HTP~. In fact, he quotes with few variations 33 stanzas from the root-tantra (tathå mülatantre bhagavån åha […]) –which in his text means from the Ådibuddha– that Vajragarbha cites from the PLH (= 6.116-121ab, 122-131ab, 132, 148cd-164).28 As we shall see later, Vajragarbha himself probably composed all –or at least a fair amount of– the stanzas he attributes to the PLH. Furthermore, the classification of phonemes on the basis of astronomical interrelationship that we find in the VP (ad LKCT 1.8) probably derives from a scheme elaborated in the HTP~ (4th pariccheda, ad HT 1.1.13-21). The above allows us to establish a likely chronological order for the composition of the texts: LT~, HTP~, VP. The LKCT and the VP were probably only works in progress when LT~ and HTP~ were completed. 3. In the light of this, we must seek to understand what it is exactly that enabled these texts to be grouped together in a single corpus as part of a founding project. We could start by saying that they derive from a common cultural environment: not only are certain themes treated in all three, but, notwithstanding the fact that the three texts were definitely composed by different people, each with an individual style,29 we find passages that literally correspond. Even if we were to suppose that Puñ∂arœka was not familiar with Vajragarbha’s work, they both mention, as we have seen above, the work by Vajrapåñi, and sometimes even seem to depend on him; for instance, several sentences of the 2nd pariccheda of the HTP~ and the tantrade†anodde†a of the VP (vol. 1, pp. 12-22) would appear to be an elaboration of the first part of the LT~ (p. 44 ff.). Moreover, the texts are also linked by the fact that their authors chose particularly meaningful pseudonyms to give themselves a certain authority, as J.R. Newman has already pointed out in the case of Ya†as and Puñ∂arœka, who declare themselves to be manifestations of Ma∞ju†rœ and Avalokite†vara respectively.30 It is a known fact that Vajrapåñi, the Bodhisattva emanating from Akßobhya, is the mediator between the Buddha and man in the imparting of tantric teachings,31 28 It is worth noting that stanza 6.155a, which in the HTP~ is mülatantreßu sarveßu, appears as ådibuddhe mahåtantre in the VP (vol. 2, p. 23426). 29 I am aware that such a statement requires a deeper analysis of the language and syntax of the LT~, the HTP~ and the VP. I shall give more information on this in the introduction to my revised edition of Sferra 1999, which is in press. 30 Cf. Newman 1998a: 314; 1987a: 70. Cf. also Gnoli 1994: 63. 31 On Vajrapåñi, see Lamotte 1966 and Snellgrove 1987, vol. I: 134-141.

262

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

while Vajragarbha is the Buddha’s interlocutor in the HT. In his HTP~, Vajragarbha openly identifies himself with the Buddha’s interlocutor in the HT (1.71). These elements, although significant, are not the prerogative of the Bodhisattva corpus and can also be found in Buddhist and nonBuddhist works, and therefore the grouping of the LT~, HTP~ and VP in a corpus cannot be due solely to them. Perhaps more important, although the foregoing should not be minimized, are the following six considerations: a) the importance these three texts give to the Ådibuddha; b) the common hermeneutic criterion they all share, which can be summed up by the concept, which appears frequently in Kålacakra literature, that a Tantra must be understood through another Tantra (tantra∫ tantråntareña boddhavyam),32 and, more specifically, that every tantra must be understood through the Ådibuddha, as we have seen in the above passage of the HTP~ (§ 2.2.3 and note 26); c) the fact that all three Bodhisattvas quote the mülatantras of the tantra that they are commenting on33 (both Vajrapåñi and Vajragarbha quote the Ådibuddha as well); d) the unusual claim made by the three authors to have unique authority in interpreting the tantra that they comment on, by right of their identity and their asserted access to the mülatantras (which would be plausible if their claims regarding their own identity were accepted), which is precluded to other commentators; e) the aim of establishing the supremacy of monks over laymen (see below, § 7); and, perhaps most important, f) the desire to introduce the new Kålacakra doctrines and practices in religious circles where the works of the Adamantine Vehicle –and in particular the yoginœtantras where the LT~ and the HTP~ are concerned– were studied. We may suppose that the idea of a corpus –which is not consonant with the original texts– was introduced as part of a broader founding project, and used to ‘incorporate’ the well-established traditions of the CST and the HT, the two most celebrated yoginœtantras, in the nascent Kålacakra system and to give it greater credibility. 3.1 This need to incorporate or associate well-established traditions in or with the new system and to give it more prestige, also explains the importance the first authors of the Kålacakra attributed to the MNS. As we know, the MNS –on which there are several commentaries in both Sanskrit and Tibetan, and which takes many of the names ‘chanted’ therein from the celebrated vaißñava Vißñusahasranåma– was already extremely well-known when the early Kålacakra teachers were active. Beginning with Puñ∂arœka, an increasingly influential role was assigned to it also with regard to the interpretation of the 32 Cf., for instance, LT~ p. 145 ; HTP~ 1.77 and parts in prose of sections four and 20 five: NAK MS A 1267/6, NGMPP Mf. A693/11, fols. 16v8-17r1, 25r7. 33 In the LT~ Vajrapåñi claims that his teaching is based on the Lakßåbhidhåna and sometimes he actually quotes from this text (cf., for instance, p. 49).

III. How to Produce, Construct and Legitimate a Tradition

263

Kålacakra doctrines and practices, as if it had actually been a text of the system. In fact, Puñ∂arœka –who would have been the author of a Ma∞ju†rœnåmasaõgœtivimalaprabhå that has come down to us only in Tibetan translation (Q #2114)–34 even goes as far as stating that the knowledge of the MNS depends on the knowledge of the Paramådibuddha, and that the knowledge of the Gnosis Body of the Vajra-holder depends on the knowledge of the MNS. Those who do not know this Body do not know the Mantrayåna and are, thus, separate from the path of the Blessed One.35 The MNS gradually became an influential source of the Kålacakra system. Even though only a few stanzas from this text are quoted in the LT~ and in the HTP~, Puñ∂arœka states in his Paramåkßaraj∞ånasiddhi (PAJS) that the essence of the Blessed One’s teaching is contained in the MNS.36 In the AK and the AKU, respectively a commentary and a sub-commentary on the MNS, the entire outer and inner worlds are said to correspond to the stanzas of the MNS.37 In these two texts, all the doctrines and practices of the school are explained in the light of the MNS. The fact that not one early Kålacakra work comments comprehensively on the GST merits further study. Here we shall limit ourselves to observing that the way in which Kålacakra authors (Anupamarakßita, Ravi†rœj∞åna, Nåropå and so on) quote and gloss parts of the GST and the Samåjottara38 diverges to varying degrees from that of Candrakœrti and other authors like Muni†rœbhadra, who adhere more closely to the GST tradition; for instance, in the commentary of the celebrated GST 2.3, which contains an allusion to the insubstantiality of the meditative practice (bhåvanå), in the explanation of the pratyåhåra limb of the sixfold yoga and in the description of the signs (nimitta) that the yogin sees while practicing the sixfold yoga. 4. It would seem plausible therefore that part of the tradition, that which is mirrored in the most ancient Tibetan historiographies, sought to endow the LT~ and the HTP~ with the same authority as the VP, by claiming that these two works were originally more extensive and complete, and had become incomplete because the second part had been hidden by the ∂åkinœs. The story is related by Bu ston and by mKhas grub rje when they describe the Rwa tradition, and also by ’Gos gz^on nu dpal. They tell us that the Bodhisattva corpus 34 Actually Bu ston questions and mKhas grub rje rejects the attribution of this work to Puñ∂arœka (J. Newman, personal communication). 35 ato ye paramådibuddha∫ na jånanti te nåmasaõgœti∫ na jånanti, ye nåmasaõgœti∫ na jånanti te vajradharaj∞ånakåya∫ na jånanti, ye vajradharaj∞ånakåya∫ na jånanti te mantrayåna∫ na jånanti, ye mantrayåna∫ na jånanti te sa∫såriñaΔ sarve vajrarabhagavato mårgarahitåΔ (VP vol. 1, p. 524-7). Cf. Newman 1987a: 83, 1987b: 93. 36 Cf. VP vol. 3, p. 100 25-26; Gnoli 1997: 78. 37 Cf. AK pp. 2-3 and AKU p. 116. 38 The Samåjottara is actually GST 18. This chapter, however, was probably added to the text and is often quoted as a separate work.

264

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

was revealed to the ancient teacher Cilu from Orissa through an incarnation of Avalokite†vara (or Ma∞ju†rœ according to another tradition). Then Cilu, prompted by the requests of his disciples, wrote down the three works of the corpus that he had memorized. Then –we read in the Deb ther sõon po– the troops of a foreign king invaded the country. They [the disciples] hid these commentaries on the tantras [the VP, the LT~ and the HTP~] in a pit, and escaped. After the war was over, they returned and searched for the [hidden texts]. [They discovered] that the second half of the two shorter commentaries was missing. The disciples again requested [him] to write down [the missing portions, but] he declined, saying that the ∂åkinœs had hidden them, and therefore it was improper to write them down [again].39

The fact that the tables of content at the beginning of the HTP~ and the LT~ list all the chapters of the commented tantras, but the commentaries are interrupted after the first ten and a half stanzas of the CST (composed of 700 stanzas) and after the first 120 stanzas of the HT (composed of 750 stanzas) would appear to confirm that part of the works disappeared as tradition would lead us to believe (even if the ∂åkinœs were probably not responsible). But it should be noted that the two works were interrupted at different times, because, as we have seen, Vajragarbha’s work contains an explicit reference to Vajrapåñi’s text, defined therein as a commentary on the ten and a half stanzas. Furthermore, a reading of the colophons of the HTP~ and the LT~ gives us the impression that the authors wished to comment on a part of the CST and the HT only, namely the part that corresponded to the piñ∂årtha of the text, and decided to interrupt their work, even though this may have been due partially to circumstances beyond their control. As regards the HTP~, the Sanskrit manuscripts that have come down to us leave no room for doubt that Vajragarbha originally glossed only the first 120 stanzas of the HT, since none of them goes beyond the commentary of HT 1.5, and the colophon of the text reads as follows: ‘Yogins should understand the meaning that has been condensed in five chapters with 120 stanzas by means of this commentary. […] This is the end of the ‡rœhevajrapiñ∂årtha™œkå’ (vi∫†atyadhika†ata†lokaiΔ piñ∂årthaΔ pa∞capa™aleßu yogibhir avagantavyo ’nayå ™œkayå Ù […] †rœhevajrapiñ∂årtha™œkå samåpteti).40 We must not forget, however, that the Tibetan translation of Vajragarbha’s commentary covers the complete text of the HT. But this translation actually contains two colophons: one at the end of the commentary on the fifth chapter 39 de nas yul der rgyal po gz^an gyi dmag ’oõs nas rgyud ’grel de dag doõ du sbas te bros so ÙÙ de nas dmag byer te slar log nas bltas pa daõ Ù ’grel pa chuõ õu g∞is kyi smad mi ’dug nas Ù slob ma rnams kyis kyaõ bri bar z^us pa daõ Ù mkha’ ’gro mas sbas pa yin pas dbrir [sic for brir ?] mi ruõ gsuõ nas ma gnaõ õo Ù (Deb ther sõon po, chapter 10, fols. 5b7-6a2 = ed. Chandra, fols. 670-671; cf. Roerich 19762: 762-763; Newman 1987a: 80-81). 40 Kaiser Library, Kathmandu, MS C 128, NGMPP Mf. C14/6, fol. 59v . 4-5

III. How to Produce, Construct and Legitimate a Tradition

265

of the first part of the HT, where the Sanskrit manuscript finishes,41 and another at the end of the work. What is contained between the two colophons is attributed to Vajragarbha and is the continuation of the commentary on the HT (1.6 ff.). Possibly, this part of the text is a later composition, written and translated into Tibetan after the Deb ther sõon po; a few lines after the above-quoted passage, the latter states the following: ‘[…] Also the statement that the last part of the two commentaries of the Dvikalpa [= HT] and the Sa∫vara [= CST] were concealed by the ∂åkinœs [is unreliable], since the present size [of these texts] in Tibetan is the one accepted by the Bodhisattvas’.42 The colophon of the LT~, in its turn, states very clearly that the work actually finishes at that point, namely after the commentary on the first ten and a half stanzas. In his History of Buddhism in India and Tibet Bu ston refers to a commentary by Vajrapåñi on the first part of the Tantra; this might also be a reference to the incompleteness of the LT~ (cf. Obermiller 19992: 225). 5. Another issue that is worthy of consideration in relation to the founding project is the choice of a definite date for its inception, which is a new and interesting occurrence in Indian religious culture. In fact, the Kålacakra was the first system that attempted to place its foundation in a historical context. The LKCT and the VP state that the system was diffused within a specific period of time, starting from the year 403 after the mlecchendravarßa, i.e. the Hijra, which corresponds to 1024/25 (or 1026/27 if we go by Abhayåkaragupta and the Kålacakrånusårigañita). The date is given, in the form of a prophecy, in LKCT 1.27. 41 This colophon (sDe dge #1180, fol. 46a ; Q vol. 53 #2310, fol. 52b ; ed. Ganden, 4-5 6-7 vol. 7 [original vol. 16], fol. 63b1-3) reads: ‘[This work] was composed by the venerable Bodhisattva Mahåsattva Vajragarbha. It was transalted [into Tibetan] by Dåna†œla, Indian teacher, and by the lotsåwa ’Bro seõ dkar †åkya ’od. Then it was revised by Subhüti†rœ†ånti, Indian teacher, and by the lotsåwa Cog gru tiõ õe ’dzin bzaõ po’; rje btsun byaõ chub sems dpa’ sems dpa’ chen po rdo rje s∞iõ pos mdzad pa’o ÙÙ rgya gar gyi mkhan po då na †œ la daõ Ù lo tså ba ’bro seõ dkar †åkya ’od kyis bsgyur ba Ù slad kyi [Q: gyis; Ganden: kyis] rgya gar gyi mkhan po su bhü ti †rœ †ånti daõ Ù lo tså ba cog gru tiõ õe ’dzin bzaõ pos z^us so ÙÙ ÙÙ. The text continues with the following words (sDe dge #1180, fol. 46a5-7; Q vol. 53 #2310, fols. 52b7-53a3; ed. Ganden, vol. 7 [original vol. 16], fol. 63b3-6): yaõ slad kyi [Q and Ganden: kyis] rgya gar gyi mkhan po rgyal po’i sras Ù dpal ’jigs [Q: ’jig] med lha’i z^al sõar lo tså ba s∞el [Q: s∞ol] cor dge sloõ pradz∞å kœrtis yul dbus ’gyur gyi dpes gtan la phab pa Ù slar yaõ dpal ldan †oõ ston rdo rje rgyal mtshan gyis legs par b†ad pa la sogs pa’i mthu las Ù brda sprod pa’i tshul rig pa’i dpaõ lo tså ba dpal ldan blo gros brtan pas Ù byaõ chub sems dpa’i ’grel ba skor [Q and Ganden: ’grel pa bskor] gsum gyi tshul la †in tu dad ciõ blo’i snaõ ba rgyas pa’i [Q and Ganden add: dge ba’i] b†es g∞en ra luõ pa chos grags dpal bzaõ pos Ù slob dpon chen po z^i ba ’tsho’i z^abs dpon slob kyis mdzad pa’i Ù dbu ma’i gz^uõ lugs chen po de kho na ∞id bsdus pa rtsa ’grel gyi glegs bam bris te yon du gnaõ nas yaõ daõ yaõ du bskul ba’i õor legs par bcos te bsgyur ciõ z^us nas gtan la [Q: las; Ganden: nas] phab pa’i yi ge pa ni mdzad [Ganden: ’dzad] ston kun dga’ rgyal mtshan z^es bya’o ÙÙ ÙÙ. 42 […] brtag g∞is daõ bde mchog gi ’grel pa g∞is kyi mjug mkha’ ’gro mas sbas par smra ba de la yaõ da ltar bod du ’gyur ba’i tshad de kho na byaõ chub sems dpas rjes su gnaõ õo (Deb ther sõon po, chapter 10, fol. 6b5-6 = ed. Chandra, fol. 672; cf. Roerich 19762: 764-765).

266

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

An indepth analysis of the reasons for choosing a date goes beyond the scope of this paper. While there may be some doubt as to the reliability of certain statements made by Kålacakra teachers, we can be sure that 1024/25 (or 1026/27) corresponds to the date that these masters intended to fix for the historical foundation of the system,43 as clearly demonstrated by J.R. Newman who has devoted an important study to this issue (1998b). There seems no good reason, however, to assume that they started producing their works and spreading their teachings precisely on that date. In fact, the system must have already been fairly widespread when Nåropå was writing (and we know that he died around 1040 CE), since he quotes widely from Vajrapåñi, Vajragarbha and Puñ∂arœka, and also from other authors, such as Anupamarakßita, who, in his turn, cites the LT~, the HTP~, the LKCT, the VP and the Tantrottara. It should also be noted that both the LKCT and the VP were written in at least two stages. Stanzas from the LKCT are quoted in the PAJS by Puñ∂arœka, but these are numbered differently from, and in some cases are not present in, the current version of the LKCT, which therefore must have been revised at some point. The PAJS, now a section of the VP (ad 5.127), probably existed as a separate and complete didactic text when the first version of the LKCT was written. This is confirmed by its specific stylistic features and by the fact that Puñ∂arœka refers to it in the opening chapters of the VP (cf., e.g., vol. 1, pp. 159, 161) and that it is the only section of the VP that contains quotations from the LKCT.44 On the basis of the foregoing, we can say that the early Kålacakra literature was produced in at least five stages: 1) stanzas of the Ådibuddha (which included the SU), LT~, stanzas of the PLH, HTP~; 2) further stanzas of the Ådibuddha, LKCT (first version), PAJS, LKCT (revised version), VP, Tantrottara, Paramårthasevå;45 3) the two ¯a∂aõgayoga treatises by Anupamarakßita; 4) SU~, Sekodde†a™ippañœ by Sådhuputra ‡rœdharånanda; 5) Sekodde†apa∞jikå,46 Guñabharañœ, AK, AKU. We cannot categorically exclude that, historically speaking, the system was founded and began to spread in the years 1024/25 (or 1026/27), as the tradition maintains. However, it is unlikely –although not impossible– that such a complex system was created in a limited period of time, and such a vast literature produced in only fifteen years or so, as Raniero Gnoli has already pointed out (1994: 62). Indeed, if we assume for the sake of argument that the 43 It is well-known that the mythic history of the system dates back to the time of the historical Buddha. 44 For more information, see Gnoli 1997: 3-4; Cicuzza, Sferra 1997: 115-118. 45 We do not know exactly when the Paramårthasevå was written, but it might have been composed after the VP, which does not make any reference to it. 46 This anonymous work is little more than a summary of the SU~.

III. How to Produce, Construct and Legitimate a Tradition

267

account given in the LKCT-VP is true, the first four of the five early stages of Kålacakra literature would have to have been produced in about 16 years (from 1024/25 to 1040).47 Probably the first two phases began before 1024/25, perhaps as early as the end of the X cent. The terminus post quem for the second stage could coincide with the first incursions the North-West India of . Sebktegin and expecially of his son MaΔmüd of Gaznœ (r. 998-1030) around the end of the X cent. and the beginning of the XI cent.,48 since significant reference to mlecchas, i.e. barbarians, is made in the works of Ya†as and Puñ∂arœka (see Newman 1998a). The fact that no mention of Islam is made in the LT~ and the HTP~ may well be in keeping with the Indian spirit, but we cannot exclude that Islam was either not present or not as strong when these texts were written. More generally, it was possibly irrelevant to the exegesis of the CST and the HT. What we can be certain of, independently of the date of the system’s founding and diffusion, is that it was ‘constructed’ gradually. We may suppose that the early authors of the Wheel of Time first began work on their texts and then chose a significant date on which to officially unveil their system; a date that coincided with the beginning of a sexagenary cycle and had a symbolic importance. This is yet another example of how these authors participated in the complex process of creating and consolidating their tradition. 6. Part of the strategy adopted with regard to the founding project of the Kålacakra was to refer to a ‘higher level’ of the tradition, namely that of the above-mentioned Paramådibuddha, the roottantra of the system. The three Bodhisattvas, along with other famous authors such as Anupamarakßita, Sådhuputra ‡rœdharånanda, Nåropå, Ravi†rœj∞åna and Vibhüticandra, quoted stanzas from this text, which is held to be the original scripture of the system and which according to the tradition consisted of 12,000 verses written down by Sucandra, emanation of the tenth-stage Bodhisattva Vajrapåñi and the first of the seven kings of the celebrated land of Sambhala.49 Although, as we shall see, the Kålacakra authors’ use of the roottantras is original in many respects, the underlying idea that the 47 According to J. Newman (1998b), the year 403 proves that the passages of the LKCT and the VP containing the year 403 were composed during or after that year, but the texts as a whole may have been composed prior to that date, and it is also possible that the LKCT and the VP were completed between 1025 ca. and 1040 CE. 48 Sebktegin and his son arrived at Pe†åvara in 987 after Råjå Jayapåla’s defeat and suicide (cf. Bernardini 2003: 78). For further information on the plundering –Bhœmanagara (Nagarkot) (1008), Thåne†vara (1011), Lahore (1014), Mathurå (1018), Kannauj (1019), . Somanåtha (1025)– carried out by MaΔmüd of Gaznœ, see Bernardini 2003: 82-85, and in particular Wink 2002: 120-135, 328-333, and Haig 1987. On the Muslim colonies in India before the raids undertaken by the Ghaznavids, see Orofino 1997: 723 and notes. 49 See Newman 1991: 51-90.

268

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

tantras that have come down to us were the abridged version of roottantras in which the spiritual teaching would have been explained in more depth is neither exclusively Buddhist nor belongs to the Kålacakra. There is a similar instance in Hindu tantric literature. The †aiva Målinœvijayottara, for example, that has come down to us in twenty-three chapters is said to be a summary of the Målinœvijaya. According to tradition, the latter, which probably never existed, amounted to thirty million stanzas.50 A similar consideration may be made with respect to other tantras, such as the Manthånabhairava, the original version of which was said to contain a billion verses (lakßako™i),51 and the Sarvavœratantra, the basic text (mülasütra) of which was said to be three hundred and fifty (chapters ?) long.52 The same applies to vaißñava texts, such as the Jayåkhyasa∫hitå (cf. 1.7079) and the Pådmasa∫hitå. The latter, in particular, is held to be the condensed teaching in 10,000 stanzas of a previous teaching in 100,000 stanzas, which, in its turn, is believed to derive from a text of 500,000 stanzas. The original work is said to have contained 15,000,000 stanzas (J∞ånapåda 1.32-33). In Buddhist texts, references to mülatantras or br¢hattantras (lit. ‘long tantras’) are to be found, for example, in the Yoginœsa∞cåratantra (pp. 5, 9, 12, 52, 55, 69, 70) and in the CST (pp. 3, 19). We know that the root-tantra of the GST should have been a work of twenty-five thousand verses, that of the Måyåjålatantra a work of sixteen thousand verses, and that of the CST a work of one hundred thousand chapters, entitled Mahålakßåbhidhåna.53 These works have never actually been found,54 that is to say they have not come down to us directly but through quotations and, in one case (= SU), in a fragment; therefore, continual references to these texts have aroused interest and sometimes perplexity among scholars.55 By general consensus both in Hindu and Buddhist circles, the roottantras were not only longer (and thus, in theory, more difficult to memorize), but also more complex and complete, and contained broader and more cryptic teachings. Usually mülatantras were only mentioned –in theistic enviroments they were used to allude to the profundity and uncontainability of divine revelation, which had to be gradually simplified to be understood by man–; but Buddhist teachers –and this is quite interesting–, especially in the Kålacakra tradition, frequently quoted directly from them. The use these authors/teachers made of the mülatantras reveals their true (and twofold) intent: to give 50

See Målinœvijayottara 1.8b ff.; cf. also Mataõgapårame†varågama, Vidyåpåda 1.30cd-

33ab. 51

See Dyczkowski 1988: 99. See Dyczkowski 1988: 110-111. 53 See Tsuda 1974: 33. 54 See Newman 1987a: 93-102. 55 Cf. Nihom 1984: 21-22; Newman 1987b. On the root-tantras, see also Reigle 1986 and Snellgrove 1959, vol. I: 15-17. 52

III. How to Produce, Construct and Legitimate a Tradition

269

authority to the new texts and to elevate their promulgators and interpreters, i.e. themselves. Indeed, the mülatantras were clearly accessible only to the few, although this was not stated specifically. It is quite possible that the root-tantras may never have existed as such, and that the Ådibuddha, in particular, consisted only of fragments and isolated stanzas. It is also possible that these stanzas and the verses attributed to other root-tantras, such as the PLH, were written by the commentators themselves in order to bolster their own works. This can be ascertained by examining one or two examples. In his HTP~ Vajragarbha quotes literally hundreds of stanzas from the PLH, which, as we have already mentioned, he claims was the basic tantra (mülatantra or åditantra) of the HT. He writes: The Blessed one, viz., Vajrasattva, taught the short tantra, which comprises seven hundred and fifty stanzas and consists of twenty-two chapters and two brief parts: Sambodhi and Måyåjåla. It is taken from the Pa∞calakßahevajra, which is its root-tantra of thirty-two extensive parts.56

However, the PLH was undoubtedly written after the HT. The eclectic nature of the former and its treatment of themes that presuppose the development of speculations subsequent to the HT reveal its later origin. R. Gnoli has proposed that root-tantras, such as the Paramådibuddha and the PLH, never existed as independent works and that the latter, in particular, may have been written by Vajragarbha himself to give more importance to his own commentary (1994: 60, 66). David Snellgrove has put forward a similar hypothesis.57 Concerning the above, we should bear four points in mind. 6.1 First of all, this work –as far as we can say at present– is only known through verses that, for the most part, can be traced to the HTP~ that, therefore, may be the source. Apart from the above-mentioned passage of the VP that does not cite the PLH, but contains stanzas that Vajragarbha attributes to it (§ 2.3), we have found verses that according to the HTP~ belonged to this text in works such as the AK,58 the AKU and the Dohåko†a™œkå by Amr¢tavajra, which although they quote Vajragarbha and/or his commentary, do not make explic56 pa∞calakßahevajrån mülatantråd dvåtri∫†anmahåkalpål laghukalpadvaya∫ sambodhimåyåjålalakßaña∫ dvåvi∫†atiparicchedåtmaka∫ sårddhasapta†atagranthapramåña∫ laghutantra∫ vajrasattvena bhagavatå sande†itam (2nd pariccheda, NAK MS A 1267/6, NGMPP Mf. A693/11, fol. 6v4-6; IsIAO, Rome, MS 1.20, 22/S, MT049, fol. 4v1-2 [see above, note 27]). On this concept see also HTP~ 1.3d-6 and 1.67ab (cf. below, § 6.4 and note 73). 57 ‘As this mülatantra in common with other works of exegesis concentrates on the figurative sense, it is probably the work of some recognized master, and not impossibly of that writer himself who goes by the name of Vajragarbha. […] I remain persuaded that this particular “basic text” is in any case later than the tantra itself and the early commentators, Saroruha, Kåñha, Bhadrapada, and Dharmakœrti and unknown to ~aõkadåsa and Ratnåkara†ånti’ (Snellgrove 1959, vol. I: 17-18). 58 In the AK (ad MNS 9.10; ed. pp. 82-83) some stanzas of the PLH (= HTP~ 5.21cd35ab) are quoted without mentioning the source.

270

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

it mention of the PLH.59 Other quotations from this text can be found in works that were composed in a later phase of the spreading of the Wheel of Time, such as the Sekodde†a™ippañœ by Sådhuputra ‡rœdharånanda, who mentions the PLH but not Vajragarbha and the HTP~,60 the SU~ by Nåropå, who, as far as we know, is the only author who cites both Vajragarbha and the PLH,61 and the rDo rje’i tshig gi s∞iõ po bsdus pa’i dka’ ’grel by s˚an grags bzaõ po.62 However, we must bear in mind that we also find quotations from the PLH that do not occur in the HTP~, such as the seven stanzas from the ’Bum phrag lõa pa’i kye’i rdo rje (= PLH) that are to be found in the *Sahajasiddhipaddhati by Lakßmœ Bha™™årikå, a commentary on Lakßmœõkarå’s Sahajasiddhi,63 and that stanzas that Vajragarbha states as deriving from the PLH are sometimes attributed to other sources. Particularly interesting in this regard is a long quotation in the Trivajraratnåvalœmålikå, an unpublished commentary on the HT. The author of this work, Kelikuli†a, quotes, with a few variant readings and in a slightly different order, several stanzas that are to be found in the fifth section of the HTP~. According to Vajragarbha these stanzas are drawn from the PLH, but Kelikuli†a refers to the source text of these verses as Låkßikadvikalparåjatantra and he even specifies the name of the chapter (cañ∂ålikåyogapa™ala; elsewhere he calls it cañdålœyogapa™ala).64 The reasons for this difference in the name of what we would expect to be the same source should be investigated more thoroughly. At present we are not able to establish the relationship between Vajragarbha and Kelikuli†a. An hypothesis worthy of note was put to me by Harunaga Isaacson, who suspects 59 In his commentary on the Dohåko†a by Kr¢ßñavajrapåda, Amr¢tavajra (alias Amitavajra, dPag med rdo rje; cf. Chimpa, Chattopadhyaya 1970: 304-305) quotes (pp. 131132) some stanzas of the PLH (= HTP~ 5.77-86ab, 90-94ab). His text, in its turn, is incorporated almost in its entirety and with only a few changes in an apparently late, still unpublished tantric composition generically entiteled Kalparåja that includes many quotations and is divided into 13 pa™alas. The ™œkå by Amr¢tavajra corresponds to pa™alas 6-8 of the Kalparåja (fols. 27r8-41v1 = pp. 130-155). Stanzas 5.77 and 5.80 are not quoted (cf. fols. 28r7-28v7) in the Kalparåja. Amr¢tavajra knew Vajragarbha and was familiar with the HTP~: a passage from the 4th pariccheda of the HTP~ is quoted with slight variations on p. 138 of his commentary (tathå ca vajragarbhapådåΔ – ‘nåsådvayarandhre […] mañ∂alacakram’). Amr¢tavajra quotes numerous stanzas of the SU and some stanzas of the LKCT; a passage from his work recurs verbatim in the SU~ (cf. ed. Carelli, p. 38). 60 In the Sekodde†a™ippañœ, Sådhuputra ‡rœdharånanda (ad SU 152-153ab; p. 142) introduces with the words: tathå ca pa∞calakßåbhidhånahevajre a stanza that corresponds to HTP~ 1.40cd-41ab. The same stanza is quoted without mentioning the source by Ravi†rœj∞åna ad MNS 10.3 (p. 90) and by Vibhüticandra in his AKU (ad AK 4.1; p. 134). 61 Cf. ed. Carelli, pp. 59-60 and p. 67, where Nåropå quotes 11 stanzas from the PLH (= HTP~ 10.4-14) introducing them with the words: yathå pa∞calakßåbhidhåne. 62 See above, note 20. 63 sDe dge #2261, fol. 14b, Q #3108, fol. 17a; see also Shendge 1967: 128, note 5. 64 The opening sentence runs as follows: tathå cokta∫ låkßikadvikalparåjatantråkr¢ß™acañ∂ålikåyogapa™ale (fol. 26r1-2). The correspondences are: HTP~ 5.2-4 (Trivajraratnåvalœmålikå, fol. 26v4-5), 5.5-17 (fols. 26r2-26v4), 5.18-26 (fols. 26v5-27v1), 5.27cd (fol. 27v1-2), 5.28cd-53 (fols. 27v2-29r1).

III. How to Produce, Construct and Legitimate a Tradition

271

that in Kelikuli†a’s time ‘there was circulating a work that called itself the Cañ∂ålœyogapa™ala of the Låkßikadvikalparåjatantra (something like the SU), which had included material attributed by Vajragarbha to the PLH’. Isaacson is aware that this must be confirmed, if confirmation is ever possible. He also pointed out to me that Kelikuli†a was familiar with the PLH and refers to it in other parts of his work as Pa∞calåkßika. 6.2 The eclectic nature of the text. Sometimes the PLH verses contain stanzas that can be found, in an identical or slightly different form, not only in the Paramådibuddha, the Piñ∂œkramasådhana attributed to Någårjuna, the CST and in other celebrated Buddhist texts (e.g., HTP~ 1.45ab = Madhyamakakårikå, 13.8ab), but also in Hindu works. For instance, Andrea Perrone has pointed out to me that some stanzas of the PLH dealing with alchemy correspond exactly to parts of the Rasahr¢dayatantra, the Rasårñava and the Ånandakanda.65 It is possible that the author of the PLH took these verses either from the abovementioned alchemical texts themselves but, more likely, from an earlier source on which these texts themselves drew. They were probably composed between the IX and the XIII cent. This would also explain why another verse of the PLH (= HTP~ 9.34) dealing with the alchemical process (but not present in the above-mentioned works) also appears in the SU (st. 134) and in the †aiva Kubjikåmatatantra (3.104). 6.3 Stanzas quoted by Vajragarbha from the PLH are evidently connected with Kålacakra themes. In a brief paper, published in DhœΔ (Cicuzza, Sferra 1997), we show the close relationship existing between the description of the arrangement of the main nå∂œs in the human body, as explained in the PLH and the Kålacakra doctrines (cf. SU 46 ff.). According to the latter, the position of the three nå∂œs changes at the navel (nåbhicakra). The rasanå, avadhütœ and lalanå, which in the upper part of the body are located on the right, in the centre and on the left, are turned towards the left, right and centre in the lower part of the body.66 We also show that differences between the HT and the PLH are sometimes evident. For example, when Vajragarbha describes the number of petals (dala) of the lotuses located in the cakras along the middle channel, he does not strictly follow the HT text (see, for instance, 1.1.23)67 but the LKCT (cf. 2.57-59), the physiological concepts of which differ slightly from those of the HT. Further similar examples can be given. For instance, in describing the different rosaries that have to be used during the rites for paralyzing, subduing and so forth, the PLH substantially agrees with the 65 For instance HTP~ 9.30 = Rasahr¢dayatantra 6.14, Ånandakanda 1.38, Rasårñava 10.17. This stanza is also quoted in the PAJS (VP vol. 3, p. 93). 66 This is illustrated with figures in Gnoli, Orofino 1994: 272, note 2; Orofino 1996: 133; Cicuzza 2001: 27. 67 Cf. also J∞ånodayatantra, p. 5.

272

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

description that we find in the PAJS, while it differs from that in the HT. The combinations of the rosaries and ritual actions are listed in the following table. HT 2.10.2-3 spha™ika raktacandana riß™ikå nira∫†uka a†vaha∂∂a brahmahåsthin gajåsthika mahißåsthi

stambhana va†ya abhicåruka vidveßa

PAJS (VP vol. 3, p. 65)

spha™ika †åntika muktåphala pauß™ika naradanta måraña uß™radanta uccå™ana [kharadanta] uccå™ana putrajœva va†ya åkarßaña padmabœja åkr¢ß™i [raktacandana] varßåpaña rudråkßa stambhana måraña riß™a mohana

PLH (= HTP~ 6.59cd-63) spha™ika mauktika nr¢danta kharadanta putrajœvaka padmabœja [candanabœja] rudråkßa riß™a

†åntika pauß™ika måraña vidveßaña [uccå™aña] va†ya åkr¢ß™i stambhana mohana

Furthermore, we should remember that the PLH deals with a certain number of themes, such as the death-sign (ariß™a) and the meditation on the Krodharåjas, which are not treated in the HT but assume considerable importance in Kålacakra texts, such as the SU, the LT~, the LKCT and also the VP.68 In the PLH, the phonemes of the Sanskrit alphabet are divided on the basis of the elements (ether, wind, etc.) and arranged according to the Kålacakra doctrine: the dentals are placed after the labials and the position of the semi-vowels LA and VA is reversed.69 Vajragarbha aligns the teachings of the HT with the doctrines of the Kålacakra by quoting the PLH, which is not in fact based on (or the basis of) the HT but inspired by those doctrines. 6.4 The use Vajragarbha makes of the PLH in his work. He actually quotes frequently and at length from it in order to comment on verses of the short tantra that he is explaining. Such usage seems to contradict Vajrapåñi’s assertion in the LT~ that the alpatantras (lit. ‘short tantras’) –even though full of difficult words (vajrapada)– were drawn from the mülatantras in order to meet man’s now limited capacity to understand (tat kasya hetor bhagavann alpatantra∫ mülatantråd de†ayasi iti Ù bhagavån åha Ù iha pa∞cakaßåyakåle jåmbüdvœpakå manußyå vi†eßeñåryavißaye alpåyußo ’lpapraj∞å bhavißyanti […] tasmåt sarvamülatantråd alpatantrade†anå buddhasya).70 As the introduction of the HTP~ shows, Vajragarbha shared with Vajrapåñi and Puñ∂arœka71 a distrust in man’s inability to understand the short tantras, but, more openly than the other two, assigned a precise explanatory role to the 68

Cf., for instance, VP vol. 2, pp. 33-34, 67. The sequence of semivowels: YA, RA, VA, LA (wind, fire, water, earth) is also present in the Cañ∂amahåroßañatantra (p. 58). 70 LT~ pp. 51 12-14, 523. 71 Cf. Newman 1987a: 313-314. 69

III. How to Produce, Construct and Legitimate a Tradition

273

mülatantras, and to the PLH in particular. This is not an obscure and incomprehensible text, but an explicative work (nirde†a) that Vajragarbha actually uses to illustrate the short HT and especially its most difficult parts.72 In the opening stanzas of the HTP~ he writes:73 Impelled by Hevajra,74 I –the glorious Vajragarbha, the Lord of the ten earths, who desire the welfare of all beings– am writing a commentary that explains the secrets of the [Hevajra]tantra, in order for the yogins to attain the [true] path. Since this short tantra of 750 [stanzas], endowed with many adamantine words, has been taken from a great tantra of 500,000 [stanzas], this [commentary], called ¯a™såhasrikå, follows the root-tantra in order to elucidate on the [short] tantra. […] The teaching of the Victorious One, which is [condensed] here, in this short tantra, the Hevajra, and which in ancient times the Buddhas taught with 500,000 stanzas in the collection of the Åditantra remains obscure to man. […] [Therefore] the meaning [of the short tantras only] becomes clear from the commentaries. He who tries to reveal, without a commentary, the obscure [meaning of a] word in a short tantra, resembles a blind man who attempts to follow the trail of a snake that has long disappeared into the water.75

Similar observations can be made about the Ådibuddha. A deep analysis of the quotations from this text would require a separate paper. Here we would limit ourselves to pointing out that numerous stanzas attributed to the Ådibuddha are found in other tantras and independent works (cf. Gnoli 1994: 60-61). 7. Essential element of the founding project, indeed one of the most important, is the description of the qualifications necessary for those whose privilege and duty it is to preserve, transmit and interpret the Scriptures and spiritual teachings. This is evident, among other things, from the space the texts devote to this topic. 7.1 It is worth noting that, as is the case with other tantric cycles, the first Kålacakra authors felt the need to define the figure of the true master (sadguru) or adamantine teacher (vajråcårya) and, above 72 The threefold division udde†a, nirde†a and pratinirde†a, which we find elsewhere (e.g. in the Yoginœsa∞cåratantra 1.2-3 and in its commentaries; pp. 37-8, 123), is said to apply to Kålacakra literature in the SU 3-5 and to the Hevajra cycle literature in the HTP~ where it is stated that the HT is the udde†a, the PLH the nirde†a and, implicitly, the HTP~ itself the pratinirde†a (cf. st. 1.77). 73 […] tantraguhyagadikå ™œkå mayå likhyate ÙÙ [3] †rœmatå vajragarbheña sarvasattvahitaißinå Ù da†abhümœ†vareñeya∫ mårgalåbhåya yoginåm ÙÙ [4] pa∞calakßamahåtantråd alpatantre samuddhr¢te Ù sårddhasapta†ate ’py asmin bahuvajrapadånvite ÙÙ [5] hevajracoditenaiva yå ßa™såhasrikå matå Ù seya∫ tantraprakå†årtha∫ mülatantrånusåriñœ ÙÙ [6] […] hevajre jinade†anåtra laghuke ’visphu™eya∫ nr¢nå∫ lakßaiΔ pa∞cabhir åditantranicaye buddhaiΔ kr¢tå yå purå Ù […] ™œkåbhir arthågamaΔ ÙÙ [67] yaß ™œkårahito ’lpatantranicaye gupta∫ pada∫ de†ayet so ’mbuny akßivivarjita† ciragatasyåheΔ pada∫ vœkßayet Ù […] [68]. 74 Cf. also st. 1.65cd. 75 This example occurs also in the Ålokamålå by Kambala (st. 280): jalaprayåtåhipadåni pa†yataΔ […] katha∫ nu lokasya na jåyate trapå ÙÙ (pp. 109-220). Another stanza of this text (st. 142) is paraphrased by Vajrapåñi (LT~ p. 143).

274

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

all, to deal with this issue at crucial points in their works, naturally giving more space to it than the masters who followed them. In fact, Vajragarbha, rather than commenting on the HT immediately, devotes the first of the ten paricchedas into which his work is divided to establishing the criteria for interpreting the Scriptures, and the qualifications necessary for imparting and glossing the teachings. Vajragarbha’s arguments are treated also at the beginning of the third section of the VP, where we find parallel passages; Puñ∂arœka deals briefly with these arguments also in other parts of the VP, but devotes the entire first part of the Paramårthasevå to them. Vajrapåñi presents the same arguments as Vajragarbha and Puñ∂arœka, but in a less systematic and detailed manner. What is characteristic of the early Kålacakra teachers is that they do not limit themselves to describing the qualities and defects of the guru; instead they refer to (or ‘stage’) a polemical debate between themselves and hypothetical exponents with a conflicting viewpoint. This is not a purely scholarly debate, and quite probably the texts of the three Bodhisattvas mirrored a debate that actually existed in Buddhist circles of the period and that must also have treated the correct interpretation of some scriptural passages on the qualifications and duties of the guru. To this end, the three Bodhisattvas cite and interpret some verses and words of the Gurupa∞cå†ikå from a standpoint asserting the superiority of ordained monks. Therefore, it is worth following the reasoning of Vajragarbha and Puñ∂arœka in more detail.76 In LKCT 3.2-3, as in other sources dealing with the vajråcåryaparœkßå, the examination of the adamantine teacher, which is a preliminary to the gurvårådhana, a list is given of the qualities that a guru must possess and of the defects that he must not possess. The wise disciple (budha) must carefully examine both his qualities and defects, and determine if he is worthy of the role he wishes to assume. It is fundamentally a question of moral qualities that he must have to be worthy of possessing the right teaching and of imparting it. Thus far there is not much difference between the LKCT-VP and what we read in the Guhyasiddhi and in other works.77 In LKCT 3.2-3 no mention is made of his cultural skills, whereas the VP does refer to them (vol. 2, p. 5), also by quoting stanzas 8 and 9 of the Gurupa∞cå†ikå; these concern the ritual (da†atattva) and the Scriptures (the master is defined as †åstrakovida). The same stanzas are quoted in the HTP~ (1.29-30). The VP raises a rather interesting objection, which is also present in the HTP~ where the vis polemica is more evident. The objector (clearly another tantric Buddhist, and possibly an exponent of the socalled årya-school) states that the Tathågata himself said that only the qualities of a master must be considered and –as specified in the 76 77

See also Wallace 2001: 8-9. For further references, see Sferra 2004.

III. How to Produce, Construct and Legitimate a Tradition

275

VP (the passage, however, seems to be partly corrupted)–78 that since in the future this point of view may be thought to be held by foolish people, that the defects of a guru must not be taken into consideration. In this regard the opponent in the debate is presented by Vajragarbha (1.14, 1.19) and by Puñ∂arœka as quoting two stanzas: the first corresponds to a stanza of the ‡rœmåyåjålamahåtantraråja quoted also by Candrakœrti in his ¯a™ko™ivyåkhyå on the GST (p. 216); the second is Gurupa∞cå†ikå 2, which is quoted to stress that the fully initiated vajråcårya is worshipped by the Tathågatas themselves (cf. GST 17, pp. 104-105; J∞ånasiddhi, p. 151) and probably to infer that no ordinary man has the right to criticize him. Vajragarbha’s and Puñ∂arœka’s reply shows that their principal concern was on the one hand to preserve, like the opponents, the guru’s high standing, but, more importantly, on the other to limit access to this elevated position, to prevent it from being ‘diminished’. The opponent’s unilateral interpretation might have finished by weakening the guru’s stature and influence, with predictable and damaging consequences. This is why, according to the Kålacakra masters, it was important to consider also the defects of the guru, since it enabled them to create a hierarchy and to ensure that authority was invested only in those who were worthy of wielding it. Both Vajragarbha and Puñ∂arœka reply to the opponent, in the first place by quoting stanza 7 of the Gurupa∞cå†ikå, which states that the disciple must not accept as a master a morally immature person, in other words, a person who is without compassion, pitiless, etc. Vajragarbha briefly glosses this, associating a lack of compassion with the secular activities of the farmer, and pitilessness with hatred for monks. In the second place, both establish a guru hierarchy and identify the highest level with the guru who is a monk (bhikßu). Both commentaries betray a certain aggressiveness, especially in their ironical portrayal of laymen and their presuming to be Vajra-holders (vajradhara). The VP and the HTP~ quote a similar stanza –the former attributing it to the Ådibuddha, the latter to the PLH– that we find with slight changes in the Sa∫varodayatantra (8.9). This stanza states that laymen, farmers, merchants and so on sell the Good Law, giving 78 The following is the reading of the editio princeps: tasmåd ‘åcåryasya guñå gråhyåΔ’, ihånågate ’dhvani yad vaktavya∫ bålajanaiΔ sanmårganaß™air ‘åcåryasya guñå gråhyå’ iti keßå∞cid mårganaß™ånå∫ vacana∫ bhavißyati, tasmåd ucyate ‘doßå naiva kadåcane’ti Ù tan na […] (vol. 2, p. 49-11); the same passage appears with different wording in a MS of the VP kept in microfilm form in the Library of IsIAO in Rome: tasmåd ‘åcåryasya guñå gråhyå’ iti keßå∞cin mårganaß™ånå∫ vacana∫ bhavißyati Ù tasmåd ucyate Ù ihånågate ’dhvani yad vaktavya∫ bålajanaiΔ sanmårganaß™air ‘åcåryasya guñå gråhyå doßå naiva kadåcane’ti Ù tan na […] (Mf. 2.3, AAC, chapter 3, fol. 4r3-4; for a description of this MS see Sferra 1995: 360); the latter reading is also that of Bu ston’s commentary on the VP (the glosses are placed between brackets): de’i phyir slob dpon gyi yon tan blaõ bar bya’o z^es pa (yaõ dag pa’i Ù) lam (las) ∞ams pa (log par ’tsho ba Ù) ’ga’ z^ig gi tshig ’byuõ bar ’gyur te Ù de’i phyir brjod par bya ste Ù ’dir ma ’oõs pa’i dus su lam ∞ams pa’i byis pa’i skye bo rnams kyis Ù slob dpon yon tan blaõ bya ste Ù skyon ni nam yaõ (brjod par bya baÙ) min pa ∞id ÙÙ ces pa gaõ brjod par bya ba de ni ma yin te Ù […] (dBaõ gi le’u’i ’grel mchan, ed. Chandra, fol. 2394-5).

276

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

us to understand that they are not truly generous; and we know from VP ad LKCT 3.4 that the disciple must avoid covetous (dhanårthin) lay masters (vol. 2, p. 718-20). Both texts quote from common sources and continue by stating that of the three types of guru (bhikßu, navaka [or cellaka], gr¢hastha) the first is by far the best, as stated also in the Kriyåsamuccaya by Jagaddarpaña (alias Darpañåcårya) (fol. 2r), who quotes from Kålacakra sources at the beginning of his work.79 The LT~ also speaks of the bhikßu’s being superior to the cellaka and the gr¢hastha, when it comes to choosing the ‘leader of the group’, the gañanåyaka (pp. 105-109). Gurupa∞cå†ikå 4a and 5d are quoted therein (p. 106), but not acknowledged, and glossed in the prose. Hence, the basic requirements for a guru are moral integrity and a profound knowledge of rituals and Scriptures, and this applies both to Kålacakra teachers and others. But this is not the real point, since, in theory, these are qualities that anyone, even laymen, can display. It is interesting to note that Vajragarbha uses the future tense in staging the above debate, to make it sound like a prophecy. In stanzas 1.10-11 we read: ‘We are the Vajra-holders since we have attained Buddhahood, the state of Vajrasattva, with effort, through the initiations’, some will say to other men. ‘All monks who observe moral precepts should not be honoured. We who wear white clothes, and who are the Vajra-holders in person, must be honoured!’.80

The crucial point is that the three Bodhisattvas establish an indisputable, objective criterion, namely that only monks, clothed in a monk’s habit, who observe the monastic precepts, can truly possess integrity and thus command respect. When fully initiated, they are the supreme vajråcåryas, i.e. Vajra-holders.81 If a bhikßu is present the layman must not perform the rites, such as the one connected with founding a monastery (vihåra), etc. Laymen must not be worshipped as the Vinaya of the Mülasarvåstivådin also reminds us;82 concerning this, stanzas 4 and 5 of the Gurupa∞cå†ikå are quoted and glossed in the VP and the HTP~. At this point, a further, particularly significant qualification of the guru is introduced in the two texts: he must have attained the 79 As pointed out by V.A. Wallace (2001: 218, note 19) the concept recurs also in the Vajramålåguhyasamåjavyåkhyåtantra. 80 buddhatva∫ vajrasattvatva∫ sekaiΔ sa∫gr¢hya yatnataΔ Ù vaya∫ vajradharåΔ kecid vadißyanti narå nr¢ñåm ÙÙ avandyå bhikßavaΔ sarve †œlasa∫varadhåriñaΔ Ù sitavastrå vaya∫ vandyåΔ svaya∫ vajradharå bhuvi ÙÙ. It is worth noting the use and the order of the words in the Sanskrit text of stanza 10, particularly the fact that it ends with narå nr¢ñåm (lit. ‘men among men’) to create a ‘surprise-effect’ and to stress that these are the words of mere men talking to their equals. 81 In this part of the VP we sometimes find the words bhikßuvajradhara (compounded or not compounded); cf. e.g. vol. 2, p. 616 and p. 71. 82 Cf. ‡ayanåsanavastu, pp. 3-5, and also VP vol. 1, p. 54 14-16.

III. How to Produce, Construct and Legitimate a Tradition

277

bhümis, otherwise the three jewels (Buddha, Dharma and Saõgha) will be weakened; the guru is a liõgin, one who bears signs of being a follower of a religious path, which here probably means a monk wearing monk’s robes, etc.83 The aforesaid, and especially the mode in which the authors expressed themselves through a particular use of words and arrangment of arguments, reveals how urgent it was to find a way of resolving a socio-cultural situation in which the Buddhist (and not only the Buddhist) culture was held by the authors to be in decline and/or seriously compromised. It was only at that point that the qualities (guña) of the guru were examined, namely after establishing the supremacy of the monks (and in particular the monks who had attained a very high level of spiritual development) over laymen, by quoting stanzas 8 and 9 of the Gurupa∞cå†ikå and the gloss on stanza 2 of the same text, which was previously used by the opponent to back up his argument. Both the HTP~ (1.20-23) and the VP give a twofold interpretation of the stanza, that is, its deep meaning (nœtårtha) and its surface meaning (neyårtha). 7.2 While it is important to establish who are the true masters to guide the Buddhist community and celebrate rituals, it is even more important to identify who is fully qualified to compose and interpret the texts. The three Bodhisattvas are perfectly clear on this as well: only Bodhisattvas can write the commentaries. This statement first appears in the LT~ (and is repeated in the HTP~ and the VP), where we read that in this period of decline (kaßåyakåla) the gr¢hasthas will write short and long commentaries (p. 51), but the wise men will follow only the commentaries written by the Bodhisattvas (p. 52).84 SU 5cd states that these works must be composed by individuals endowed with supernatural powers (abhij∞å) and not merely by learned men (naiva pañ∂itaiΔ).85 Vajragarbha says the same thing, not at all modestly, in his HTP~: ™œkå mayå likhyate […] keci™ ™œkå∫ karißyanti pa∞cåbhij∞ådibhir vinå (stt. 1.3d, 9ab). The syllogism is not perfect, but it is evident what the author wishes to convey. Thus, it is not a question of knowledge, but first and foremost of spiritual achievement. Knowledge alone may even be an obstacle. The concept recurs in LKCT-VP 5.243, where it is stated that in India, the Vajra-holder (= ‡åkyamuni) hid the adamantine words (namely, 83 This meaning of liõgin is known from Brahmanical, ‡aiva, Vaißñava and Buddhist texts; it is common to contrast gr¢hins/gr¢hasthas and liõgins, cf., e.g., Soma†ambhupaddhati, †råddhasåmånyalakßaña section, verse 2, pp. 624-625 (H. Isaacson, personal communication). 84 While introducing SU 63ab, in his SU~ Nåropå makes reference to those who are confused and deceived (vipralabdha) for having heard laghutantras whitout the salvific assistance of Bodhisattvas. 85 In the Sekodde†a™ippañœ we read: abhij∞ålåbhibhir iti Ù da†abhümœ†varamahåbodhisattvair eva […] pañ∂itair iti alabdhabhümikaiΔ (p. 11915-17).

278

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

those concerning the ‘supreme unchanging pleasure’) in the Royal Tantras (yogatantra and yoginœtantra), so that learned Buddhists who presume to have knowledge will not understand the meaning by simply reading, and without listening to a master. The writings of the early Kålacakra authors reflected a kind of clericalism, since they sought to create a strong, united front to face the new powerful enemies, and to establish a strong and solid orthodoxy with the necessary hierarchy to sustain it. There was no room for doubt within the tradition. Thus, the Kålacakra rather than codifying a popular movement –although it does assimilate and systematize the insights of the Siddhas, whose Apabhra∫†a stanzas are quoted by the dozen– or representing the apex of the gradual evolution of Buddhist tantrism, is, primarily, the product of a number of individuals who consciously developed a system that drew and elaborated on existing practices and doctrines. The Kålacakra authors placed themselves at the centre of the Buddhist tantric movement by defining a new orthodoxy. They did not set themselves up as one of the tantric currents, but as the orthodox tantric current par excellence. They sought, perhaps for the first time in tantric Buddhist circles, to create a consistent, unifying system, capable of holding its own, firstly at the intrabuddhist and interreligious level (especially with Hindu tantric traditions); secondly, with Islam that was an ever-growing presence. The Kålacakra authors maintained that it was necessary to ally Buddhism with a subordinated Hinduism to defeat Islam (cf. Newman 1995).86 Thus it was no accident that Buddha was described as the promulgator of the Veda (LKCT 1.156) and that Ya†as warned the bråhmañas against becoming barbarians (VP vol. 1, p. 24 ff.).87

ABBREVIATIONS AND BIBLIOGRAPHY Sigla CIHTS Central Institute of Higher Tibetan Studies sDe dge sDe dge edition of the bKa’ ’gyur and the bsTan ’gyur. Cf. Chibetto Daiz≠ky≠ S≠mokuroku / A Catalogue-Index of the Tibetan Buddhist Canons, by Ui H., Suzuki M., Kanakura Y., Tada T., Sendai, T≠hoku Imperial University aided by Sait≠ Gratitude Foundation, Sendai 1934. 86 On the relationship between the Kålacakra and Islam, see Hoffmann 1960, 1969; Grönbold 1992: 277-278, 284, 292-293; 1996; Orofino 1997; Newman 1998a. 87 Cf. Newman 19912: 60.

III. How to Produce, Construct and Legitimate a Tradition

DhœΔ IIJ IsIAO NAK NGMPP Q

RBTS RSO SOR WZKS

279

DhœΔ. Journal of Rare Buddhist Texts Research Project Indo-Iranian Journal Istituto Italiano per l’Africa e l’Oriente National Archives, Kathmandu Nepal-German Manuscript Preservation Project Qianlong: The Tibetan Tripi™aka. Peking Edition. Reprinted under the supervision of the Otani University, Kyoto, vols. 1-168. Edited by Suzuki D.T., Suzuki Research Foundation, Tokyo-Kyoto 1955-1961 Rare Buddhist Text Series Rivista degli Studi Orientali Serie Orientale Roma Wiener Zeitschrift fr die Kunde Sdasiens

Sources Ravi†rœj∞åna, Amr¢takañikå: Åryama∞ju†rœnåmasa∫gœti with Amr¢takañikå™ippañœ by Bhikßu Ravi†rœj∞åna and Amr¢takañikodyotanibandha of Vibhüticandra, edited by B. Lål, Bibliotheca Indo-Tibetica 30, CIHTS, Sarnath 1994. AKU Vibhüticandra, Amr¢takañikoddyotanibandha: see AK. Amr¢tavajra, Dohåko†a™œkå, edited in DhœΔ, vol. 32 (2001), pp. 127-155; cf. Q vol. 87, #5049, fols. 31b6-48a3. Anaõgayogin, ˘åkinœjålasa∫vararahasya, edited by S. Rinpoche, V.V. Dwivedi, RBTS 8, CIHTS, Sarnath 1990. Ånandakanda, Madras Government Oriental Series 59, Madras 1952. Anupamarakßita, ¯a∂aõgayoga: The ¯a∂aõgayoga by Anupamarakßita with Ravi†rœj∞åna’s Guñabharañœnåmaßa∂aõgayoga™ippañœ, text edition and annotated translation by F. Sferra, SOR 85, IsIAO, Roma 2000. Åryadeva, Gurupa∞cå†ikå: 1) edition of the first 33 stanzas by S. Lévi in «Journal Asiatique», vol. 215 (1929), pp. 255-263; 2) edition and retranslation of stanzas 34-50 by J. Pandey, DhœΔ, vol. 13 (1992), pp. 16-20 [reprinted in Bauddhalaghugranthasa∫graha, RBTS 14, CIHTS, Sarnath 1997, pp. 33-53]. Bu ston Rin chen grub, dBaõ gi le’u’i ’grel mchan: The Collected Works of Bu-ston. Part 2 (Kha), edited [= reproduced from an original xylogaph] by L. Chandra, ‡ata-Pi™aka Series, Indo-Asian Literatures 42, International Academy of Indian Culture, New Delhi 1965, fols. 231-274. ______ rGyud sde’i zab don sgo ’byed rin chen gces pa’i lde mig ces bya ba: The Collected Works of Bu-ston. Part 4 (Õa), edited [= reproduced from an original xylograph] by L. Chandra, ‡ataPi™aka Series, Indo-Asian Literatures 44, International Academy of Indian Culture, New Delhi 1965, fols. 1-92. Cañ∂amahåroßañatantra: The Cañ∂amahåroßaña Tantra, Chapters IVIII. A Critical Edition and English Translation by Ch. S. AK

280

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

George, American Oriental Series 56, American Oriental Society, New Haven (Conn.) 1974. Candrakœrti, ¯a™ko™ivyåkhyå: Guhyasamåjatantrapradœpodyotana™œkåßa™ko™ivyåkhyå, edited by C. Chakravarti, Kashi Prasad Jayaswal Research Institute, Patna 1984. CST Cakrasa∫varatantra: ‡rœherukåbhidhånam Cakrasa∫varatantram with the Vivr¢ti Commentary of Bhavabha™™a, edited by J.S. Pandey, RBTS 26, 2 vols., CIHTS, Sarnath 2002. ˘åkinœvajrapa∞jaratantra: Årya∂åkinœvajrapa∞jaramahåtantraråjakalpanåma, Q vol. 1, #11, fols. 262a6-301b3. Dharmåkara†ånti, Kålacakrabhagavatsådhanavidhi, edited in DhœΔ, vol. 24 (1997), pp. 127-174. ’Gos lo tså va gz^on nu dpal, Deb ther sõon po: The Blue Annals, reproduced by L. Chandra, ‡ata-Pi™aka Series, Indo-Asian Literatures 212, International Academy of Indian Culture, New Delhi 1976. See also Roerich 19762. Govinda Bhågavat, Rasahr¢daya, edited by J. Tricumji, Åyurvedœya Granthamålå 1, Bombay 1910-1911. GST Guhyasamåjatantra: The Guhyasamåja Tantra. A New Critical Edition, edited by Matsunaga Y., Toho Shuppan, Osaka 1978. Guhyådi-Aß™asiddhi-Saõgraha, edited by S. Rinpoche, V.V. Dwivedi, Sanskrit and Tibetan text, RBTS 1, CIHTS, Sarnath 1987. HT Hevajratantra: see Snellgrove 1959. HTP~ Vajragarbha, Hevajratantrapiñ∂årtha™œkå: see Sferra 1999. Indrabhüti, J∞ånasiddhi, see Guhyådi-Aß™asiddhi-Saõgraha, pp. 89-157. Jagaddarpaña, Kriyåsamuccaya: Kriya-samuccaya. A Sanskrit Manuscript from Nepal Containing a Collection of Tantric Ritual by Jagaddarpaña, reproduced by L. Chandra, International Academy of Indian Culture, ‡ata-pi™aka Series, vol. 237, New Delhi 1977. Jayåkhyasa∫hitå: Jayåkhyasaμhitå of På∞caråtra Ågama, critically edited with an Introduction in Sanskrit, Indices etc. by E. Krishnamacharya, Gaekwad’s Oriental Series 54, Oriental Institute, Baroda 1967. J∞ånodayatantra: edited by S. Rinpoche, V.V. Dwivedi, RBTS 2, CIHTS, Sarnath 1988. Kalparåja: IsIAO, Rome, MS FGT V/1, 263. Kambala, Ålokamålå: Kambala’s Ålokamålå, in Miscellanea Buddhica, edited by Chr. Lindtner, Indiske Studier, vol. V, Copenhagen 1985, pp. 109-220. Kelikuli†a, Trivajraratnåvalœmålikå: Hevajra∂åkinœjålasa∫varapa∞jikå, IsIAO, Rome, MS 3.23, folder 43 [photos of the same MS taken by R. Såõkr¢tyåyana and listed by him as MS XXV, 118; cf. Såõkr¢tyåyana 1935: 38]. mKhas grub rje, Dus ’khor ™œk chen: rGyud thams cad kyi rgyal po bcom ldan ’das dpal dus kyi ’khor lo mchog gi daõ po’i saõs rgyas kyi rtsa ba’i rgyud las phyuõ ba bsdus pa’i rgyud kyi ’grel chen rtsa

III. How to Produce, Construct and Legitimate a Tradition

281

ba’i rgyud kyi rjes su ’jug pa stoõ phrag bcu g∞is pa dri ma med pa’i ’od kyi rgya cher b†ad pa de kho na ∞id snaõ bar byed pa z^es bya ba, in mKhas grub dGe legs pa, Yab sras gsuõ ’bum: mKhas grub. Kha, Tibetan Cultural Printing Press (‡es rig bar khaõ), Dharamsala 1983, pp. 97-1113. Lakßmœõkarå, Sahajasiddhi: see Shendge 1967. Lakßmœ Bha™™årikå, *Sahajasiddhipaddhati: sDe dge #2261, fols. 4a325a1; Q vol. 69, #3108, fols. 4b8-29a7. Kubjikåmatatantra: The Kubjikåmatatantra. Kulålikåmnåya Version, Critical edition by T. Goudriaan and J.A. Schoterman, Orientalia Rheno-Traiectina 30, E.J. Brill, Leiden 1988. LKCT Laghukålacakratantra: 1) ‡rœkålacakratantraråja. A critical Edition, edited by B. Banerjee, The Asiatic Society, Bibliotheca Indica, Calcutta 1985; 2) Kålacakra-Tantra and Other Texts: Part I, edited by L. Chandra, R. Vira, International Academy of Indian Culture, ‡ata-pi™aka Series, vol. 69, New Delhi 1966, pp. 53-378. LT~ Vajrapåñi, Laghutantra™œkå: see Cicuzza 2001. Mahimabha™™a, Vyaktiviveka: Vyaktiviveka edited with a Sanskrit Commentary of Råjånaka Ruyyaka and the Madhusüdani Commentary, edited by M. Mi†ra, The Kashi Sanskrit Series 121, Chaukhamba Sanskrit Sansthan, Varanasi 1936. Målinœvijayottaratantra: ‡rœmålinœvijayottaratantram, edited by M.K. Shâstrî, Kashmir Series of Texts and Studies 37, Bombay 1922. Mataõgapårame†varågama. (Vidyåpåda) avec le commentaire de Bha™™a Råmakañ™ha, édition critique par N.R. Bhatt, Publications de l’Institut Français d’Indologie 56, Pondichéry 1977. MNS Ma∞ju†rœnåmasaõgœti, see AK. Någårjuna, Mülamadhyamakakårikå: Mülamadhyamakakårikå, edited by J.W. de Jong, The Adyar Library Series 109, The Adyar Library and Research Centre, Adyar 1977. Någårjuna, Piñ∂œkramasådhana: Piñ∂œkrama: 1) edited by L. de la Vallée Poussin in Pa∞cakrama, Grand-Louvain 1896, pp. 114; 2) edited by R.S. Tripathi in Piñ∂œkrama and Pa∞cakrama of Åcårya Någårjuna, Bibliotheca Indo-Tibetica Series 25, CIHTS, Sarnath 2001, pp. 3-32. Nåropå, Paramårthasaõgraha = Sekodde†a™œkå: see SU~. Pådmasa∫hitå: Pådma Sa∫hitå [prathamo bhågaΔ], critically edited by S. Padmanabhan, R.N. Sampath, Pancaratra Parisodhana Parisad Series 3, Madras 1974. Padmavajrapåda, Guhyasiddhi: see Guhyådi-Aß™asiddhi-Saõgraha, pp. 1-62. Puñ∂arœka, dPal don dam pa’i bs∞en pa (*Paramårthasevå), Q vol. 47, #2065, fols. 1-25a. ______ Vimalaprabhå: see VP. Rasahr¢dayatantra: see Govinda Bhågavat.

282

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Rasårñavakalpa: Rasårñavakalpa, edited and translated by M. Roy in collaboration with B.V. Subbarayappa, Indian National Science Academy, New Delhi 1976. Ravi†rœj∞åna, Guñabhårañœ: see Anupamarakßita, ¯a∂aõgayoga. Sådhuputra ‡rœdharånanda, Sekodde†a™ippañœ: La Sekodde†a™ippañœ di Sådhuputra ‡rœdharånanda. Il testo sanscrito, a cura di R. Gnoli, in RSO, vol. 70 (1997), n. 1-2, pp. 115-146. Sa∫varodayatantra: see Tsuda 1974. ‡ayanåsanavastu: The Gilgit Manuscript of the ‡ayanåsanavastu and the Adhikarañavastu. Being the 15th and 16th Sections of the Vinaya of the Mülasarvåstivådin, edited by R. Gnoli, SOR 50, Istituto Italiano per il Medio ed Estremo Oriente, Roma 1978, pp. 1-56. Sekodde†apa∞jikå: Sekodde†apa∞jikå, edition of the Sanskrit text, in «Annual of the Institute for Comprehensive Studies of Buddhism», Taish≠ University, vol. 16 (1994), pp. 354-289. Soma†ambhupaddhati. Troisième partie. Rituels occasionnels dans la tradition †ivaïte de l’Inde du Sud selon Soma†ambhu. II: dœkßå, abhißeka, vratoddhåra, antyeß™i, †råddha, Texte, Traduction et Notes par H. Brunner-Lachaux, Publications de l’Institut Français d’Indologie No. 25.III, Institut Français d’Indologie, Pondichéry 1977. SU Sekodde†a, see Orofino 1994; Gnoli 1999. SU~ Nåropå, Sekodde†a™œkå: 1) editio princeps: Sekodde†a™œkå of Na∂apåda (Nåropå), edited by M. Carelli, Gaekwad’s Oriental Series, vol. 90, Baroda 1941; 2) The Sekodde†a™œkå by Nåropå (Paramårthasa∫graha), critical edition of the Sanskrit text by F. Sferra and critical edition of the Tibetan translation by S. Merzagora, SOR, IsIAO, Roma (forthcoming). Vißñusahasranåma: Vißñusahasranåma with the Bhåßya of ‡rœ ‡a∫karåcårya. Transaletd into English in the Light of ‡rœ ‡a∫kara’s Bhåßya by R.A. Sastry, The Adyar Library and Research Centre, The Adyar-Library General Series 8, Adyar 1980. VP Puñ∂arœka, Vimalaprabhå: Vimalaprabh噜kå of Kalkin ‡rœpuñ∂arœka on ‡rœlaghukålacakratantraråja by ‡rœma∞ju†rœya†as, vol. I, edited by J. Upadhyaya, Bibliotheca Indo-Tibetica Series 11, CIHTS, Sarnath 1986; vols. II-III, edited by V.V. Dwivedi, S.S. Bahulkar, RBTS 12-13, CIHTS, Sarnath 1994. Yoginœsa∞cåratantra: Yoginœsa∞cåratantram with Nibandha of Tathågatarakßita and Upade†ånusåriñœvyåkhyå of Alakakala†a, edited by J.S. Pandey, RBTS 21, CIHTS, Sarnath 1998. Studies Bernardini, M. (2003), Storia del mondo islamico (VII-XVI secolo). Volume secondo. Il mondo iranico e turco dall’avvento dell’Islàm all’affer-

III. How to Produce, Construct and Legitimate a Tradition

283

mazione dei Safavidi, Piccola Biblioteca Einaudi, Storia e geografia 252, Torino. Chimpa, L., Chattopadhyaya, A. (transl.) (1970), Tåranåtha’s History of Buddhism in India, Motilal Banarsidass, Delhi. Cicuzza, C. (ed.) (2001), The Laghutantra™œkå by Vajrapåñi, SOR 86, IsIAO, Roma. Cicuzza, C., Sferra, F. (1997), Brief Notes on the Beginning of the Kålacakra Literature, in DhœΔ, vol. 23, pp. 113-126. De Rossi Filibeck, E. (1994), Catalogue of the Tucci Tibetan Fund in the Library of IsMEO, Volume 1, Istituto Italiano per il Medio ed Estremo Oriente, Rome. Dyczkowski, M.S.G. (1988), The Canon of the ‡aivågama and the Kubjikå Tantras of the Western Kaula Tradition, State University of New York Press, Albany (N.Y.). Gnoli, R. (1994), Introduzione, in Gnoli, Orofino 1994, pp. 11-103. ______ (1997), La realizzazione della conoscenza del Supremo immoto (Paramåkßaraj∞ånasiddhi), in RSO, vol. 70, Supplemento n. 1. ______ (1999), Sekodde†aΔ [Edition of the Sanskrit Text], in DhœΔ, vol. 28, pp. 143-166. Gnoli, R., Orofino, G. (transl.) (1994), Nåropå. Iniziazione (Kålacakra), Biblioteca Orientale 1, Adelphi, Milano. Grönbold, G. (1992), Heterodoxe Lehren und ihre Widerlegung im Kålacakra-Tantra, in IIJ, vol. 35, n. 4, pp. 273-297. ______ (1996), Kriegsmaschinen in einem buddhistischen Tantra, in Wilhelm, F. (ed.), Festschrift Dieter Schlingloff, Verlag fr Orientalistische Fachpublikationen, Reinbek, pp. 63-97. Haig, T.W. (1987), Cambridge Shorter History of India, Cambridge University Press, Cambridge. Hoffmann, H.H.R. (1960), Manichaeism and Islam in the Buddhist Kålacakra System, in Proceedings of the 9th International Congress for the History of Religions, Tokyo, pp. 96-99. ______ (1969), Kålacakra Studies I. Manichaeism, Christianity, and Islam in the Kålacakra Tantra, in «Central Asiatic Journal», vol. 13, n. 1, pp. 52-73. Lål, B. (1994), Bauddh tantr vå∫may kå paricay. (Kålacakratantra), in DhœΔ, vol. 18, pp. 19-34. Lamotte, É. (1966), Vajrapåñi en Inde, in Mélanges offerts à Monsieur Paul Demiéville, Presses Universitaires de France, Paris, pp. 113-159. Newman, J.R. (1987a), The outer Wheel of Time: Vajrayåna Buddhist cosmology in the Kålacakra tantra, Ann Arbor, University Microfilms International (University of Wisconsin, Madison, Ph.D. Thesis). ______ (1987b), The Paramådibuddha (The Kålacakra mülatantra) and its Relation to the Early Kålacakra Literature, in IIJ, vol. 30, n. 2, pp. 93-102. ______ (19912), A brief history of Kålacakra, in Simon, B. (ed.), The Wheel of Time. The Kalachakra In Context, Snow Lion Publications,

284

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Ithaca (N.Y.), pp. 51-90 [Deer Park Books, Madison, Wisconsin 19851]. ______ (1995), Eschatology in the Wheel of Time Tantra, in Lopez, D.S. Jr. (ed.), Buddhism in Practice, Princeton University Press, Princeton (N.J.), pp. 284-289. ______ (1998a), Islam in the Kålacakra Tantra, in «Journal of the International Association of Buddhist Studies», vol. 21, n. 2, pp. 311-371. ______ (1998b), The Epoch of the Kålacakra Tantra, in IIJ, vol. 41, n. 4, pp. 319-349. ______ (2004), Kålacakra, in Buswell, R.E. Jr. (ed.), Encyclopedia of Buddhism, vol. 1 (A-L), Thomson, New York, pp. 408-411. Nihom, M. (1984), Notes on the Origin of Some Quotations in the Sekodde†a™œkå of Nå∂apåda, in IIJ, vol. 27, n. 1, pp. 17-26. Obermiller, E. (transl.) (19992), The History of Buddhism in India and Tibet by Bu-ston, Classic India Publications, Delhi [19271]. Orofino, G. (1994), Sekodde†a. A Critical Edition of the Tibetan Translation. With an Appendix by Raniero Gnoli On the Sanskrit Text, SOR 72, Istituto Italiano per il Medio ed Estremo Oriente, Roma. ______ (1997), Apropos of Some Foreign Elements in the Kålacakratantra, in Krasser, H., Much, M.T., Steinkellner, E., Tauscher, H. (eds.), Tibetan Studies. Volume II, Verlag der Österreichischen Akademie der Wissenschaften, Wien 1997, pp. 717-724. Reigle, D. (1986), The Lost Kålacakra Müla Tantra on the Kings of ‡ambhala, Kålacakra Research Publications 1, Eastern School 1, Talent (Oregon), pp. 1-14. Roerich, G.N. (transl.) (19762), The Blue Annals. Parts I and II (Bound in One), Motilal Banarsidass, Delhi [Calcutta 19491]. Samphel, ~h. (1995), Bauddh tantr vå∫may kå paricay. Kålacakratantra (Bho™ khañ∂), in DhœΔ, vol. 20, pp. 99-136. Såõkr¢tyåyana, R. (1935), Sanskrit Palm-Leaf MSS. in Tibet, in «Journal of the Bihar and Orissa Resarch Society», vol. 21, part. 1, pp. 21-43. Sferra, F. (1995), Textual Criticism Notes on the Vimalaprabhå by Puñ∂arœka, in «East and West», vol. 45, pp. 359-364. ______ (1999), The ¯a™såhasrikåkhyå Hevajratantrapiñ∂årtha™œkå by Vajragarbha, critical edition and annotated translation by F. Sferra, Università degli Studi di Roma “La Sapienza”, Doctorate Thesis [a revised edition of this work is in press (Firenze University Press)]. ______ (2001), Alcune note sulla Hevajratantrapiñ∂årtha™œkå di Vajragarbha, in Botto, O. (ed.), Atti dell’Ottavo Convegno Nazionale di Studi Sanscriti (Torino 20-21 ottobre 1995), Associazione Italiana di Studi Sanscriti, Torino, pp. 125-135. ______ (2004), Teaching and Spiritual Counselling in Indian Buddhist Traditions. Some Considerations on the Role of the Kalyåñamitra,

III. How to Produce, Construct and Legitimate a Tradition

285

in Rigopoulos, A. (ed.), Guru. The Spiritual Master in Eastern and Western Traditions: Authority and Charisma, Indoasiatica 2, Cafoscarina, Venezia, pp. 345-371. Shendge, M.J. (1967), ‡rœsahajasiddhi, in IIJ, vol. 10, n. 2-3, pp. 126-149. Snellgrove, D.L. (1959), The Hevajra Tantra. A Critical Study. Part I. Introduction and Translation; Part II. Sanskrit and Tibetan Texts, London Oriental Series 6, London. ______ (1987), Indo-Tibetan Buddhism. Indian Buddhists & their Tibetan Successors, 2 vols., Shambala, Boston-London. Tsuda S. (1974), The Saμvarodaya-Tantra. Selected Chapters, The Hokuseido Press, Tokyo. Wallace, V.A. (2001), The Inner Kålacakratantra. A Buddhist Tantric View of the Individual, Oxford University Press, New York. Wink, A. (20022), Al-Hind. The Making of the Indo-Islamic World. Vol. II. The Slave Kings and the Islamic Conquest. 11th-13th Centuries, Brill Academic Publishers, Boston-Leiden [19961].

ELISA FRESCHI, ALESSANDRO GRAHELI Bhå™™amœmå∫så and Nyåya on Veda and Tradition*

Introduction This paper deals with the development of two philosophical traditions (i.e. successions of thinkers), namely Nyåya and Bhå™™amœmå∫så. Our attempt is to juxtapose the two schools on a specific philosophical problem, to highlight points of contact and departure, and possibly to get a better picture of the evolution of their thought by underlining inherited concepts and mutual influences. In doing so we will show how the representatives of the two streams further explained the original arguments of the sütras, added new perspectives, and replied to issues raised by opponent schools. Particularly, our focus is on the defense of the validity (pråmåñya) of Veda. The reason for this choice is that Veda is held as an autonomous source of information about religious matters by the two schools under exam1 and is widely accepted in their social context as a source of knowledge. In this sense it constitutes the tradition (i.e. the cultural horizon) of both. In their bid to defend Veda’s validity, Nyåya and Mœmå∫så have also a further purpose (for Nyåya, arguably, the main purpose): justifying the epistemic role of tradition (i.e. word-conveyed knowledge). * This paper is the product of discussions, mutual advise and criticism shared by the two authors. Specifically, the introduction, chapters 2, 4, and 6 must be ascribed to Alessandro Graheli; the conclusions, chapters 1, 3, and 5 to Elisa Freschi. To facilitate the distinction between our comments and the original sources, all direct quotations from Sanskrit texts are given as set-off block quotations in smaller type. Our experiment has been, as far as possible, to ‘let the authors speak’. 1 And, as George Chemparathy states, “[…] les Mœmå∫sakas […] sont les partenaires les plus acharnés des Nyåya-Vai†eßikas dans leur controverses sur la source de l’autorité du Veda” (Chemparathy 1983: 7).

288

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

In the following pages the different acceptations of the word ‘tradition’ –respectively employed, from the strictly grammatical point of view, as a class-noun, a noun denoting a state of affairs, and an action noun– will clearly surface: we are going to talk of traditions that, within the horizon of a cultural tradition, discuss the epistemic role of tradition. That is, our subject matter concerns (1) the development and interaction of two philosophical schools, (2) the cultural horizon –partially departing and partially overlapping– surrounding the two schools, (3) the epistemic function attributed by them to linguistic communication. We will treat the authors concerned in a chronological order. In the Mœmå∫så camp we will deal with Jaimini (II century B.C.), ‡abara (IV-V century A.D.?), and Kumårila Bha™™a (VII century A.D.). In the Nyåya camp we will examine the positions of Gautama (II century A.D.?), Våtsyåyana (V century A.D.?), Uddyotakara (VI century A.D.), and finally Jayanta Bha™™a (IX century A.D.).2 The order of the paragraphs respects the chronology of the authors, thus alternating Mœmå∫sakas and Naiyåyikas, to underline the mutual interaction of the two schools. To get a better view of the inner development of each school, however, the reader might choose to read first the Mœmå∫saka paragraphs in a sequence, and then the Naiyåyika ones (or vice versa). Our goal is to assess the degree of continuity, or discontinuity, in each of the two schools, rather than extensively present the thought of each author involved. Hence, within the borders of epistemic issues related to Veda’s validity, we selected a few themes which were kept alive by the two traditions throughout the centuries; their development is schematized in each Mœmå∫så and Nyåya paragraph through a symmetric articulation of sub-paragraphs (e.g. §§ 1.1, 3.1, 5.1). This linearity becomes less evident with the latest authors examined in the paper, Kumårila and, even more so, Jayanta. In fact, their works cover a wider range of philosophical issues and are clearly more indebted to the mutual influences of the two traditions. 1. Jaimini Although Jaimini’s time and role in the Mœmå∫så school are still a matter of debate (see Parpola 1981; 1994), it can be safely said that the Mœmå∫såsütra (MS) attributed to him is the most ancient text of Mœmå∫så and is most probably older than the roottexts of the other philosophical systems; despite its antiquity, how2 In order to treat the issue with sufficient depth we decided to examine a limited number of authors and to narrow down the covered period, leaving necessarily out many important sources. Other schools which said much on these topics –especially Pråbhåkaramœmå∫så, Buddhist Epistemology, and Vedånta– were left out. The aim of the paper is obviously not at exhaustiveness, so we hope the specialists will bear with us.

III. How to Produce, Construct and Legitimate a Tradition

289

ever, it mentions former authors and views. This suggests an early existence of different opinions in the larger picture of an already developed Mœmå∫så system. The MS does not problematize the Veda’s validity, which rather is the implicit assumption behind its teachings. Nevertheless, embedded in the text are already present the chief components of the defense of Veda’s validity. First of all there is a restriction of the scope of Veda: Veda is accepted as instrument of knowledge only within the sphere of dharma and, among Vedic statements, only the prescriptive ones are considered means of knowledge. Then, MS sets the ground for the assertion of an authorless Veda by stating the beginninglessness of Veda’s transmission, and for the foundation of Veda’s validity by postulating the intrinsic relation of word and meaning. Here and in all the next Mœmå∫så paragraphs (§ 3 and § 5) I shall follow the development of these arguments according to the order in which they are found in MS and MS’s commentaries. 1.1. The word-meaning relation is not human-made The MS opens with a declaration of its goal: the examination of dharma. Which are the means for this inquiry? Sense perception, says Jaimini, is not fit since it can only reach present objects. The right instrument to know dharma is instead Vedic word (†abda). MS 1.1.5 insulates the sphere of dharma from that of present objects, and states the intimate relationship of word (†abda) and meaning (artha):3 But the relation of word and meaning4 is originary (autpattika).5 Its (dharma’s)6 knowledge is the Instruction (i.e. Veda) and it is infallible as far as unknown entities. It is an instrument of knowledge [to know dharma], according to Bådaråyaña, because it is independent.7

Noticeably, word is here treated as a given datum. Its possible author and etiology are secondary problems, unessential for its actual reality, that is the relationship with a (mental or concrete) meaning. 3 I render artha with ‘meaning’, in want of a better option. ‘Meaning’, in fact, lacks the ontological commitment often present in artha and can be ambiguous because of the ending -ing, suggesting an active role as if it were a våcaka and not a våcya. 4 Francis X. Clooney maintains that here artha means also ‘purpose’ because of the ritual background of the MS, where objects are all viewed in the framework of (ritual) action (Clooney 1990: 104). 5 On autpattika see § 3.1, and Frauwallner 1961: 119ff; Clooney 1990: 67 n. 29; Bilimoria 1994: 190ff. 6 What does ‘its’ (tasya) refer to? The ‡Bh understands it as connected with the previous sütras and glosses it as ‘of the dharma’ (dharmasya). See also Francis X. Clooney (1990: 136). On the other hand, Peter M. Scharf links it to the immediately preceeding ‘relation’ (sambandha) (Scharf 1996: 269). In any case, this does not invalidate the connection of MS 1.1.5 with dharma. 7 autpattikas tu †abdasyårthena sambandhas tasya j∞ånam upade†o ’vyatireka† cårthe ’nupalabdhe tat pramåñam bådaråyañasyånapekßatvåt (MS 1.1.5).

290

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Epistemological independence is needed because an instrument of knowledge, according to Mœmå∫sakas, must provide fresh information, i.e. it must cause to know a referent not previously known through other means of knowledge. By stating at the outset the nexus with dharma, Jaimini focuses on †abda as Vedic word; †abda as acoustic entity and as ordinary language is dealt with in the following sütras (1.1.6-23). No matter if composed by Jaimini or not,8 these look like a more recent layer of the MS, where Vai†eßika and Naiyåyika objections are presented and refuted. The proposal that †abda is a product (karman) is discarded as follows:9 On the contrary [words] must be permanent (nitya)[in their relation to their meaning], because [their] appearance10 is for the purpose of [communicating with] someone else.11

In Jaimini’s view, if words were ephemeral, how could we recognize them and hence understand their meaning? Their relationship with meant entities must be understood as fixed, in order to explain communication. Common linguistic communication, however, is rather neglected in MS, which chiefly discusses dharma-related issues. 1.2. Veda is an instrument of knowledge only in relation to dharma Although most Mœmå∫saka justifications of (Vedic) word as an instrument of knowledge (†abdapramåña) will later be based on this assumption, an explicit restriction appears only in the sixth book of MS: It is only in the case of what cannot be known [by ordinary means of knowledge] that the sacred text (†åstra) can serve a useful purpose.12

1.3. Only prescriptions are means of knowledge MS 1.2 deals with the epistemic status of non prescriptive passages (mantras and arthavådas). These Vedic portions, says an opponent, cannot be considered means of knowledge because they contradict real experience, are inconsistent, etc. (MS 1.2.2-5). Jaimini’s solution is to separate mantras and arthavådas (which may be read as eulogies, figurative descriptions, and so on) from prescriptions (vidhis), which are genuine means of knowledge. Nonetheless, mantras and arthavådas may still serve an ancillary role to prescriptions.13 8 On this issue, see Frauwallner 1961; Frauwallner 1968: 17; D’Sa 1980: 113ff; Clooney 1990: 77ff. 9 Similar Mœmå∫saka objections and Naiyåyika replies are, vice versa, quoted in NS 2.2.13-39. According to Arthur Berriedale Keith (1921: 37) the MS version is older. 10 ‡Bh ad 1.1.18 glosses: ‘appearance means utterance’ (dar†anam uccårañam). 11 nityas tu syåd dar†anasya parårthatvåt (MS 1.1.18). 12 […] apråpte vå †åstram arthavat (MS 6.2.18). 13 vidhinå tv ekavåkyatvåt stutyarthena vidhœnå∫ syüΔ (MS 1.2.7).

III. How to Produce, Construct and Legitimate a Tradition

291

1.4. Transmission as beginningless succession In answer to the criticism against mantras and arthavådas, Jaimini says that they cannot be simply cut out from the Veda, because they have been handed down to us along with prescriptions by the same teacher-pupil tradition. And it is equally transmitted in a tradition.14

This argument displays the typical Mœmå∫saka commitment to the empirical state of things –we have to deal with the Veda as it is–, the refusal to postulate whatever contradicts today’s reality (e.g. an arthavåda-free Veda), and the acceptance of the teacher-pupil transmission. Since he does not contemplate the option of an originally different Veda which was later corrupted, Jaimini seems to regard transmission as a dynamic but everlasting reality. In the next Mœmå∫så paragraphs (§ 3 and § 5) these very arguments will be unfolded, albeit in a background where the obviousness of Veda’s validity cannot anymore be taken as undoubtedly granted. 2. Gautama The date and historicity of Gautama, or Akßapåda, are by no means settled. The current version of the Nyåyasütra (NS) could probably trace back to 200 A.D. This dating is tentative, because there are no external evidences for it. It is inferred from the assumed crossrelation of Någårjuna’s writings and NS, although there is no consensus on the relative chronology of the two authors. Whatever its date and no matter if the work of a single author or the sum of many layers, however, the NS is the unquestioned roottext of the Nyåya tradition and its earliest known authority. NS, with a typical structure that will be adopted in most successive Naiyåyika writings, treats every subject in three phases: enumeration of the items, definition, and examination of the definition. I will follow a shortened version of the same pattern, by discussing in §§ 2.1, 4.1, 6.1 the definition of †abda, and in §§ 2.2, 4.2-3, 6.2-3 its examination. 2.1. †abda as testimony. In NS 1.1.3 the four instruments of knowledge admitted by Naiyåyikas are enumerated as sense perception, inference, analogy and testimony. Testimony15 (†abda) is defined as follows: tulya∫ ca såmpradåyikam (MS 1.2.8). It seems to me that the epistemic value of †abda in older Nyåya is quite akin to the Western concept of testimony. With Jayanta the analysis becomes less speaker-centered, and the same term, †abda, is more often used in linguistic, psychological, and lay contexts, 14 15

292

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Testimony is the statement (upade†a) of a reliable speaker. / It is of two kinds, for it can have a perceivable or an unperceivable object.16

The principle of the speaker’s reliability concerns human word in general, leaving unspecified the distinction between common and Vedic speech. On the other hand, this distinction is implicitly suggested by the division according to the type of object, perceivable and unperceivable. 2.2. Reasons for the validity of testimony NS 2.1.49-67 examines the definition given in NS 1.1.7, quoted above. Firstly, Gautama measures himself with a Vai†eßika opponent: [O.] Testimony is [reducible to] inference, because it has an inferable object which is not [directly] perceived. For, since [both their objects are inferentially] perceived, there is no different performance of the two (testimony and inference). And also because of the relation [functional both in testimony and inference]. [R.] Thorough knowledge of the object derives from testimony because of the efficacy of a reliable speaker’s statement. And [in testimony] there is not such a [word-meaning] relation because there is no perception of filling, burning, splitting [upon uttering ‘food’, ‘fire’, ‘sword’]. [O.] [The word-meaning relation] is indisputable, because of the fixed word-meaning structure.17

This last sentence, if taken in context, is to be ascribed to a Vai†eßika opponent, according to whom the word-meaning fixedness is akin to the invariable concomitance in inferential processes. But it can also be taken as an argumentative bridge that, as interpreted by Våtsyåyana, conceptually evokes the Mœmå∫saka position (see § 4.2), although the term for ‘fixedness’ is vyavasthå and not nityatå. The Naiyåyika answer is that the apparently fixed structure is just based on a commonly shared convention, and that the tenet of a fixed word-meaning relation bears unwanted consequences: No, [there is no fixed word-meaning structure] because the proper knowledge of a meaning from its word is due to a convention. in which cases ‘testimony’ is not anymore a suitable translation. Therefore in the Nyåya environment I will render †abdapramåña with ‘testimony’ when the context is clearly epistemological. In linguistic occurrences, I will render †abda with ‘speech’ when a broad linguistic category, and with ‘word’ when justaposed to ‘meaning’. For intriguing inputs on translating †abdapramåña, see Mohanty (2001: 5-18), who believes that the concept of pramåña does not imply a principle of authority, but is rather a critical norm, and especially denies the interpretation of †abdapramåña as ‘revelation’. 16 åptopade†aΔ †abdaΔ. sa dvividho, dr¢ß™ådr¢ß™årthatvåt (NS 1.1.7-8). 17 †abdo ’numånam arthasyånupalabdher anumeyatvåt. upalabdher advipravr¢ttatvåt. sambandhåc ca. åptopade†asåmarthåc chabdåd arthasampratyayaΔ. pürañapradåhapå™anånupalabdhe† ca sambandhåbhåvaΔ. †abdårthavyavasthånåd apratißedhaΔ (NS 2.1.49-54).

III. How to Produce, Construct and Legitimate a Tradition

293

And because, being [any word-meaning relation valid for] specific classes of people, there is no limitation [of word-meaning relations].18

The next argument is of particular interest because, according to later commentators, it is here that Gautama shifts from the context of testimony in general to Vedic testimony, beginning with the stock objection already tackled by Jaimini (see § 1.3): It (Veda) is not an instrument of knowledge because of the defects of falsity, inconsistency, and tautology.19

NS 2.1.58-68 replies to this objection by an array of Mœmåμsaka arguments (see § 1.3), mainly stating the need of distinguishing Vedic prescriptions (vidhi) from other statements such as descriptive statements (arthavåda) and reiterations (anuvåda), and contextualizing the above-mentioned three defects in this perspective. Eventually Gautama states the only positive argument in favor of Vedapråmåñya given in NS: And that [Veda] is a valid source of knowledge because its reliable speaker is a valid source, like with the validity of incantations and Ayurveda.20

The question of impermanence of sound (†abda, as a broader physical class) is dealt with in NS 2.2.13-39 (see § 1.1). Gautama affirms three reasons why sound must be considered impermanent (a-nitya): Because it has a beginning, because it is sensible matter, because it is considered artificial.21

These justifications of the definition of †abda (NS 1.1.7) were here given in a nutshell, in adherence to the sütra style. They will be further clarified in the next Nyåya sections (§ 4.1-3). 3. ‡abara and the V®ttikåra ‡abara is the author of the first extant commentary on MS, the ‡åbarabhåßya (‡Bh). Quite different hypothesis have been formulated on his date, ranging from II century B.C. (Devasthali 1942) to VI century A.D. (Taber 1998). For the present purpose, enough to say that he did not directly precede Kumårila. An older commentator, the Vr¢ttikåra (the ‘author of the gloss’), is extensively quoted by ‡abara in his commentary on MS 1.1.5. Since these quotations are embedded in ‡abara’s text, it is not always easy na såmayikatvåc chabdårthasampratyayasya. jåtivi†eße cåniyamåt (NS 2.1.55-56). tadapråmåñyam anr¢tavyåghåtapunaruktadoßebhyaΔ (NS 2.1.57). 20 mantråyurvedapråmåñyavac ca tatpråmåñyam åptapråmåñyåt (NS 2.1.68). 21 ådimatvåd aindriyakatvåt kr¢takavadupacåråc ca (NS 2.2.13). 18

19

294

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

to distinguish ‡abara’s views from the Vr¢ttikåra’s. But since ‡abara did not dispute the Vr¢ttikåra’s views, we can assume that he did not disagree with them. Basically, ‡abara’s arguments are the same found in MS, but more philosophically developed. As in MS, also in the ‡Bh the main focus is on ritual application, rather than on theoretical analysis; hence Veda’s validity is only discussed in the relatively short tarkapåda (one chapter out of sixty). ‡abara also hints at the ‘intrinsic validity’ theory which will be later elaborated by Kumårila, and assumes that no clearly understood Vedic notion can turn to be false (Kumårila will reject this thesis; see § 5.1). 3.1. The word-meaning relation is not human-made ‡abara glosses ‘originary’ (autpattika, see § 1.1) with nitya. This term can be interpreted in at least three senses: ‘eternal’, ‘continual’, and ‘fixed’. An ‘eternal’ relation between word and meaning would imply that words and meant entities are themselves metaphysically and ontologically eternal, i.e. existing in time but without beginning or end. A ‘continual’ relation would refer to the diachronic continuity of words and meant entities from the point of view of speakers, who never think of a word independently of its meaning; conversely,22 every time a person thought, created, conceived, touched an object, there had always been word for it. A ‘fixed’ relation would entail that it cannot be changed by the speakers. The second and third options are treated in ‡Bh ad 1.1.5 (§ 3.4). As to the first option, in my view the strictly temporal aspect of nitya is discarded by ‡abara: The relation is the non-separate existence of word and meaning. The relation does not take place afterwards, when they (word and meaning) have been [already] originated.23

This means that it is impossible to conceive a separate existence of word and meaning. The relationship of word and meaning is originary in the sense of ‘mutually inborn’, as they can never exist independently. As for common language, in MS, words (†abda) were said to be nitya because otherwise there could be no intercommunication (see 22 ßabda and artha, however, are not exactly symmetrical, and –at least in Kumårila– the external world is logically independent of the language which describes it. 23 aviyuktaΔ †abdårthayor bhåvaΔ sambandho. notpannayoΔ pa†cåt sambandhaΔ (‡Bh ad MS 1.1.5, F 24.4). Although most editions and manuscripts agree on this reading, F prefers bhåvaΔ sambandhena and translates it accordingly:“Das Wesen von Wort und Gegenstand ist also von der Verknpfung nicht getrennt”. Yet the meaning does not change.

III. How to Produce, Construct and Legitimate a Tradition

295

§ 1.1). ‡abara elaborates further: we recognize (pratyabhij∞å) words and do not think that a new word is uttered every time we hear a repetition of the same word: We say: ‘the word ‘cow’ has been uttered eight times’ and not ‘eight words ‘cow’ have been uttered’ […] And, like us, other people recognize [it] too, [hence] we can safely assert that it is not another [word].24

Going back to Vedic word, the objection might be raised that the Veda cannot be an instrument to know dharma, because it is fallible, like any human word. Accordingly, the Vr¢ttikåra glosses MS 1.1.5 as follows: The word ‘but’ counteracts this opinion (namely, that Veda is fallible like human words). The relation between word and meaning is not human.25

The Vr¢ttikåra believes that Jaimini’s intention is to dispel the view that every word is fallible. Vedic and human words, instead, have altogether different natures. Accordingly, the Vr¢ttikåra glosses autpattika with ‘non-human’ (apaurußeya). The idea of the non-human relation is attacked by Naiyåyikas. The Vr¢ttikåra quotes a Naiyåyika objection, which echoes the argument given in NS 2.1.53 (see § 2.2). This cross-reference with NS gives further justification to the identification of the NS opponent as a Mœmåμsaka (see § 2.2). [O.] There is no relation whatsoever between word and meaning […]. If there were a relation with the meaning, upon uttering the word ‘razor’ or ‘sweetmeat’ there would be splitting or filling of the mouth, provided the relation is meant as a conjunction (sa∫†leßa). And [other kinds of] relations –cause-effect, condition-conditioned, support-supported, etc.–, do not apply to words. [R.] You fail to mention the only relation which applies to the case, the relation between knowledge and cause of knowledge, characterized as denoter-denoted.26

3.2. Veda is an instrument of knowledge only in relation to dharma Vedic words convey knowledge of dharma, i.e. knowledge related to sacrifices. This is why the Vr¢ttikåra (see ‡Bh ad 1.1.5, F 34. 15-17) stress24 aß™akr¢tvo go†abda uccaritaΔ iti vadanti, nåß™au go†abdåΔ iti. […] pratyabhijånånå vayam ivånye ’pi ‘nånyaΔ’ iti vaktum arhanti (‡Bh ad 1.1.20). 25 tu†abdaΔ pakßa∫ vyåvartayati. apaurußeyaΔ †abdasyårthena sambandhaΔ [...] (‡Bh ad 1.1.5, F 34.13-15). 26 naiva †abdasyårthena asti sambandhaΔ […] syåc ced arthena sambandhaΔ, kßuramodaka†abdoccårañe mukhasya på™anapürañe syåtåm, yadi sa∫†leßalakßaña∫ sambandham abhipretya ucyate. kåryakårañanimittanaimittikå†rayå†rayibhåvådayas tu sambandhåΔ †abdasya anupapannå eva iti. ucyate: yo ’tra vyapade†yaΔ sambandhaΔ, tam eka∫ na vyapadi†ati bhavån, pratyåyyasya pratyåyakasya ca, yaΔ sa∫j∞åsa∫j∞ilakßañaΔ sambandhas tam iti (‡Bh ad 1.1.5, F 36.6-14).

296

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

es an important feature of Vedic knowledge: it must refer to something not accessible through other instruments of knowledge, such as sense perception or inference. Hence Vedic knowledge can never be contradicted because it is the only source to know dharma. Consequently, objections based on human experience cannot challenge the Veda: [O.] [Vedic] word is not the cause [of knowledge of dharma]. It is not an instrument of knowledge because at the time of sacrifice we do not see any result [of that sacrifice] and in a successive time [when the result is there] sacrifice is not [anymore] extant. [R.] [...] An instrument of knowledge is that through which one acquires knowledge, and one acquires knowledge also through [Vedic] words. Thus, also [Vedic] words are means of knowledge, just like sense perception. And if something is known by an instrument of knowledge and not known by another, this does not mean that it is not known. Nor does the Veda say that once the sacrifice is done the result has to immediately accrue. It is instead said that a result is achieved by means of sacrifice.27

Thus, Mœmå∫sakas deny the need and even the possibility of testing Veda-originated knowledge through other instruments of knowledge. But one may object that also Vedic utterances –like human utterances– can be false. Thus, a clear differentiation between human and Vedic words is an important premise for the validity of the latter. By themselves, human words are not a source of knowledge and, when based on an erroneous cognition, they are erroneous, too. Vedic words, instead, are never conditioned by someone’s cognition. Commenting on MS 1.1.2,28 ‡abara stresses their different status: [O.] [Here] an inference based on similarity is possible. Human utterance is perceived to be incorrect. Vedic word is also inferred to be false because it shares [with human utterance] the same character of being an utterance. [R.] No, because it is different. If one is incorrect, indeed, it is not that also the other has to be so.29

Rejecting the similarity of human utterance and Vedic word, ‡abara instead proposes the equation of Vedic word and sense perception. In connection to the extra-sensorial sphere, the Veda has the 27 atha yad uktam ‘animitta∫ †abdaΔ; karmakåle phalådar†anåt kålåntare ca karmåbhåvåt pramåña∫ nåsti’ iti. [...] yena yena hi pramœyate tat tat pramåñam. †abdenåpi pramœyate. tataΔ †abdo ’pi pramåñam, yathaiva pratyakßam. na ca pramåñenåvagata∫ pramåñåntareñånavagatam ity etåvatånavagata∫ bhavati. na caiva∫ †rüyate ‘kr¢te karmañi tåvaty eva phala∫ bhavati’, ki∫ tu ‘karmañå phala∫ pråpyata’ iti (‡Bh ad 1.1.5, F 48.16-23). 28 “dharma is the purpose designated by prescriptions” (codanålakßanårtho dharmaΔ). Different translations of this sütra have been given, see Frauwallner 1968: 16; D’Sa 1980: 49-54, especially p. 50; Taber 1983. 29 nanu såmånyato dr¢ß™a∫ bhavißyati. paurußeya∫ vacana∫ vitatham upalabhya vacanasåmånyåd vedavacana∫ mithyeti anumœyate. –na, anyatvåt. na hy anyasya vitathabhåve ’nyasya vaitathya∫ bhavitum arhati. (‡Bh ad 1.1.2, F 18.16-18).

III. How to Produce, Construct and Legitimate a Tradition

297

function of directly showing its referents, just like perception does in the realm of perceivable objects. The evidence of meaning strikes the knower with the same immediate strength of the image of an external object before his eyes. When someone clearly perceives an external object through sense perception, he cannot avoid believing in it. A similar process occurs when someone grasps the meaning of a Vedic prescription. For Vedic statements, the cognition is a direct one.30

Of course, a necessary condition for the direct impact of a Vedic statement is the clear understanding of what is said: Nor can that which is clearly understood be false.31

3.3. Only prescriptions are means of knowledge Jaimini explained (§ 1.3) that prescriptions (vidhis)32 are the only Vedic sentences which yield knowledge. The absence of a human author and the prescriptive character of Vedic statements jointly invalidate the charge of falsity. On one side, the ignorance, or the will to mislead, of an author could jeopardize the validity of Veda. On the other, a prescriptive statement does not stand in need of an external verification, at least not in the same sense of a descriptive one. In fact the predominant aspect in a prescription is the direct link established with the hearer, rather than a description of truth about the world. Asks the Vr¢ttikåra: “On the other hand, [unlike human words,] if the [Vedic] word speaks, how [can it] be false? [In fact,] in that case we do not seek for a cognition coming from another person. ‘It speaks’ means ‘it causes to know’, ‘it is the cause of a person’s knowing’. With the word as cause, he knows33 by himself […]”.34 As the ‡Bh admits (§ 3.2), it is possible for a prescription not to be understood, but it is quite far-fetched to claim it is ‘false’. pratyakßas tu vedavacane pratyayaΔ (‡Bh ad 1.1.2, F 20.1-2). na ca ni†citam avagamyamånam ida∫ mithyå syåt (‡Bh ad 1.1.2, F 16. 25-26). 32 Interestingly, a strenuous opponent of verbal communication as a separate source of knowledge, J.N. Mohanty, agrees that “[…] †abda alone gives us knowledge of moral rules, of what one ought to or ought not to do, of vidhi and nißedha [prohibition]. I am not only saying that ought-sentences cannot be derived from is-sentences (so that perception and inferences are incapable of yielding knowledge of moral rules), but much more. [...] What I am saying, rather, is that we learn the rules only from hearing (or reading) verbal or written instructions. There is no other means of knowing them” (Mohanty 1992: 256-7). 33 budhyamånasya and avabudhyate could be read also as passive voices and translated accordingly. See Jhå’s translation: “[…] it makes that thing known, i.e. it becomes the means of that thing becoming known, […] it becomes known by itself” (Jhå 1933: 17). See also D’Sa 1980: 67. 34 atha †abde bruvati katha∫ mithyå iti? na hi tadånœm anyataΔ purußåd avagåmaΔ. ‘bravœti’ ity ucyate ‘avabodhayati, budhyamånasya nimittam bhavati’ iti. †abde cen nimittabhüte svayam avabudhyate (‡Bh ad 1.1.5 F 34.19-21). On the views ascribed here to ‡abara and the 30 31

298

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

3.4. Transmission as beginningless succession The Naiyåyikas try to establish the validity of Veda in a different way. Their view is quoted in ‡Bh ad 1.1.5: [O.] What is the [word-meaning] relation? The fact that, once the word is known, the meaning is known. This [relation], however, is artificial […]. Therefore we believe that Vedas have been composed by the same person who, before that, made the relation between words and meant entities for the sake of communication.35

The Vr¢ttikåra replies that nobody remembers such a person, and that all we know from our ancestors is that a tradition of speakers always existed:36 [R.] It is commonly seen that juniors, hearing seniors talking to each other for their own purposes, understand a directly perceivable meaning. Even those seniors, when they were children, [understood] it from other seniors, and these from others. So there is no beginning. […] There is actually nothing incongruous in what we see. What we see is that juniors learn from seniors […].37

Thus, the Vr¢ttikåra denies the possibility of justifying a convention-maker through presumption (arthåpatti).38 There is, he maintains, nothing incongruous in the data we have, and we do not need anything extra to account for them. The problem of the origin of Veda is similarly discussed. There are no living people who have ever seen an author of the Veda, nor has tradition recorded such a person. [O.] But since [the convention-maker did it] a long time ago, he ought not be object of sense perception for today’s people. V®ttikåra a few words of caution are due. ‡abara often paraphrases prescriptive statements with descriptive ones, and this suggests an unclear prescriptive/descriptive distinction on his part ( see Kataoka 1995). 35 atha sa∫bandhaΔ ka iti? – yat †abde vij∞åte ’rtho vij∞åyate. sa tu kr¢taka […]. tasmån manyåmahe kenåpi purußeña †abdånåm arthaiΔ saha sambandha∫ kr¢två sa∫vyavahartu∫ vedåΔ prañœtå iti (‡Bh ad 1.1.5, F 42. 12-15). 36 This argument based on tradition, or succession (paramparå), is one of the bases of Mœmå∫så and is linked with its commitment to empiricism. As often stressed by many scholars (e.g. Taber 1992: 205ff), this commitment is among the main characteristics of Mœmå∫så. As empiricists, Mœmå∫sakas deny creationism as well as the hypothesis of a convention-maker in the sphere of language. 37 vr¢ddhånå∫ svårthena vyavaharamånånåm upa†r¢ñvanto bålåΔ pratyakßam artha∫ pratipadyamånå dr¢ßyante. te ’pi vr¢ddhå yadå bålå åsa∫s, tadånyebhyo vr¢ddhebhyas, te ’py anyebhya iti nåsty ådir ity [...] na hi dr¢ß™e ’nupapanna∫ nåma. dr¢ß™å hi bålå vr¢ddhebhyaΔ pratipadyamånåΔ (‡Bh ad 1.1.5, F 46. 2-2; 47. 3-4). 38 The instrument of knowledge (not reducible to inference, according to Mœmå∫sakas) through which one knows that ‘if Devadatta is alive and is not at home, he must be outside’. The gist of the reasoning is that an apparent inconsistency leads to the only other possible solution (anyathånupapatti).

III. How to Produce, Construct and Legitimate a Tradition

299

[R.] Just because he [lived] a long time ago, it is not that he should not be remembered. Nor this type of forgetfulness can be likened to that of a well or garden in the Himålaya; in fact in such [cases] a separation of people [from them] has occurred because of remoteness of the area or disappearance of the community. But a separation of people from the usage of words and meanings never occurs. [O.] It may be so: those who only use the relation and do not care for the useless recollection of the maker [of the relation] could well forget him. [R.] It is not so. Indeed, if a person created a relation and caused [people] to communicate, he would be doubtlessly remembered at the time of communication. As a matter of fact, a meaning is settled [only] when there is agreement between the doer and the user [of the relation], not when there is disagreement. Indeed, ådaic (i.e. å, ai and au)39 would not be understood through the word vr¢ddhi by someone who communicates without [knowing] Påñini or by someone who does not accept Påñini’s terminology.40

In this passage, the Vr¢ttikåra maintains that people learn the Veda from teachers who themselves previously learnt it from other teachers, and so on (Taber 1992: 205). There is no evidence of the opposite; both personal and collective memories show that Veda has always been there and language has never been ‘created’. Thus, in the Vr¢ttikåra’s view, there is no point in postulating something that goes against common experience without gaining any advantage to account for Veda and language. Moreover, the uniformity of usage among native speakers would be difficult to justify, if language were a convention. This argument may be countering (or countered by) the one expressed in NS 2.1.56: a justification of differences among languages is as problematic as that of uniformity within a single language. Like in this region the word ‘cow’ [refers to] an [animal] endowed with dew-lap, etc., so it is in every [region], even the most remote. How could have many relation-makers met [to establish that]? A single one could not [have spread the same relation in such a huge area]. Therefore there is no relation-maker.41 39 The Vr¢ttikåra here refers to Påñini’s sütra vr¢ddhir ådaic (Åß™ådhyåyœ, 1.1.1), which could not be understood if we would not know Påñini’s rules. Hence, wherever there is a convention, the convention-doer must be remembered, otherwise it cannot be deciphered. 40 nanu ciravr¢ttatvåt pratyakßasya avißayo bhaved idånœntanånåm. – na hi ciravr¢ttaΔ san na smaryeta. na ca hœmavadådißu küpåråmådivad asmaraña∫ bhavitum arhati. purußaviyogo hi teßu bhavati de†otsådena kulotsådena vå. na tu †abdårthavyavahåraviyogaΔ purußåñåm asti. – syåd etat: sa∫bandhamåtravyavahåriño nißprayojana∫ kartr¢smarañam anådriyamåñåΔ vismareyur iti. – tan na. yadi hi purußaΔ kr¢två sa∫bandham vyavahårayet, vyavahårakåle ’va†ya∫ smartavyo bhavet. sa∫pratipattau hi kartr¢vyavahartror arthaΔ sidhyati, na vipratipattau. na hi vr¢ddhi†abdena apåñiner vyavahårata ådaicaΔ pratiyeran påñinikr¢tim ananumanyamånasya vå (‡Bh ad 1.1.5, F 42.20-44.6). 41 yathå asmin de†e såsnådimati go†abdaΔ, eva∫ sarveßu durgameßv api. bahavaΔ sambandhårah katha∫ sa∫ga∫syante. eko ’pi na †aknuyåt. ato nåsti sambandhå (‡Bh ad 1.1.5, F 46.12-14).

300

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Finally, the Vr¢ttikåra reports an alternative argument against the convention theory. Interestingly enough, it is based on the interpretation of autpattika as ‘fixed’: Indeed, there was no time when a relation was not there, when some words were not connected with a meaning. Why? Because ‘relationmaking’ just does not make sense. Surely whoever made the relation must have done it by means of words. Who made [the relation] of those [words] through which [that relation] was made? Was it [also] made by someone else through [words]? And that by whom? There would be no end. Thus, beyond any doubt, we must admit some words whose relation has not been made by a relation-maker and which are established through seniors’ usage. [But] if there are [words] established through seniors’ usage, then a relation-maker is not needed.42

To sum up, ‡abara confirms that Veda’s validity is limited to prescriptions and to the sphere of dharma. He defines the word-meaning relation as fixed (nitya) and stresses its independence from human convention. Through this relation, †abda directly conveys the knowledge of its meaning, and a beginningless succession of language users and Veda reciters corroborates this fact. 4. Våtsyåyana and Uddyotakara Våtsyåyana (Pakßilasvåmin) is the author of the oldest extant commentary on NS, the Nyåyabhåßya (NBh). Some (Ingalls) propose a collocation as early as in the III century A.D., while Oberhammer and others argue for the V century A.D. The latter view is currently more popular (see Potter 1977: 239), so in the present paper we abide by it and put Våtsyåyana after ‡abara. Uddyotakara, the author of the Nyåyavårttika (NV), is unanimously assigned to the VI century. 4.1. †abda as testimony Våtsyåyana defines the reliable speaker as follows: Reliable is the speaker (upadeß™®) who has directly experienced the essential quality (dharma) [of things], and is moved by the desire to describe [things] as they are or they are not. […] The definition applies all the same to seers (r¢ßi), people of Åryavarta and barbarians.43 42 na hi sa∫bandhavyatiriktaΔ ka†cit kålo ’sti, yasmin na ka†cid api †abdaΔ kenacid arthena sa∫baddha åsœt. –katham?– sa∫bandhakriyaiva hi nopapadyate. ava†yam anena sa∫bandha∫ kurvatå kenacic chabdena kartavyaΔ. yena kriyeta, tasya kena kr¢taΔ? athånyena kenacit kr¢taΔ, tasya keneti, tasya keneti naivåvatiß™hate. tasmåd ava†yam anena sa∫bandha∫ kurvatå akr¢tasa∫bandhåΔ kecana †abdå vr¢ddhavyavahårasiddhå abhyupagantavyåΔ. asti ced vyavahårasiddhiΔ, na niyogataΔ sa∫bandhrå bhavitavyam (‡Bh ad 1.1.5, F 46.17-47.2). 43 åptaΔ khalu såkßåtkr¢tadharmå yathådr¢ß™asyådr¢ß™asya cikhyåpayißayå prayukta upadeß™å [...] r¢ßyåryamlecchånå∫ samåna∫ lakßañam (NBh ad 1.1.7).

III. How to Produce, Construct and Legitimate a Tradition

301

As for the subdivision of speech –speech related to perceivable and unperceivable objects–, Våtsyåyana says that this subdivision entails a partition between common language and language of the seers.44 According to him, the two types of objects are explicitly mentioned by Gautama because one should not think that the statement of a reliable person is valid only for perceivable objects, which the speaker perceives through sense perception, but also for unperceivable objects, which he knows through inference. Admitting inference as a root source of reliability in the speaker is necessary to account for the reliability in common speakers like us, adds Våcaspati. In fact, divine seers can directly perceive also objects not directly perceivable by us. 4.2. Reasons for the validity of testimony. The Vai†eßikas dispute the distinct character of inference and testimony (see § 2.2). Våtsyåyana explicitly points at the speaker’s reliability as the distinguishing factor: Heaven, celestial nymphs, [mythical regions such as] the Northern Kurus, the Seven Islands, the Ocean, etc.: the knowledge of such directly unperceivable objects does not come from words alone. From what, then? Thorough knowledge comes from words uttered by reliable speakers: otherwise there is no thorough knowledge. On the contrary this does not apply to inference.45

Vai†eßikas maintain that inference and testimony share the indistinct character of being based upon a relation. Introducing a principle which will come of use in the debate with Mœmå∫sakas, Våtsyåyana explains Gautama’s denial of a relation: Again, as for ‘because of relation’ (NS 2.1.51), there is one word-meaning relation which we accept, and another which we reject. […] The relation among word and meaning as a relation of contact46 is rejected.47

As far as rejection of the naive idea of a word-meaning relation of contact, Våtsyåyana and the Vr¢ttikåra (§ 3.1) show no differences of sort. As we shall see, however, they disagree on the type of relation to be accepted. Uddyotakara rephrases the word-meaning relation accepted by Naiyåyikas, and criticizes the tenet of a natural (svabhåvika) relation, thus entering the Naiyåyika vs Mœmå∫saka debate: […] evam r¢ßilaukikavåkyånå∫ vibhåga iti (NBh ad 1.1.8). svargaΔ apsarasaΔ uttaråΔ kuravaΔ sapta dvœpåΔ samudra ity evam åder apratyakßasyårthasya na †abdamåtråt pratyayaΔ, ki∫ tarhi? åptair ayam uktaΔ †abda ity ataΔ sampratyayaΔ viparyayeña sampratyayåbhåvåt. na tv evam anumånam iti (NBh ad 2.1.52). 46 I render pråpti with ‘contact’ following Våcaspati Mi†ra’s commentary; Chattopadhyaya and Gangopadhyaya (1968) translate it as ‘natural relation’. 47 yat punar ida∫ sambandhåc ceti, asti ca †abdårthayoΔ sambandho ’nuj∞åtaΔ, asti ca pratißiddhaΔ. […] pråptilakßañas tu †abdårthayoΔ sambandhaΔ pratißiddhaΔ (NBh ad 2.1.52). 44 45

302

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

We agree on the signified-signifier relation: ‘this word has this signified meaning’. But we refute that word-meaning relation which is accepted by someone as natural.48

One of the main Mœmå∫saka arguments is the commonly felt fixedness of the word-meaning relation, without which mutual understanding among speakers could not be secured. Våtsyåyana explains the Mœmå∫saka position, hinted at in NS 2.1.54, as follows: Since we witness a fixed pattern in comprehension of a [given] meaning from a [given] word, we infer that the word-meaning relation is the cause of this fixed pattern. In fact, if there were no relation, there would be the consequence of comprehending any meaning from any word. Therefore the relation cannot be ruled out.49

The Naiyåyika answer is that the only existing relation is a conventional one. Våtsyåyana explains what convention means: The fixed pattern of word and meaning is not produced by a [natural] relation. From what then? It is produced by convention. […] But what is this ‘convention’? It is the mandate of a restriction of signifier and signified: ‘by this word this meaning is to be signified’. When this [mandate] is respected the understanding of a meaning from a word takes place. Otherwise, indeed, even upon hearing a word there is no understanding. Also the supporters of [a natural] relation50 cannot deny this [fact]. And common people use convention by taking it from other users; to preserve this convention there is grammar.51

Noticeably, Naiyåyikas agree with Mœmå∫sakas on the role of a tradition of language users (vr¢ddhavyavahåra) in the transmission of linguistic competence, but conceive a prime cause behind it which Mœmå∫ßakas do not. There is another argument against a natural word-meaning relation: differences of idiom according to social belonging (see NS 2.1.55-6, § 2.2). Våtsyåyana states: The understanding of meaning from words is conventional and not naturally intrinsic (svåbhåvika). The arbitrary use of words to convey mean48 våcyavåcakabhåvalakßañaΔ sambandho ’nuj∞åtaΔ. asya †abdasyåyam artho våcya iti. yaΔ punar aya∫ svåbhåvikaΔ †abdårthayoΔ sambandha iti kai†cid abhyupagamyate sa pratißiddha iti (NV ad 2.1.52). 49 †abdåd arthapratyayasya vyavasthådar†anåd anumœyate asti †abdårthasambandho vyavasthåkårañam. asambandhe hi †abdamåtråd arthamåtre pratyayaprasaõgaΔ. tasmåd apratißedhaΔ sambandhasyeti (NBh ad 2.1.54). 50 Våcaspati glosses ‘supporters of relation’ (sambandhavådinaΔ) with ‘Mœmå∫sakas or Vaiyåkarañas’ (NVT ad NS 2.1.55). 51 na sambandhakårita∫ †abdårthavyavasthånam. ki∫ tarhi? samayakåritam. […] kaΔ punar aya∫ samayaΔ? asya †abdasyedam arthajåtam abhidheyam iti abhidhånåbhidheyaniyamaniyogaΔ. tasminn upayukte †abdåd arthasampratyayo bhavati. viparyaye hi †abda†ravañe ’pi pratyayåbhåvaΔ. sambandhavådino ’pi cåyam avarjanœya iti. [...] prayujyamånagrahañåc ca samayopayogo laukikånåm. samayaparipålanårtha∫ ceda∫ [...] vyåkarañam. (NBh ad 2.1.55).

III. How to Produce, Construct and Legitimate a Tradition

303

ings applies to seers, inhabitants of Åryavarta, or barbarians. Indeed if a word naturally conveyed its meaning, there would be no arbitrariness, just like the cognition of a form, produced by bright light, does not vary according to specific classes of people.52

Uddyotakara makes clear that the difference hinted at in NS 2.1.56 has a geographical value: By the expression ‘specific classes of people’ the region is denoted.53

4.3. Reasons for the validity of Veda In NS 2.1.68 the validity of Veda was inferentially established. The inferential mark (hetu) given there was the reliability of the speaker, and the illustration (dr¢ß™ånta) was the validity of incantations and Åyurveda. But are these hetu and d®ß™ånta solidly grounded? Våtsyåyana, with a soteriological flavor, enumerates three qualities of a reliable speaker: direct experience of the essence (dharma) of things; compassion towards living beings; desire to speak of things for what they are.54 As for the validity of incantations and Åyurveda, whatever is taught by Åyurveda –by doing so and so you can achieve the desirable outcomes, by giving up that and that you can avoid the undesirable outcomes– is consistent, truthful and infallible for one who follows it. And in the usage of the words of incantations meant to contrast poison, ghosts and thunderbolt there is consistency with the outcomes, and this proves [their] validity.55

The inference, as expressed by Gautama, is that Veda is an instrument of knowledge, because it has reliable speakers, just like Åyurveda is a means of knowledge because it has reliable speakers. Våtsyåyana seems to understand Gautama’s argument as an implicitly larger argument, akin to an hypothetical syllogism (if p then q; if q then r; therefore, if p then r) which he develops accordingly: (p) The Åyurveda, which deals with perceivable objects, is an instrument of knowledge; because it is the statement of a reliable speaker. 52 samayikaΔ †abdåd arthasampratyayo na svåbhåvikaΔ. r¢ßyåryamlecchånå∫ yathåkåma∫ †abdaviniyogo ’rthåpratyåyanåya pravartate. svåbhåvike hi †abdasyårthapratyåyakatve yathåkåma∫ na syåd, yathå taijasasya prakå†asya rüpapratyayahetutva∫ na jåtivi†eße vyabhicaratœti (NBh ad 2.1.56). 53 jåtivi†eßa†abdena punar de†o ’bhidhœyata iti (NV ad 2.1.56); here, according to Vacaspati, Uddyotakara means that, for instance, a stranger living in Åryavarta would conform to the usage of the Åryas. 54 ki∫ punar åptånå∫ pråmåñyam? såkßåtkr¢tadharmatå bhütadayå yathåbhütårthacikhyåpayißeti [...] evam åptopade†aΔ pramåñam evam åptåΔ pramåñam (NBh ad 2.1.68). 55 yat tad åyurvedenopadi†yate ida∫ kr¢tveß™am adhigacchati ida∫ varjayitvåniß™a∫ jahåti, tasyånuß™hœyamånasya tathåbhåvaΔ satyårthatåviparyayaΔ. mantrapadånå∫ ca vißabhütå†anipratißedhårthånå∫ prayoge ’rthasya tathåbhåvaΔ, etat pråmåñyam (NBh ad 2.1.68).

304

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

(q) The portion of the Veda which deals with perceivable objects is an instrument of knowledge; because it is the statement of a reliable speaker; like the Åyurveda. (r) The portion of the Veda which deals with unperceivable objects is an instrument of knowledge; because it is the statement of a reliable speaker; like the portion of the Veda which deals with perceiveable objects56

(p) and (q) share the common character of dealing with visible objects; (q) and (r) share the character of belonging to the same work, i.e. the Veda. The kinship of Åyurveda and Veda is further stressed by the character of being both spoken by reliable seers: And the [validity of the] inference owes to the sameness of seers and speakers [of Veda and Åyurveda]. Indeed the reliable seers and speakers of Vedic matter are also those of Åyurveda, etc.; thus the validity of Veda should be inferred in the same fashion of the validity of Åyurveda.57

The whole passage also shows that Våtsyåyana had in mind, as Uddyotakara confirms, sections of the Veda which deal with perceived objects and others which deal with unperceived ones. Interestingly enough, in (p), (q), (r) the inferential mark remains the reliability of the speaker, which in (p) and (q) can be cross-checked by dint of perceivable outcomes. Våtsyåyana raises some Mœmå∫saka objections, which point to the inconsistency of speakers’ reliability and to the permanence of Veda: [O.] Being the validity of Vedic statements based on [their] permanence (nitya), it is not correct to say that their validity is based on reliability of the speakers. [R.] This is wrong. The validity in the knowledge of meanings depends on the signifying capacity of words, not on [their] permanence. […] [O. ] Without permanence there cannot be signification [see § 1.1, MS 1.1.18]. [R.] No, it is not so, because there is no evidence of this [permanence] in common language. [O.] That (common language) is also permanent. [R.] No, because if this were the case, it would be uncongrous to say that wrong meanings are understood from unreliable speakers (being all words permanent, these should invariably produce valid knowledge). In day-to-day language, wherever a word is assigned to an object, it becomes a factor of knowledge owing to the power of its mandate, not because of permanence. ‘Permanence’ of Vedas is the unbroken conti56 See NBh ad 2.1.68: dr¢ß™årthenåptopade†eñåyurvedenådr¢ß™årtho vedabhågo ’numåtavyaΔ pramåñam iti åptapråmåñyasya hetoΔ samånatvåd iti. asyåpti† caikade†o gråmakåmo yajeta ity evam ådir dr¢ß™årthaΔ, tenånumåtavyam iti. 57 draß™r¢pravaktr¢såmånyåc cånumånam. ya evåptå vedårthånå∫ draß™åraΔ pravaktåra† ca ta evåyurvedaprabhr¢tœnå∫ ity åyurvedapråmåñyavad vedapråmåñyam anumåtavyam iti (NBh ad 2.1.68).

III. How to Produce, Construct and Legitimate a Tradition

305

nuity of tradition in sequential past and future ages, and [their] validity is due to the worthiness of a reliable speaker. And this applies also to day-to-day language.58

If Veda is not permanent it must be the product of an author. But the Mœmå∫saka might ask: where is the proof of the existence of such a person? Uddyotakara retorts that, whatever the case, this is certainly not an evidence in favor of permanence: in fact, proofs of permanence are also hardly available.59 Nonetheless, the eternity of Veda is a wide-spread and commonly accepted concept. How can this fact be accounted for? [O.] What about the common conviction that Vedas are eternal? [R.] The reason [behind it] is the non-interruption of tradition. These very Vedic statements were handed down throughout the ages without interruption of tradition, and in reference to this [fact] people use words such as ‘Vedas are permanent’, similar to [commonly used expressions such as] ‘permanent mountains, permanent rivers’, etc. This applies all the same to statements such as those of Manu [for which there is record of an author]. [O.] Why talking of permanence that depends on non-interruption of tradition, and not just of permanence? [R.] Because of non-falsity of the above-said evidences in favor of impermanence: the evidences by me implemented in relation to impermanence are not disproved, and since they are not false, being there impermanence, ‘permanence’ is figuratively used in the sense of ‘continuity of tradition’.60

Noticeably, in these last lines of Våtsyåyana and Uddyotakara the Mœmå∫saka objector closely mirrors the thesis of the Vr¢ttikåra, who argued in favor of a beginningless tradition (see § 3.4). 5. Kumårila Kumårila Bha™™a, the upholder of the Bhå™™a school of Mœmå∫så, can be assigned with relative certainty to the VII century, as proven by several passages found in the works of Dharmakœrti, ‡aõkara, Mañ∂ana Mi†ra, etc. 58 nityatvåd vedavåkyånå∫ pramåñatve tatpråmåñyam åptapråmåñyåd ity ayuktam. †abdasya våcakatvåd arthapratipattau pramåñatva∫ na nityatvåt. […] nånityatve våcakatvam iti cen na, laukikeßv adar†anåt. te ’pi nityå iti cen na, anåptopade†åd arthavisa∫vådo ’nupapannaΔ [...] yatrårthe nåmadheya†abdo niyujyate loke tasya niyogasåmarthyåt pratyåyako bhavati, na nityatvåt. manvantarayugåntareßu cåtœtånågateßu sampradåyåbhyåsaprayogåvicchedo vedånå∫ nityatvam, åptapråmåñyåc ca pråmåñyam. laukikeßu †abdeßu caitat samånam iti (NBh ad 2.1.68). 59 paurußeyatvam asiddha∫, nityatvåd iti cet, atha manyase nityåni vedavåkyåni nityatvåc caißå∫ pråmåñyam tasmåt paurußeyatvam asiddham, na, asiddhatvåt (NV ad 2.1.68). 60 atha yo ’yam abhilåpo nityå vedå iti, sa katham? sampradåyasyåvicchedåt. tany eva vedavåkyåni manvantaracaturyugåntareßu sampradåyåbhyåsåvicchedena pravartante, tadapekßayå laukikåΔ †abdån prayu∞jate nityå vedå iti, yathå nityåΔ parvatåΔ nityåΔ sarita iti, tan manvådivåkyeßu samånam. kuta eva∫ sampradåyåvicchedåt nityatva∫ na punar nityatvåd eva? anityatvoktapramåñåvighåtåt. yåni mayånityatve pramåñåny upanyaståni tåni na vihanyante, teßåm avighåtåd anityatve sati sampradåyåvicchedåt nityatvopacåra iti (NV ad 2.1.68).

306

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Concerning Veda’s validity, Kumårila follows in the footsteps of Jaimini and ‡abara. Nevertheless one should keep in mind that also the antagonistic school of Pråbhåkaramœ∫å∫så, whose interpretation of the tradition sharply diverges from Kumårila’s, claims faithfulness to MS and ‡Bh. We have seen how the Vr¢ttikåra put emphasis on the non-human61 (apaurußeya) character of the word-meaning relation. Kumårila further develops the issue, discarding the very possibility of an extra-ordinary human being who could have created language and known dharma. Even the ‘beginningless transmission’ argument (see §§ 1.4, 3.4) is not directly employed to establish Veda’s validity but only to prove the absence of its author. The main positive argument adopted by Kumårila in favor of Veda’s validity, instead, is the ‘intrinsic validity’ theory of knowledge. 5.1. Veda is not human-made (apaurußeyatva)62 In his systematizing effort, Kumårila dismisses ‡abara’s main argument on the infallibility of Vedic knowledge as ‘ineffective’ (jåti).63 In fact propositions like “if the word speaks, how [can it] be false?” (§ 3.3) or “It is contradictory to say ‘spoken’ and ‘false’’’,64 could justify, if consistently applied, even Buddha’s words: There (in the ‡Bh) the contradiction [between ‘spoken’ and ‘false’] would apply to Buddha’s statements as well; because also from those [statements] cognitions do arise.65

Therefore, if on one side he develops a more sophisticated justification of the Veda’s validity, on the other he rejects the possibility of an omniscient human being. 61 a-paurußeya means ‘non-human’, ‘non-personal’ and may refer to both human beings and personal, though superhuman, beings, e.g. Gods. 62 On how should ‘authorlessness’ of the Veda be understood, views diverge. J. N. Mohanty (1992: 259) states that “the concept of apaurußeyatva […] is […] the concept of the primacy and autonomy of the eminent text over the subjective intentions of the author. It is also the concept of the role the eminent texts such as the †ruti play in delimiting the horizon within which our tradition has understood itself and, within the tradition, we have understood ourselves” and (1992: 258): “The words are prior to experiences […] The implication is that the same [mystical] experiences would be made to tell a different story, in another tradition, where a different set of words fulfil that foundational role”. Somehow similar is V.K. Chari’s argument: “Apaurußeyatva […] can only mean that because of the self-explanatory character of the verbal sign itself, the meaning of the text is not dependent upon its author” (Chari 1992: 104). Finally, Purußottama Bilimoria (1989: 149) lists seven possible meanings of apaurußeyatva. On the other side, J. Bronkhorst (especially Bronkhorst 1998) argues for apaurußeya as meaning ‘literally’ without author and relates it to the claim of the Veda being ‘literally’ beginningless. 63 This weak point was already noticed by Våtsyåyana (see § 4.3). 64 vipratißiddham idam abhidhœyate ‘bravœti ca vitatha∫ ca’ (‡Bh ad 1.1.2, F 16.18). 65 tatra vipratißiddhatva∫ buddhavåkye ’pi yujyate Ù tato ’pi pratyayotpattes (‡V codanå 32a-c).

III. How to Produce, Construct and Legitimate a Tradition

307

Following the Bhåßya arguments (§ 3.2, ‡Bh ad 1.1.2, F 18. 3-15.), he asserts that the Veda cannot be proven to be false because of an author’s defects, as there is no author:66 Therefore in the case of [statements of individuals such as Buddha, etc.], the falsity is justified by their human origin. This character [of falsity] does not apply to the Veda, because [in its case] there is no author. Since it is not hampered by any author’s notion, the meaning [of a Vedic sentence] is comprehended directly through the meanings [of its words]; consequently it does not depend on [any author’s] cognition nor can it be false.67

But what is the evidence that an author of the Veda never existed? Kumårila’s answer can be reduced to three main arguments: 1. there is no need to postulate anything unseen (see § 1.4; n. 34); 2. there is no need to postulate the existence, in the past or in the future, of something different from what we experience today (this point is developed from an analogous argumentation of the Vr¢ttikåra; see §§ 3.4, 5.2); 3. as the Vr¢ttikåra asserted (§ 3.4), if there were an author of the Veda, we would have heard about such an illustrious personality. Hence, the non-existence of an author is proven by absence (abhåvapramåña), considered an independent instrument of knowledge by Bhå™™amœmå∫sakas.68 The basic idea behind the concept of abhåva is that the absence of something is the positive content of a distinct type of cognition, ‘there is absence of x in y’. Finally, as far as words in common language are concerned, Kumårila maintains69 that they are permanent (nitya) on the basis of presumption (arthåpatti), since communication would otherwise be unexplainable (see §§ 1.1, 3.1). 66 What Hans-Georg Gadamer (1960: 299) called the Vorgriff der Vollkommenheit (‘the presupposal of perfection’), once applied to the Veda (and it is applied to it by almost the whole Indian community), can never be doubted, since no chances of an historical (Veda is said to be timeless) or personal (since it has no author) understanding are left. 67 ato ’tra pu∫nimittatvåd upapannå mr¢ßårthatå Ù na tu syåt tatsvabhåvatva∫ vede vaktur abhåvataΔ ÙÙ tadbuddhyantarayo nåstœty artho ’rthai† ca pratœyate Ù ato na j∞ånapürvatvam apekßya∫ nåyathårthatå ÙÙ (‡V codanå 169-170). I adopted the reading of U∫veka Bha™™a, the earliest commentator, instead of tadbuddhyantarayor (perhaps a scribal error due to the infrequent term antaraya, changed into the dual antarayor) found in the published versions of ‡V. Despite this difficulty, the meaning of the verse is quite clear thanks to both the ‡V context and the comments, who all seem to have read tadbuddhyantarayo. U∫veka Bha™™a (‡VV): ato vaktur abhåvena ca vaktr¢j∞ånaparatvåbhåvåt padårthapratipådanadvåreña bhåvanåyam eva tåtparyam ity åha –tatbuddhyantaraya iti. Pårthasårathimi†ra (NR): vaktr¢buddhyantaråyåbhåvåt padarthair eva såkßåd våkyårtha pratœyata iti. Sucaritami†ra (K): vaktr¢buddhyantarayor vyavadhånam api vede nåstœti padårthair eva kevalair nityanirdoßair våkyårthaΔ pratœyata ity åha tadbuddhœti. 68 The Bhå™™amœmå∫sakas accept six instruments of knowledge: sense perception (pratyakßa), inference (anumåna), verbal communication (†abda), analogy (upamåna), presumption (arthåpatti), absence (abhåva). For the means of knowledge accepted by Naiyåyikas see n. 82. 69 ‡V †abdanityatå, especially 236-7.

308

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

5.2. Veda is an instrument of knowledge only in relation to dharma Like Jaimini, ‡abara and the Vr¢ttikåra, also Kumårila believes that Veda is an instrument of knowledge only for objects beyond human faculties, e.g. dharma. No cognition based on sense perception could ever be invalidated by Vedic utterances. He even declares that for objects beyond its sphere of concern, even the Veda [can] be false.70

In other words, whenever the empirical reality is at stake, Kumårila rejects the possibility of a conflict between Veda and the other sources of knowledge. Only the Veda, vice versa, can bestow knowledge of dharma, so both the likelihood of inference playing a role in the ascertainment of unperceivable objects and the idea that human beings can intellectually perceive dharma are ruled out (see § 4.1). In order to exclude the possibility of different conditions for human beings in the past or in the future, he says: A class of objects which is presently known through its [specific] instrument of knowledge was known through that same instrument of knowledge also in the past. […] As far as future objects [such as dharma] are concerned, there is not the slightest room for sense perception, nor for inference, for want of an inferential mark.71

5.3. An omniscient author does not exist The Naiyåyika and Buddhist ideas72 that the validity of sacred texts has to be founded on their author’s reliability is criticized by Kumårila. He rejects the inference (a) the author’s assertions in the matter of dharma are true; because they are his assertions, like his true assertions in many matters of ordinary perception (see § 4.3).

Such an argument is logically vulnerable, as it could be easily used against its propounder. In fact, a disputant would argue at once, having undoubtedly the same grounds [at his disposal] against you: ‘My assertion that ‘Buddha is not omniscient’ is true; because it is my assertion, like my [true] assertion ‘fire is hot and bright.73 70

svavyåpåråtirikte ’rthe vedasyåpi mr¢ßårthatå (‡V codanå 173ab).

71 yajjåtœyaiΔ pramåñais tu yajjåtœyårthadar†anam Ù bhaved idånœ∫ lokasya tathå kålåntare

’py abhüt ÙÙ […] bhavißyati na dr¢ß™a∫ ca pratyakßasya manåg api Ù såmarthyam nånumånåder liõgådirahite kvacit ÙÙ (‡V codanå 113, 115). 72 Eli Franco highlights the link between the Naiyåyika and the Buddhist theses (Franco 1997: first chapter). 73 vaded eva∫ ca yo nåma vådœ prathamasaõgataΔ Ù tasyåpi hetuΔ syåd eßa bhavanta∫ praty asa∫†ayam ÙÙ buddhådœnåm asårvaj∞am iti satya∫ vaco mama Ù maduktatvåd yathaivågnir ußño bhåsvara ity api ÙÙ (‡V codanå 129-130).

III. How to Produce, Construct and Legitimate a Tradition

309

Kumårila’s sarcasm, however, would miss the mark if in (a) the author were qualified as ‘omniscient’. In fact, Buddhists and later Naiyåyikas such as Jayanta (§ 6.1) claim that the one who said such true things (e.g. the four noble truths, or the Åyurveda) could not but be omniscient, hence his words are to be held true even for unperceivable matters. Therefore, Kumårila aims at disproving the logical possibility of an omniscient human being, and consequently he proposes many circumstantial criticisms which presuppose a fundamental reduction: the theorization of an omniscient being is the theorization of something unseen. According to Kumårila the proposal of an omniscient author is nothing more than fiction, an appeal to imaginary entities. As for an omniscient person, at present we do not see him; nor it is possible to hypothesize –while it is easy to disprove– that [he] ever existed.74

Moreover any human statement could be invalidated whenever defects are found to be present, or qualities found to be lacking, in the speaker, and in such cases it would not yield knowledge anymore. Any human being (including Buddha and seers) who speaks about supernatural subjects like dharma, which cannot be known through human faculties, cannot be reliable. Hence, on extra-sensorial matters only an autonomous instrument of knowledge, i.e. Veda, can serve the purpose. In sum, Kumårila aims at securing the autonomy (§ 1.1) of Veda from other means of knowledge. As a matter of fact, if the validity of Veda were in need of an external justification (be it an author or an empirical test), it would depend on it and become only a ‘secondorder’ instrument of knowledge: just as the validity [of assertions regarding ordinary matters] is due to sense-discrimination, etc., so also [the validity of assertions] in the case of matters of faith [would depend on sense –discrimination, etc.] and it would never be autonomous.75

Kumårila dismisses also the possibility of founding the reliability of an omniscient author on the sacred text itself, which would end up in a petitio principii: Nor can an omniscient being [be known] through the sacred text (ågama); for in that case there would be mutual interdependence.76

If instead we assume the autonomous validity of Veda, then the postulation of an author becomes unnecessary. 74 sarvaj∞o dr¢†yate tåvan nedånœm asmadådibhiΔ Ù niråkarañavac chakyå na cåsœd iti kalpanå ÙÙ(‡V codanå 117). 75 yathaivåtrendriyådibhyaΔ paricchedåt pramåñatå Ù †raddheye ’pi tathaiva syån na svåtantryeña labhyate (‡V codanå 123). 76 na cågamena sarvaj∞aΔ tadœye ’nyonyasa∫†rayåt Ù (‡V codanå 118ab).

310

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Thus, Naiyåyikas77 are criticized because, by resorting to an author, they implicitly state that Veda is not an autonomous instrument of knowledge: therefore, since validity [of the Veda] does not depend on an author, the worship of its [author] is wrong. Because one could address him only with prayers if he assumes that the Veda is [in itself] devoid of epistemic validity.78

5.4. Transmission as beginningless succession As ‡abara, also Kumårila maintains that every epistemic enterprise must be based on the world as it is actually experienced, and that an alternative state of affairs needs not to be imagined: every [student’s] learning of the Veda was preceded by the learning of the Veda by his teacher, because it is called learning of the Veda [which is necessarily done under a teacher], like today’s learning [of the Veda].79

Therefore the postulate of a time when the Veda was not transmitted by a teacher but composed ex nihilo is unwarranted. An objector may say that the teacher-student transmission applies also to authored compositions such as the Mahåbhårata. But in that case, replies Kumårila, the existence of an author is well recorded, while an author of the Veda is not remembered. Neither do arthavådas referring to Prajåpati may be said to prove his authorship of the Veda, since they are arthavådas and hence ineffective cognitive tools. Therefore, a tradition based on them is nothing but a tradition of blind people transmitting to each other opinions about colors. This shows that for Kumårila ‘tradition’ is not necessarily a value, because when the foundation of a tradition turns out to be flawed, it is not anymore valuable. In this connection it is worth mentioning that Kumårila rejects (TV ad 1.3.4., v. 269) the argument based on consensus of ‘many/great people’80 (mahåjanaparigraha, akin to consensus gentium), that is the deduction of the validity of Veda etc. from its acceptance by many people. The validity of sacred texts, he maintains, must be independently established, in order to avoid the risk of relativism. 5.5. Validity is intrinsic Kumårila’s positive argument for Veda as an instrument of knowledge is the intrinsic validity (svataΔ pråmånya) theory.81 This is most 77

Explicitely identified by Pårthasårathi in NR ad ‡V codanå 69. ato vaktranadhœnatvåt pråmåñye tadupåsanam Ù na yuktam apramåñatve kalpye tatprårthanå bhavet ÙÙ (‡V codanå 69). 79 vedasyådhyayana∫ sarva∫ gurvadhyayanapürvakam Ù vedådhyayanavåcyatvåd adhunådhyayana∫ yathå ÙÙ (‡V våkya 366). 80 The two aspects are mostly intertwined, see Chemparathy 1983: 59-63. 81 Exemplarily discussed by Taber 1992 and (following Taber) by Arnold 2001. 78

III. How to Produce, Construct and Legitimate a Tradition

311

likely a development of ‡abara’s concept of Veda-originated cognition as self-evident, just like sense perception (see § 3.2). The theory is defended by Kumårila from the attacks of Buddhists and Naiyåyikas. The basic tenet is that a cognition can be denied only if and when defeated by a subsequent one, but unless that happens, and until then, it must be regarded as wholly valid. This position is most likely rooted in the Mœmå∫saka commitment to common experience, in which no one waits for further justification before acting (this would lead to a complete paralysis). Moreover, if no cognition were admitted as valid without the verification of a second one, also this latter verifying cognition, qua cognition, would need a verification, and so on. This regressus ad infinitum could be stopped, as Buddhists and Naiyåyikas claim, by postulating that some cognitions, e.g. empirical tests, do not need further verification. But if a limit in the infinite series of verifications must be anyway postulated, why not accepting the very first cognition as valid? According to Kumårila, every cognition must be assumed as valid, and can be invalidated only if, thereafter, its causes are found to be defective or the piece of information produced is proven to be false. As far as Veda is concerned, it has no author and should be considered an instrument of knowledge only in relation to dharma and other extra-sensorial matters. The absence of an author precludes the possibility of finding any defect in the Veda and, since no other instrument of knowledge could invalidate knowledge of dharma achieved through the Veda, its intrinsic validity can never be overthrown. 6. Jayanta Bha™™a The magnum opus of Jayanta (IX century) is the encyclopedic Nyåyama∞jarœ (NM), in which three books out of twelve are devoted to †abdapramåña.82 6.1. †abda as testimony The third book of NM comments on NS 1.1.7. Unlike previous commentaries on the word upade†a (statement, see §§ 2.1, 4.1) Jayanta, probably influenced by Mœmå∫så, also brings into the picture the receptive hearer’s side of the coin, and not just the productive speaker’s as previous Naiyåyikas did: What is the meaning of ‘stating’? It is ‘doing an act of denotation’. […] An act of denotation implies that a meaning produced by a hearable enti82 Naiyåyikas accept four instruments of knowledge: sense perception (pratyakßa), inference (anumåna), verbal communication (†abda), analogy (upamåna). For the instruments of knowledge accepted by the Bhåttas, see n. 68.

312

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

ty is understood. Commonly a ‘said’ or ‘denoted’ meaning is called that which conveys the content (vißayatå)83 of a corresponding cognition.84

As said above (§ 4.1), experience of the essential quality of things is the foundation of reliability. Jayanta confirms that the reliable speaker does not necessarily rely only on direct perception, but also on inference and other instruments.85 This is consistent with Jayanta’s tenet of the possibility of more instruments of knowledge concurring (pramåñasamplava) to the cognition of a same object,86 and it opposes the Mœmå∫saka principle of autonomous instruments of knowledge (see §§ 3.2, 5.3). The most significant innovation in Jayanta’s justification of the validity of Vedas is that he brings into the picture God (œ†vara), applying to him the three characteristics of reliability mentioned by Våtsyåyana (§ 4.1) and reading dharma in a moral sense: This same definition, as we heard it [from Våtsyåyana], applies to God, the reliable propounder of sacred traditions (ågama). Because, like someone who experiences directly the essential quality of things (dharma), God perceives directly [the universal] dharma; as to the desire to communicate things, it will be explained how God is compassionate; and he is a teacher (upadeß™®) because, as we will establish, he is the propounder of Vedic tradition, etc.87

The introduction of God in the epistemological discussion at Jayanta’s time is open to further social and historical research. Jayanta and the coeval Naiyåyika Bhåsarvaj∞a, who developed the discussion on similar lines, were both hailing from Kå†mœr and affiliated to ‡aiva circles. As a direct consequence, Jayanta generalizes the scriptural validity also to other extra-Vedic traditions such as ‡aiva Ågamas and Pa∞caråtras. In this connection, a few words on ‘consensus of many/great people’ (mahåjanaparigraha) are due (see also § 5.4). Jayanta discards the 83 In Matilal’s words,“the hearer’s knowledge or the cognitive episode arising in the hearer from the utterance of a sentence is said to grasp the ‘meaning’ (artha) of the sentence uttered. […] What this episode grasps has a ‘structured content’ (vi†ayatå) which we can make more intelligible by calling it the structure of a thought. When we say that a particular hearer a understands the meaning, we mean thereby that a particular hearer a has a particular ‘structured’ thought” (Matilal 1985: 417). 84 upadi†yate iti ko ’rthaΔ? abhidhånakriyå kriyate […] †rotragråhyavastukarañikå tadarthapratœtir abhidhånakriyå, ittha∫ loke vyavahåråt. ukto ’bhihita† ca sa evårtho loke vyapadi†yate yas tu tathåvidhapratœtivißayatå∫ pratipannaΔ. (NM 3, vol. 1, p. 218). 85 na tu pratyakßeñaiva grahañam iti niyamaΔ, anumånådini†citårthopade†ino ’py åptatvånapåyåt (ibidem p. 219). 86 On pramåñasamplava, see NM 1, vol. 1, pp. 48-52. Against this concept, see §§ 1.1 and 3.2. 87 tatprañetur åptasye†varasya yathå†rutam eveda∫ lakßañam såkßåtkr¢tadharmeva dharmasye†varapratyakßagocaratvåt, cikhyåpayißayå prayukta iti kåruñika eva bhagavån iti vakßyate. upadeß™å ca vedådyågamånå∫ tatprañœtatvasya samarthayißyamåñatvåd iti (NM 3, vol. 1, pp. 219-220).

III. How to Produce, Construct and Legitimate a Tradition

313

possibility of including Buddhist scriptures among genuine sacred texts by appealing to the fact that ‘many/great people’ do not accept them. ‘Many/great people’ are defined as those who live in the country of the Åryas and who abide by the system of four social classes and four stages of life (cåturvarñya∫ cåturå†ramyam). The validity of ‡aiva Ågamas and Vaißñava Pa∞caråtras is accepted by them because these scriptures belong to the greater Vedic tradition. Buddhist texts, however, are out of this sphere and therefore not considered instruments of knowledge.88 6.2. Reasons for the validity of testimony 6.2.1. Validity: intrinsic or extrinsic? Before an analysis of the reasons that lead to believe in God’s existence, Jayanta discusses the Mœmå∫saka concept of intrinsic validation/extrinsic invalidation. In reference to NS 1.1.8, he begins with a fundamental distinction of two main classes of knowledge, according to the kind of object (perceivable/unperceivable) involved, confining the issue of intrinsic/extrinsic validation to religious knowledge alone: Regarding instruments of knowledge such as direct perception, which have perceivable objects, normal activities are carried on without ascertaining the validity of knowledge. […] But in relation to unperceiveable objects, without the ascertainment of the validity of Vedic rituals –whose difficult goals are achieved [only] with the employment of enormous wealth, etc.– it would be improper for a wise person to undertake them. Thus, the ascertainment of its (of the Veda) validity must necessarily be done.89

He maintains that validity cannot be intrinsic because it is commonly experienced as a development of a previous condition of doubt.90 Doubt effectively explains the knowing process of everybody, despite the bad image it sometimes has in the eyes of tradition.91 However, Jayanta clarifies that Naiyåyikas are not obsessed with the concept of doubt. Their position, instead, is that it is not possible to determinate the validity of knowledge at the time of its production, and with ‘doubt’ they mean the state of epoché deriving from that incapacity to validate knowledge.92 88 See

NM 4, vol. 1, pp. 379-382. Chemparathy 1983: 59-72 treats the subject at length. pratyakßådißu dr¢ß™årtheßu pramåñeßu pråmåñyani†cayam antareñaiva vyavahårasiddhes […] adr¢ß™e tu vißaye vaidikeßv agañitadraviñavitarañådikle†asådhyeßu karmasu tatpråmåñyåvadhårañam antareña prekßåvatå∫ pravarttanam anucitam iti tasya pråmåñyani†cayo ’va†yakarttavyaΔ (NM 3, vol. 1, p. 240). 90 Doubt is an important concept already in NS (1.1.1), where it is listed among the sixteen categories. 91 Such as ‘the doubtful one meets with destruction’ (sa∫†ayåtmå vina†yati) in Bhagavadgœtå, 4.20, comments Cakradhara (NM 3, vol. 1, p. 243). 92 Ibidem, p. 241-243. 89

314

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Jayanta explains that we can talk of knowledge in two senses: ‘operational’ (vyåpåra) and ‘resulting’ (phala). The ascertainment of validity never occurs during the operational stage, but only later, when the practical efficacy (pravr¢ttisåmarthya) of the resulting knowledge can be appreciated.93 Kumårila’s charge of uselessness and of regressus ad infinitum (see § 5.4) may stand for common knowledge related to perceivable objects, but Jayanta makes clear that this is not always the case. In fact validation of knowledge has a role in the Åyurveda argument (§§ 2.2, 4.3) even when the context pertains to perceivable objects. In response to Kumårila, who argued that the author’s qualities cannot secure the validity of Vedas (see § 5.3), Jayanta says that the negative argument of the absence of defects does not suffice, unless we are ready to accept the validity of every passage of the Vedas, including those which appear to be fictional. In such cases also, being the object unperceivable, their truth or falsity cannot be determined. Moreover, it is wrong to claim: “the Veda must be autonomous because only in absence of an author there can be absence of defects”. In fact this argument can be reversed to say that if there were no author, due to the absence of his qualities the validity of the Veda would also be absent.94

Then, the Mœmå∫sakas may ask: “Tell us who is the author of the Veda”. Jayanta answers: The author of the Veda, indeed, is not an ordinary person, but the supreme lord, skilful in the craft of universal creation.95

Some Mœmå∫sakas may accept God’s existence, as this does not undermine their theory. But what evidences are there that God authored the Veda? In fact, they say, in the context of words, wordmeaning relations, or Veda itself, the hypothesis of an author is equally far-fetched. This is how Jayanta sums up the Mœmå∫saka position: We call ‘word’ a group of phonemic sounds manifested in sequence and, if these sounds are indestructible, how is it possible that they were created by God? And also the [word-meaning] relation is not his production; indeed it has its own natural power. And the signifying power of words is fixed (nitya), just like heat in fire. Furthermore, Vedic compositions are not made by a person, because they are altogether different from compositions such as poetry.96 93

Ibidem, p. 244-45. asati vaktari pråmåñyahetünå∫ guñånåm apy abhåvena tatpråmåñyasyåpy abhåvåt (ibidem, p. 267). 95 vedasya purußaΔ kartå na hi yådr¢†atådr¢ßaΔ. kintu trailokyanirmåñanipuñaΔ parame†varaΔ (ibidem, p. 267). 96 varñarå†iΔ kramavyaktaΔ padam ity abhidhœyate Ù varñånå∞ cåvinå†itvåt katham œ†varakåryatå ÙÙ sambandho ’pi na tatkåryaΔ sa hi †aktisvabhåvakaΔ Ù †abde våcaka†akti† ca 94

III. How to Produce, Construct and Legitimate a Tradition

315

6.2.2. Words are a creation According to Jayanta, Bhå™™amœmå∫sakas maintain that permanence (nityatå) of phonemes is proven through recognition (pratyabhij∞å) and through presumption (arthåpatti), because signification could not be explained otherwise (see §§ 1.1, 3.1, 5.1). However, the two arguments collapse if we can establish a universal property for each phoneme. When a speaker says gagana (‘sky’), how is a difference between the two g-s produced?97 If we say that there is one single g we end up with a Bhartr¢hari-type of linguistic monism, which is not acceptable to Naiyåyikas and Mœmå∫sakas. If we say that they are two different g-s we must admit that they have a same property, and that property is g-ness, the universal of the phoneme g. In summary, says Jayanta, this is the real matter: either notions of universal and particular are neglected in all instances, or, just like in the case of the universal cow-ness, the universal g-ness which is present in all the different [phonemes] g-s is to be accepted.98

Jayanta’s conclusion is that phonemes are a product, and if they are a product we must admit a creator behind them.99 If phonemes are created, also words and sentences, which are made of phonemes, cannot be considered eternal. Not only that: Even if words are maintained to have fixed (nitya) meanings, sentences are composed of words. Being there consistency with authorship of a person, why should the Veda be not authored?.100

6.3. Reasons for the validity of Veda: vedic words are a creation More particularly, to prove that Vedic words have an author the following inference is formulated by Jayanta: (a) Vedic compositions have an author, because they are compositions, like common compositions.101

Kumårila (§ 5.4) built a symmetrical argument which proves exactly the opposite, i.e. that the Veda is permanent, and therefore it has no author. But Jayanta replies that his inference has an inconclunityaivågnåv ivoßñatå ÙÙ racanå api vaidikyo naitåΔ purußanirmitåΔ Ù kaviprañœtakåvyådiracanåbhyo vilakßañåΔ ÙÙ (ibidem, p. 287). 97 vaktrekatve tu gaganådau kutas tatkr¢to bhedaΔ? (ibidem, p. 306). 98 tad aya∫ vastusa∫kßepaΔ, upekßyatå∫ vå sarvatra såmånyavi†eßavyavahåraΔ, ißyatå∫ vå gotvådivad gakårabhedavr¢tti gatvasåmånyam (ibidem, p. 310). 99 sati ca kr¢takabhåve tasya kartå puråñaΔ (ibidem, p. 326). 100 padanityatvapakße ’pi våkye tadracanåtmake. kartr¢tvasambhavåt pu∫so vedaΔ katham akr¢trimaΔ (NM 4, vol. 1, p. 327). 101 vaidikyo racanåΔ kartr¢pürvikåΔ, racanåtvåt, laukikaracanåvat (ibidem, p. 327).

316

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

sive (anaikåntika) inferential mark which fails to prove the beginninglessness of the Veda, since we can apply it also to compositions which Mœmå∫sakas do not deem permanent, for instance to the Mahåbhårata: Every learning of the Mahåbhårata implies the learning of the Mahåbhårata by the teacher, because it is called learning of the Mahåbhårata, like today’s learning of the Mahåbhårata.102

In this connection, Kumårila (§ 5.4) retorted that for the Mahåbhårata the situation is altogether different, because the record of its author is quite settled. On the other hand, there is no record of an author of the Veda in the Vedic tradition of studies. The only record occurs in arthavåda passages, which cannot be taken as evidences. Now, asks Jayanta, is this new argument of the ‘unrecorded author’ devised as a positive and autonomous evidence, or as a refutation of (a)? If the latter is the case, the refutation is ineffective, for a valid inference cannot be refuted by another inference. So, says Jayanta, the issue is which of the two arguments is less cogent: But which of the two inferential marks is less consequential (prayojaka),103 the ‘composition’ or the ‘unrecorded author’ one? Only the ‘composition’ mark is [always] consequential; in fact without a person nowhere there is evidence of literary compositions.104

But even if we accept compositions in general as artificial, we may still hold that Vedas have the peculiar character of being naturally and permanently existing, may say the Mœmå∫sakas. Answers Jayanta: Respected gentlemen, where is it seen or heard in this world that composition of words into sentences is a natural one? If in the Veda the composition of words had a natural character, then why is it not natural also [the composition] of threads in a cloth?105

In other words, what is this distinction that makes you think that Vedas have the peculiar character of not being authored? Mœmå∫sakas may easily answer: 102 anaikåntika† cåya∫ hetur bhårate ’py evam abhidhåtu∫ †akyatvåt, bhåratådhyayana∫ sarva∫ gurvadhyayanapürvakam, bhåratådhyayanavåcyatvåd, idånœntanabhåratådhyayanavad iti (ibidem, p. 328). 103 For possible meanings of the word prayojaka, see NK, s.v. 104 nanu katarad anayoΔ sådhanayor aprayojaka∫ racanåtvåd asmaryamåñakartr¢katvåd iti ca. ucyate. racanåtvam eva prayojaka∫ na hi purußam antareña kvacid akßaravinyåso dr¢ß™aΔ (NM 4, vol. 1, p. 330). 105 bho bhagavantaΔ sabhyåΔ kveda∫ dr¢ß™a∫ kva vå †ruta∫ loke yad våkyeßu padånåm racanå naisargikœ bhavati. yadi svåbhåvikœ vede padånå∫ racanå bhavet. pa™e hi hanta tantünå∫ katha∫ naisargikœ na så (ibidem, p. 330).

III. How to Produce, Construct and Legitimate a Tradition

317

The Vedic composition appears clearly different from compositions such as Kålidåsa’s which have an author. […] Peculiar uses of nouns, verbs, verbal prefixes; praising formulas, warnings on wrong acts, prescriptions relating to past events or other people;106 description of items refuted in relation to other Vedic branches. These and others are not seen in common compositions. Therefore all scholars believe Veda to have no author because of its form. Mœmå∫sakas have drunk glory in this world. Morons like you [Naiyåyikas] who did not study Vedas, however, talk of ‘production’, being confused by the resemblance to [ordinary] compositions.107

This may be, answers Jayanta, but it does not affect the Naiyåyika stance; the ‘composition’ argument still stands: This is our verdict: no matter what Mœmå∫sakas may drink –glory, or milk, or BraΔmœ-ghee to improve their dull minds–, Vedas are composed by a person and here there is no confusion. Like the status of mountains, etc., depends on a creator although different from the status of a jar, so it is for the composition of the Veda.108

Enough, says Jayanta, with the ‘composition’ argument. What about the cogency of the ‘unrecorded author’ argument? It is inconsequential, as it implies a ‘false cause’ fallacy, because the forgetfulness of the author could be just caused by the long time elapsed since the composition of the Veda (see also § 3.4). Moreover, Jayanta shows his skepticism for the unduly universalization of subjective belief attempted by the Mœmå∫sakas: How do you know that every person does not remember the author of the Veda? Indeed you do not have experience of everyone’s heart, otherwise you would be omniscient.109

And even the validity of the Veda is undermined if no author is recognized for it. Can there be an instruction without an author? And even if there can be one, will there be anyone who develops faith in such abstract teachings? On the word-meaning relation held high by the Mœmå∫sakas, Jayanta says that the understanding of a meaning from a word is not 106 The translation abides the gloss of Cakradhara, who follows an opponent quoted in MS, see MS 6.7.26-30. 107 nanu yåΔ kålidåsådiracanåΔ kartr¢pürvikåΔ Ù tåbhyo vilakßañaiveya∫ racanå bhåti vaidikœ ÙÙ […] nåmåkhyåtopasargådiprayogagatayo navåΔ Ù stutinindåpuråkalpaparakr¢tyådinœtayaΔ ÙÙ †åkhåntaroktasåpekßavikßiptårthopavarñanam Ù ity ådayo na dr¢†yante laukike sannibandhane ÙÙ tenådhyetr¢gañåΔ sarve rüpåd vedam akr¢trimam Ù manyante eva loke tu pœta∫ mœmå∫sakair ya†aΔ ÙÙ vedå na pa™hitå yais tu tvådr¢†aiΔ kuñ™habuddhibhiΔ Ù kåryatva∫ bruvate te ’sya racanåsåmyamohitåΔ ÙÙ (NM 4, vol. 1, pp. 331f). 108 ucyate. mœmå∫sakå ya†aΔ pibantu payo va pibantu buddhijå∂yåpanayanåya bråΔmœghr¢ta∫ vå pibantu vedas tu purußaprañœta eva nåtra bhråntiΔ. yathå gha™ådisa∫sthånåd bhinnam apy acalådißu sa∫sthåna∫ kartr¢mat siddha∫ vede ’pi racanå tathå (ibidem, p. 332). 109 sarve pumå∫saΔ karttåra∫ vedasya na smarantœti katha∫ jånåti bhavån? na hi tava sakalalokahr¢dayåni pratyakßåñi, sarvaj∞atvaprasaõgåt (ibidem, p. 333).

318

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

justifiable without a conventional relation established by a person. What is convention? Convention is the prescribed restriction of signifier and signified.110

A fixed word-meaning relation is not admissible, because with sense perception and the other instruments of knowledge we cannot cognize it. But when is this convention made? At every utterance? By every person? Or once for all by God? Jayanta’s position is that it is made once at the beginning of creation by God. Therefore a whimsical use of words is not admissible. Jayanta leaves room, however, for neologisms which are created by modern convention, and he calls such usages ‘optional, extemporary’ (yadr¢cchå). Finally, the Mœmå∫sakas wittily object (§ 3.4) that whoever established the convention also needed words for that purpose. Who established the convention of those words? This leads to an infinite regress. Answers Jayanta: You have the weapon but you did not hit the target. This fault applies to us but not to God. Who can fathom his skill, if with his might he creates at will such a universe, a place of multifarious actions and results?.111

In conclusion, the Veda is a valid instrument of knowledge because it is spoken by a reliable source and not because it is beginningless. Reliability of the author,112 says Jayanta, is the inferential reason behind the statement of validity of any piece of verbal knowledge, and religious matters are not excluded. There is an invariable relation of concomitance between reliability of the author and validity of a sacred text. Conclusion In the Introduction three possible meanings of the word tradition were mentioned: cultural horizon, succession of thinkers, word-conveyed knowledge. Let us briefly review the issues examined in this paper in the perspective of this paradigm. Tradition as cultural horizon. Nyåya and Mœmå∫så become aware of the Vedic cultural horizon and explicitly problematize it in their sys110

abhidhånåbhidheyaniyamaniyogaΔ samaya ucyate (ibidem, p. 339).

111 astram åyußmatå j∞åta∫ vißayas tu na lakßitaΔ Ù asmadådißu doßo ’yam œ†vare tu na yujy-

ate ÙÙ nånåkarmaphalasthånam icchayaivedr¢†a∫ jagat Ù sraß™u∫ prabhavatas tasya kau†ala∫ ko vikalpayet ÙÙ (ibidem, p. 345). 112 According to George Chemparathy, author of the only book on vedapråmåñya in Nyåya, Jayanta stresses the importance of the agreement among mahåjana (many/important people) to prove the validity of the Veda (Chemparathy 1983: 58-72). However, this argument has little space in NM and is only used in his skirmishes with Buddhists. On the other hand, Kumårila defeats it in TV ad 1.3.4., v. 269 (see § 5.4).

III. How to Produce, Construct and Legitimate a Tradition

319

tems firstly as an epistemic tool. This does not mean that social or religious reasons have no role in the defense of the Veda, but these influences are not expressed, at least not in the period examined herewith. Both traditions agree that knowledge of dharma (moral authority) exceeds the precincts of human knowledge. But Mœmå∫sakas (see also MS 1.3.4-9) maintain that the Veda is the source of knowledge of dharma. Naiyåyikas, instead, believe that the root-source of information about dharma and Veda is an extra-ordinary human being (such as divine seers, or God himself). Nyåya’s and Mœmå∫så’s horizons of acceptance of the Veda, however, does not exactly coincide. As Wilhelm Halbfass113 said, unlike Mœmå∫så, which has in the Veda its very raison d’être, Nyåya faced the problem of its justification as an already well-developed system. Moreover, for Mœmå∫sakas only Vedic prescriptions, i.e. the bulk of the Bråhmañas, are genuine means of knowledge. On the contrary, Naiyåyikas refer to Vedic passages celebrating Prajåpati, i.e. parts of the Sa∫hitås which would be regarded by Mœmå∫sakas as arthavådas, as internal evidence for the existence of an author of the Veda. Noticeably, it is only with Jayanta that this difference becomes evident (see Daya Krishna 2001: 74-79). Tradition as succession of thinkers. These different motives determined a separate development of the two schools in the matter of justification of Veda’s validity. Nyåya tries to demonstrate that language and Veda must have a beginning and therefore an author. Mœmå∫så argues that without and until contrary evidence one must adhere to the present state of things: the Veda is universally recognized as valid; no Veda’s author is recorded; a beginning in the use of language and in the transmission of Veda just contradicts the very premises of our understanding of the world. The mutual comprehension of the two schools is sometimes threatened by different assumptions and linguistic ambiguities, such as the use of the same technical terms (†abda, veda, artha, pråmåñya, etc.) in a different acceptation. For instance, a major difficulty derives from the interpretation of the nityatva of Veda and language, explicitly criticized by Uddyotakara as referring to a temporal eterni113 “The Nyåya does not try to compete with the Mœmå∫så in the technical field of Vedic exegesis […] [Naiyåyikas’] teachings which are not inherently affiliated with the ‘Vedic’ or ‘Vedicizing’ traditions, and are at least potentially neutral, serve an increasingly apologetic function. In the history of Mœmå∫så, the development seems to have been the reverse. A genuinely and originally exegetic and text-oriented tradition opens itself increasingly to epistemology and logic, and to inherently ‘neutral’ and universal methods of thought and argumentation” (Halbfass 1992: 31). And: “Der Nyåya behauptet seine eigene Relevanz fr die Verteidigung des Veda, indem er sich eine von der Mœmå∫så nicht voll wahrgenommene Funktion zuweist […] Er tritt sozusagen von aussen an den Veda heran, stellt die von ihm entwickelten, ursprnglich aus anderen Quellen stammenden und auch fr andere Zwecke eingestzten Mittel des Denkens und Argumentierens in den Dienst der Verteidigung des Veda. Er bernimmt die Aufgabe, die bermenschliche Offenbarung des Veda mit den Mitteln menschlicher, nicht ihrerseits aus der Offenbarung stammender Einsicht zu klären und zu sichern” (Halbfass 1991: 130).

320

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

ty. But for Mœmåμsakas nitya means primarily ‘fixed’, ‘permanent’ (and not ‘eternal’), and it firstly qualifies the relation between word and meaning, rather than words as such.114 Tradition as word-conveyed knowledge. As a consequence of their inner link with Veda, Mœmå∫sakas never shift their focus from Vedic to common language. So human testimony, if it is at all considered to be an instrument of knowledge,115 does not play a role in the discussion on Veda’s validity. Naiyåyikas, on the other hand, are primarily interested in justifying human testimony, and as far as Vedic statements are concerned they just apply to them the same criteria.116 Similarly, Mœmå∫så discusses language as a general theme departing from the crucial specimen of Vedic language, while Nyåya discusses Veda as an instance of ordinary language. From this different approach many misunderstandings originate, such as the one about the role of a speaker/author.117 Another meaningful difference is that Mœmå∫sakas’ perspective is on the side of the hearer, and hence they do not really care for the origin or production of words. Conversely, Naiyåyikas, at least until Jayanta, mainly deal with the production of words, both as a linguistic and as an epistemic instrument.

114 “Jaimini’s system requires not a demonstration of the ‘eternity of the word’, but simply the certainty that the word-meaning relationship exists prior to the action of any speaker” (Clooney 1990: 78). 115 Prabhåkara, for instance, considers it as a kind of inference. 116 Quite appropriately, I think, Jonardon Ganeri compares the Naiyåyika discussion on †abdapramåña to Hume’s one on miracles and states: “The Nyåya quite sensibly restrict the domain of testimony to natural extensions of our already existing beliefs. They are interested in the role of testimony in everyday life, not in the special epistemological problems posed by religious witnesses” (Ganeri 1999: 80). 117 As synthesized by Arindam Chakrabarti: “There are two ways in which […] the essentiality of commitment to the tradition as a necessary condition for rationality can be brought about. The first way can be called the Nyåya-Dummett way and the second the Mœmå∫så-Gadamer way. By emphasizing the irreducible role of knowledge from words (†abda pramåña) in the acquisition and use of language, Nyåya epistemology exposes a fundamental error of Lockean individualistic epistemology. […] [T]he other way, adopted by Kumårila Bha™™a […] turns on an underlying distrust of individual speakers and treats a speaker-less body of received tradition to be the only possible source of moral knowledge” (Chakrabarti 1997: 266-267).

III. How to Produce, Construct and Legitimate a Tradition

321

BIBLIOGRAPHY Abbreviations and Sources F MS NBh NM NS NV NVT NK NR ‡Bh

K ‡V TV ‡VV

All quotations preceded by F refer to page and line in Frauwallner 1968. Jaimini, Mœmå∫såsütra, in Frauwallner 1968. Vatsyåyana, Nyåyabhåßya, see NS. Jayanta Bha™™a, Nyåyama∞jarœ with the comentary of Granthibhaõga by Cakradhara, edited by Gaurinath Sastri, ‡ivakumåra†åstri-granthamålå, Varanasi 1982. Gautama, Nyåyasütra, in Nyåyadar†anam, edited by Taranatha Nyaya-Tarkatirtha, Munshiram Manoharlal, Calcutta 1985 [first ed. 1936-1944]. Uddyotakara, Nyåyavårttika, see NS. Våcaspati, Nyåyavårttikatåtparya™œkå, see NS. Nyåyako†a, edited by Bhœmåcårya Jhalakœkar, Bhandarkar Oriental Research Institute, Pune 1996. Pårthasårathimi†ra, Nyåyaratnåkara, see ‡V. ‡abara, ‡åbarabhåßya, I.1.1-5 in Frauwallner 1968. I.1.6-32 in The Mœmånså Dar†ana, edited by Mahe†a Chandra Nyåyaratna, Bishop’s College Press, Calcutta 1863. From I.2. onwards, in Mœmå∫sådar†anam, edited by Måhåprabhulåla Gosvåmœ, Tara Printing Works, Varanasi 1984. Sucaritami†ra, Kå†ikå, edited by K. Såmba†iva ‡åstrœ, Trivandrum Sankrit Series n°XC, XCIX, Trivandrum 1926; 1929. Kumårila Bha™™a, ‡lokavårttika with the commentary Nyåyaratnåkara of ‡rœ Pårthsårathi Mi†ra, edited by Gaõgå Sågar Råy, Ratna Publications, Varanasi 1993. Kumårila Bha™™a, Tantravårttika, in Mœmå∫sådar†anam, edited by Måhåprabhulåla Gosvåmœ, Tara Printing Works, Varanasi 1984. U∫veka Bha™™a, ‡lokavårttikavyåkhyatåtparya™œkå, edited by S.K. Råmanåtha ‡åstrœ, University of Madras, Madras 1971 [first ed. 1940].

Studies Arnold, D. (2001), Intrinsic Validity Reconsidered: a Sympathethic Study of the Mœmå∫saka Inversion of Buddhist Epistemology, in «Journal of Indian Philosophy», vol. 29, n. 5/6, pp. 589-675. Bilimoria, P. (1989), On the Idea of Authorless Revelation (Apaurußeya), in Roy W. Perrett (ed.), Indian Philosophy of Religion, Kluwer, pp. 143-166. ______ (1994) Autpattika: the ‘originary’ signifier-signified relation in Mœmå∫så and deconstructive semiology, in Dwivedi, R.C. (ed.),

322

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Studies in Mœmå∫så: Dr. Mañ∂an Mishra Felicitation Volume, Motilal Banarsidass, Delhi, pp. 187-205. Bronkhorst, J. (1998), Does the Veda have an Author?, in «Asiatische Studien», vol. 52, n. 1, pp. 5-14. Chakrabarti, A. (1997), Rationality in Indian Philosophy, in Deutsch, E., Bontekoe, R. (eds.) Blackwell’s Companion to World Philosophy, Blackwell, Oxford, pp. 259-278. Chari, V.K. (1992), ‡abda-Pråmåñya: an Analysis of the Mœmå∫så Argument, in «Journal of Oriental Research», Madras, 56-62 (1986-1992), pp. 96-105. Chattopadhyaya, D., Gangopadhyaya, M. (1981) [1968], Nyåya Philosophy. Literal Translation of Gautama’s Nyåya-sütra & Vatsyåyana’s Bhåßya, Firma KLM Private, Calcutta. Chemparathy, G. (1983), L’autorité du Veda selon les Nyåya-Vai†eßikas, Centre d’Histoire des Religions, Louvain-La-Neuve. Clooney, F.X. (1990), Thinking Ritually. Rediscovering the Pürva Mœmå∫så of Jaimini, Publications of the De Nobili Research Library, Wien. Daya Krishna (2001), New Perspectives in Indian Philosophy, Rawat Publications, Jaipur. Devasthali, G.V. (1942), On the probable date of ‡abara-Svåmin, in «Annals of Bhandarkar Oriental Research Institute», vol. 23, pp. 84-97. D’Sa, F.X. (1980), ‡abdapråmåñyam in ‡abara and Kumårila. Towards a Study of the Mœmå∫så Experience of Language, Publications of the De Nobili Research Library, Wien. Franco, E. (1997), Dharmakirti on Compassion and Rebirth, Arbeitskreis für tibetische und Buddhistiche Studien, Wien. Frauwallner, E. (1968), Materialen zur ältesten Erkenntnislehre der Karmamœmå∫så, Österreichische Akademie der Wissenschaften, Wien. ______ (1961), Mœmå∫såsütram I, 1, 6-23, in «Wiener Zeitschrift fr die Kunde Sd- und Ostasiens», 5, pp. 113-124. Gadamer, H.-G. (19906)[1960], Wahrheit und Methode, Mohr, Tbingen. Ganeri, J. (1999), Semantic Powers. Meaning and the Means of Knowing in Classical Indian Philosophy, Clarendon Press, Oxford. Halbfass, W.(1991), Tradition und Reflexion. Zur Gegenwart des Veda in indischen Philosophie, in Oberhammer, G. (hrsg), Beiträge zur Hermeneutik indischer und abendländischer Religionstraditionen. Arbeitsdokumentation eines Symposiums, Verlag der Österreichischen Akademie der Wissenschaften, Wien, pp. 123-46. ______ (1992), Observations on the Relationship between Vedic Exegesis and Philosophical Reflection, in «Journal of Oriental Research», Madras, 56-62 (1986-1992), pp. 31-40. Jhå G. (1933), ‡åbarabhåßya, Oriental Institute, Baroda. Kataoka, K (1995), Naraseru ho Raishaku-gaku (Mœmå∫så Theory of Causal Action: ‡abara’s Concept of Bhava, Kriyå and Bhåvanå

III. How to Produce, Construct and Legitimate a Tradition

323

[in Japanese with an English Summary]), in «Indo-Tetsugaku Bukkyo-gaku Kenyu», vol. 3, pp. 47-60. Keith, A.B. (1921), The Karma-Mimamsa, Oxford University Press, London. Matilal, B.K. (1985), Logic, Language and Reality, Motilal Banarsidass, Delhi. Mohanty, J.N. (1992), Reason and Tradition in Indian Thought. An Essay on the Nature of Indian Philosophical Thinking, Clarendon Press, Oxford. ______ (2001), Exploration in Philosophy, Oxford University Press, New Delhi. Parpola, A. (1981), On the Formation of the Mœmå∫så and the Problems concerning Jaimini with particular Reference to the Teacher Quotations and the Vedic Schools I, in «Wiener Zeitschrift fr die Kunde Sd-Asiens», 25, pp. 145-177. ______ (1994), On the Formation of the Mœmå∫så and the Problems concerning Jaimini with particular Reference to the Teacher Quotations and the Vedic Schools II, in «Wiener Zeitschrift fr die Kunde Sd-Asiens», 38, pp. 293-308. Potter, Karl H. (1977), Encyclopedia of Indian Philosophy. Volume II: The Tradition of Nyåya-Vai†eßika up to Gaõge†a, Motilal Banarsidass, Delhi. Scharf, P.M. (1996), The Denotation of Generic Terms in Ancient Indian Philosophy: Grammar, Nyåya, and Mœmå∫så, Transactions of the American Philosophical Society, Philadelphia, vol. 86, pt. 3. Taber, J.A. (1983), Book review: ‡abdapråmåñyam in ‡abara and Kumårila, in «Philosophy East and West», vol. 33, n. 4, pp. 408-411. ______ (1992), What did Kumårila Bha™™a mean by svataΔ pråmåñya?, in «Journal of the American Oriental Society», 112, pp. 204-221. ______ (1998) Mœmå∫så in Routledge Encyclopedia of Philosophy, Routledge, London et al., vol. 6, pp. 376-381.

CEZARY GALEWICZ Why should the Flower of dharma be Invisible? Såyaña’s Vision of the Unity of the Veda

Une littérature diffère d’une autre moins par le texte que par la façon dont elle est lue (M. de Certeau quoting Borges in Lire: un braconnage)

The ‘flower of dharma’ (dharmapußpa) is a metaphor used by Såyañåcårya1 in the scholastic introduction to his R¢gvedårthaprakå†a, known also as R¢gvedabhåßya –a Vedic commentary which established itself among Indian scholiasts as a canonical one beginning from the late Middle Ages. The metaphor is well attested prior to Såyaña in the tradition of Mœmå∫så thinkers and is associated with a bipartite image of the Veda divided into that of karma- and brahma- khåñ∂a. This rough division had been well established within the Mœmå∫så scholars but rather shunned by Vedic studies outside India. Medieval and pre-modern Vedic exegesis somehow did not attract much attention of modern Vedic studies2 in Europe, usually being dismissed as either anachronistic or strongly biased. While often referred to in passing for the sake of comparison, Såyaña’s commentaries seldom have been given pride of place as original pieces of independent scholarly work. They used to be judged as eclectic3 and generally not offering much worth lasting academic attention.4 1 Ca. 1315-1385 AD. In this essay all the quotes from Såyaña’s R¢gvedabhåßya is from the Mller (1983) edition. 2 This seems to be changing recently. 3 Gonda who describes Såyaña’s commentary as: “proposing many anachronistic misinterpretations, defending wrong etymological or grammatical explanations, failing in the coherent explanation of stanzas and sentences as wholes, and inserting unnecessary discussions” (Gonda 1975: 41). More about the history of early European reception of Såyaña’s work see Figueira 1994; Patton 1994. 4 On a somewhat later, pre-modern example of exegesis and re-interpretation of Vedic texts, as well as some methodological problems with reading and evaluating indigenous pre-modern commentaries, see Minkowski 2002: 28-30.

326

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Studies focusing on his work have been rather rare, with a few exceptions, like that by Oertel, highly valuable and informative, today already rather a thing of the past in terms of methodology deriving from the time of its conception (Oertel 1930). Even less lucky with the attention of modern scholars was the R¢gbhåßya Upodghåta –the introduction to the commentary on R¢gveda–, an independent work in itself and a sophisticated piece of learned †åstric literature.5 Author of numerous commentaries on primary and secondary Vedic texts,6 as well as those of vyåkaraña, ala∫kåra, ayurveda and dharma,7 Såyaña became somehow an icon of himself –the image substituted the historical person.8 If all those works attributed to him were actually written or authored by him, Såyaña must have been a scholar of rare refinement, amazingly profound learning and widest horizon of interest. The purpose of this contribution is to draw attention to a few aspects of historical and socio-cultural nature pertaining to the commentator’s rationale expressed primarily in Såyaña’s R¢gbhåßya Upodghåta.9 It is its relation to the centres of political power and religious authority as shaping its success in establishing itself as a canonical commentary which are of my main interest here. Those aspects will be seen as reflecting and conditioning the phenomena that shall be provisionally termed as ‘appropriation’ and ‘reinterpretation’ pertaining to the Vedic canon and identified otherwise on textual grounds within the R¢gbhåßya itself. While entering the realm of history and historiography of South Asia, certain general problems are to be faced concerning the nature of available data sources. How to interpret different sort of sources in reconstruction of the South Indian history? Could any literary text of a period be equally a source for the history of that period? Certainly not. But each and every one has something to tell us. And this is probably what we should expect while reading them today. In this respect I 5 Only a few attempts at translation (Peterson 1974; Ray 1961; Bali 1999) had been done and actually little in terms of study of the text (with an exception of Oertel 1930 and to some extent Ray 1961). Introductions to other Vedic commentaries received almost no attention from modern scholars. Stray references to some of Såyaña’s ideas expressed in his Upodghåta can be found in Kahrs 1998. 6 He is believed to have authored commentaries to sixteen different Vedic works. Whether he was indeed a sole author of all the works attributed to him is sometimes doubted because of the extraordinary volume of those texts. Perhaps some of the Vedic commentaries were in fact authored or compiled by him while actually a collective body of pandits may have, to a certain extant, jointly work on their production. That a body of Vedic experts like that indeed worked on commentaries is suggested by inscriptions (e.g. copper plate grant dated 1386 [Modak 1995: 30; Kulke 1985: 131]) recording donations of land to those bråhmañas for their work on commenting the Vedas. Såyaña however does not mention any names of his collaborators, except for that of his brother Mådhava. 7 For a full list of works attributed to Såyaña see Modak 1995: 46. 8 For the reconstruction of historical background of Såyaña’s writing see Narasimhachar 1916; Kulke 1985; Kripacharyulu 1986; Modak 1995. 9And, secondarily, also in the introduction to Taittirœyasa∫hitåbhåßya.

III. How to Produce, Construct and Legitimate a Tradition

327

can not help recalling a quotation from C.C. Berg referred to by Kulke, in his Functional Interpretation of a South Indian Mahåtmya: In Berg’s opinion […] for the reconstruction of Javanese history it is not enough just to read a text and take the details which appear suitable for the reconstruction. The main question must always be: ‘Why did the man write his book, and why did he write it thus?’. It seems that in such perspective it is extra-textual context that is being focused upon rather than the content of the text. Yet, not all sorts of texts should perhaps be approached in this way. Is the question formulated by Berg meaningful for Såyaña? Is the kind of reading it entails the right choice in our case? We have every reason to believe that at least an important as it is a historical embedding of Såyaña’s commentaries make such reading that pays a closer attention to the relation between the content, formal elements of composition and socio-historical context a worthwhile one. (Kulke 2001: 193)

Såyaña’s work stands on the crossroad of two different worlds: on one hand a piece of commentatorial work –a Vedic exegesis– it belongs to the highly hermetic world of Sanskrit traditional †åstras on the other. Even more so it is the case with the text of the Upodghåta to the R¢gvedabhåßya. It is by no means an easy data source for a historian. True to medieval Indian style of learned disputes within their scholastically defined convention, it does not present itself as an easy pray for a hunger of proper names and dates.10 Yet, it is a point of this short study to stress that texts as intellectual productions which take form in a certain time and socio-cultural environments must somehow bear the stamp of the historical moment as well as of the institutions, places and people who shaped the culture of the time and influenced the production, authorship and reception11 of those texts. Treated cautiously, they bear witness to and reflect at least some of the relations between themselves, their historical authors, producers, editors, patrons, disseminators, readers, listeners and users. An attempt at a historical reading of Såyaña’s texts here does not aim at any sort of history writing. It aims rather at giving justice to historically specific intellectual and social context of Såyaña’s work which 10 Full of intertextual references and direct quotations, Såyaña’s introduction brings only a handful of proper names, often presented in stereotyped, abridged manner, for instance ‘guru’ for Prabhåkara or ‘bha™™a’ for Kumårila. 11 See Pollock 2003: 102: “Literary culture is a phenomenon that exists not just in time but also in space […]. The sites of production and consumption concern the social locations (court, temple, school, and so on) that help shape the primary meanings and significations of literature. […] The sociotextual community for which a literature is produced derives a portion of its self-understanding as a community from the very act of hearing, performing, reproducing, and circulating literary texts. The conceptualization of space in literature and the embodiment of this concept in people are often importantly related to political formations, which exercise power over persons in space”. While Pollock does not count Vedic texts (he does not say it explicitly about medieval commentaries) as ‘literature’, which he considers to be only that labeled kåvya, I think it may well be extended here to the commentarial traditions which are nor Vedic themselves any more and belong to the realm of literary productions of a †åstric type.

328

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

might have added to, or stimulated, his motivations and decisions which otherwise would remain for us highly unclear. It is hardly thinkable, in my opinion, that a work of that volume and ambition remained untouched by the intricate patterns of mutual relations between intellectual production and political power that must have been at work during early years of Vijayanagara, as they are everywhere and anytime, though culturally and historically specific. My own inspiration draws here in part from the contemporary interest in the paradigms and strategies of state formation in South Asia and anxiety towards finding a suitable model construction for describing the working of Sanskrit and vernacular intellectual traditions in their own historicity and their historical associations with centres of power. And specifically also from the hypothesis formulated once by Paul Hacker and referred to by H. Kulke in his study on early Vijayanagara: “Together with his famous brother Såyaña, the author of the Vedabhåßya, Vidyårañya, tried to establish in an act of intentional cultural policy (in Einer Art bewußter Hindu-Kulturpolitik [Hacker 1978: 478 ff]), a new system of orthodoxy in order to counteract the influence of Islamic inroads into South India” (Kulke 2001: 236). However, while the wording of the idea of “intentional cultural policy” sounds to me still promising as a tool for interpretation, that of “counteracting the influence of Islamic inroads” as an exclusive motivation of such policy should perhaps be verified and substituted with a general motivation for legitimizing imperial hegemony of early Vijayanagara over vast areas belonging to various political and civilizational centres of South India,12 as well as in the broader political and cultural context of South Asia in the second half of 14th century AD. With these general presuppositions let us come back to the case of Såyaña’s vision of the unitary Veda as discussed in the introduction to his commentary to the R¢gveda, composed as part of an imperial project instigated by the early rulers of South Indian kingdom of Vijayanagara and aiming at commenting on the whole of the Veda. As not only the case of the introduction to the R¢gvedabhåßya but also those to other bhåßyas to Vedic collections ascribed to Såyaña show, the definition of what actually the Veda is must have been considered quite of a problem to be undertaken time and again for a dispute with real or imagined opponents in a learned treatise as our work in question certainly is. 1. What is visible and what is not in the Veda? The case of Månavaka and Purñikå The textual body of the R¢gveda commented by Såyaña belongs in the Mœmå∫så perspective to that part of the realm of the Veda which 12 We should definitely take into account a wider than only Hindü sphere of reference for the political decisions taken by the Vijayanagara rulers in their policy towards safeguarding the legitimation for their imperial power (Wagoner 1996: 863).

III. How to Produce, Construct and Legitimate a Tradition

329

is recognized by the name of mantra.13 As such it makes a part of the overall project of vedårthaprakå†a. In contradistinction to other Vedic texts commented by Såyaña, for instance that of Taittirœyasa∫hitå, it is made up only of mantra type of text and one would suspect that whenever Såyaña is speaking about mantras in his introduction to R¢gvedabhåßya he means R¢gveda mantras. But this is not the case at all. There is something disquieting in the way Såyaña builds relations between the concepts of mantra, bråhmaña, vedårtha and dharma. A contemporary reader of his Upodghåta may even get impression that Såyaña happens to contradict his own ideas while proceeding further on with his introduction.14 At first, he strongly opposes the idea that mantras have an invisible (adr¢ß™a)15 meaning only.16 He several times explicitly declares his (and his school?) position, namely that the Vedic mantras do have a visible, intended meaning.17 Especially while they are recited at the sacrifices.18 What then about mantras being recited in other circumstances? Here an example is brought under consideration in which Månavaka, a young brahmacårin, is reciting the avåghata mantra (a mantra enjoining the action of ‘pounding [the grain]’) during his daily Vedic exercise (vedåkßaragrahåña) in the vicinity of a girl named Purñikå who is pounding corn with a pestle. In such a situation, admits Såyaña, the avåghata mantra, which is meant in the system of Vedic ritualism to accompany the ritual threshing of the corn in a mortar according to a fixed rhythmical procedure does not express its meaning. Why? Because it is not intended to do so.19 Does it mean that Vedic mantras can or can not express meaning, and that whether they do so depends on the context of the recitation (type of ‘textual lecture’)? The pürvapåkßin –imagined opponent– does not pose such a The other being that of bråhmaña. In a few cases it is not quite clear whether particular argument comes from pürvapåkßa or uttarapåkßa. In fact we cannot exclude that in certain parts the text of his introduction as well as the commentary itself lacks consistency due to corruptions and bad preservation of manuscripts. See e.g. Mller 1983: xviii. 15 The idea of the pürvapåkßa is that Vedic mantras have (only) invisible meaning if any as many of them may be quoted that do not seems to have any referent in the world of common experience. The so called Kautsa argument (the argument of a thinker named Kautsa) that mantras are meaningless is an old motive present already in Yåska’s Nirukta 1.15 (anarthakå hi mantra). See Pürvamœmå∫såsütra, 1.2.1. 16 More about the problem of meaninglessness of mantras and ritual see Staal 1979 and Staal 1996. 17 mantrasyåvivakßitårthatve tu ki∫ nåma tåtparya∫ mantre vyåkhyåyeta Ù tasmådvivakßitårthå mantråΔ prayogakåle svårthaprakå†anåyaivoccårayitavyåΔ ÙÙ (Mller 1983: 6, ll. 28-29). 18 See Taber 1989: 26, n. 6. Perhaps Taber is right while stating that “Såyaña works out an interesting intermediate position between Mœmå∫så and Vedånta” as concerns mantras and their meaning. 19 vedavidyågrahañakåle ’rthasya yadavacana∫ tadayaj∞asa∫yogåd [ayaj∞asa∫gåd] upapadyate Ù na hi purñikåyå avaghåto yajñasayuktaΔ Ù nåpi måñavako yaj∞amanutiß™hati Ù ato yaj∞ånupakårånna tatrårthavivakßå [yatra-] ÙÙ yadapyukta∫ amyakså ta indra sr¢ñyeva jarbharœ turpharœtœ ityådåvarthasya j∞atuma†akyatvånnåstyevårtha iti tatrottara∫ sütrayati (Mller 1983: 5, ll. 32-35). If there occurs no communication of meaning (artha) during 13

14

330

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

question, and Såyaña does not answer it. It is clear however, that the Veda is not taken by Såyaña as a textual body existing anywhere and anytime in the same manner. Though, on the other hand it is deemed as eternal and unchangeable by the Mœmå∫sakas. It seems rather to be a sort of reservoir, or a store-room, to draw from according to a certain procedure of access which is determined by a pre-set niyama. I am tempted here even to recall a general remark expressed once on quite different occasion by M. de Certeau –toute lecture modifie son objet (Certeau 1990: 245). In this type of thinking represented by Såyaña there is absolutely no one textual lecture of the Veda, but rather multiple lectures having different aims and these different lectures (protocols of lecture) modify the Vedic text (or its status) in certain way constituting differently their object.20 There is no one and the same Veda for every occasion. There is a Veda for studying,21 a Veda for japa-reciting (Mller 1983: 15, l. 4), or for so called brahmayaj∞a, and a Veda for sacrifice.22 There are different Vedas seen by textual manuals belonging to different Vedic †åkhås.23 Could we perhaps even say in this respect paraphrasing Derrida that il n’ya pas de texte –there is no (one) text of the Veda? Accordingly, any question pertaining to the meaning of Veda has to be related to a particular situation or a particular type of lecture (with its pre-set protocol). And yet, Såyaña strongly defends the idea of one unitary Veda and its artha. So does he that of its mantra portion. But let us stop for a moment at the issue of the visible/invisible meaning of the Vedic mantras. Såyaña time and again assures us that the meaning of the Vedic mantras (mantrårtha) is something visible (dr¢ß™a) and intended (vivakßita) and can be experienced directly. At the same time he however does not deny that additionally the time of ‘drawing the knowledge of the Veda’ this is due to lack of association with the sacrifice. The Purñikå’s threshing of the corn is not connected with sacrifice. Neither Måñavaka is carrying a sacrifice. Thence due to no assistance to a sacrifice there is no intention of expressing meaning here. 20 Mœmå∫så, of course, conceives of the text (of the Veda) as something to do with rather than something to read. 21 See Mller 1983: 14, ll. 14-16. But there is another reality behind the argument of Såyaña and the Mœmå∫sakas: the studying of the Veda as svådhyåya is a ritual, and as such may be carried on only by qualified persons, by the adhikårin. On the early notion of svådhyåya as an independent ritual and that of brahmayaj∞a see Taittœriyårañyaka, 2.1 and Malamoud 1977: 24. 22 The problem of different arrangement of Vedic texts for different use is reflected partly in the twin notions of mantrakåñ∂a and karmakåñ∂a on one hand and dharmakåñ∂a and brahmakåñ∂a on the other. See, for instance, Mller 1983: 19, l. 3 (mantrakåñ∂o brahmayaj∞ådijapakrameña pravr¢tto na tu yågånußthåna-krameña) where kalpa is pointed out as necessary to determine a proper arrangement of the text for a particular ceremonial occasion in contradistinction to the arrangement ‘by mantra’ characteristic of applications known as brahmayaj∞a and japa. 23 Såyaña seems to be perplexed by this plurality himself and thematizes it in a series of pre-arranged doubts voiced by pürvapåkßin: ityådißu kr¢tsnamantrakåñ∂aviniyogeßu sa∫pradåya-påra∫paryågata eva krama ådarañœyaΔ Ù vi†eßaviniyogå∫stu mantravi†eßåñå∫ †rutiliõgavåkyådipramåñåny-upajœvya å†valåyano dar†ayati Ù ato mantrakåñ∂akramåbhåve ’pi na ka†cid-virodhaΔ (Mller 1983: 19, l. 8).

III. How to Produce, Construct and Legitimate a Tradition

331

mantras may have also adr¢ß™a meaning, an invisible (spiritual?) one.24 It is the former one which dominates and there is no use to look for the invisible when the visible is clear and at hand.25 To support his position he quotes from his brother Mådhava’s Jaiminœyanyåyamålåviståra.26 It stands out to reason, however, that, according to him, the visible artha of the Vedic mantras is something that can be seen when they are duly recited as accompanying the Vedic rites. This perspective seems to explain, or at least account for, a somewhat ambiguous formulation from the beginning of the Upodghåta, where Såyaña brings to the fore the primacy of knowledge of the sacrifices for gaining an understanding of the meaning of the Veda (vedårthaj∞åna).27 If the mantrårtha is something visible from one point of view and invisible from another, so what about dharma? Do the mantras express dharma or something else than that? What is the dr¢ß™årtha of the mantras such as Såyaña sees it? What actually is dr¢ß™a and what is adr¢ß™a in the Veda according to him? The commentator seems not to allow for finding answers to those questions otherwise than by respecting the order of successive questions and answers he himself adopts.28 Could there be any reason for Såyaña not to lead his reader straight along the line of the sütras of Jaimini, whom he quotes extensively from time to time, but along clusters of pre-formulated questions rather that often contradict the original sequence of the sütras? Quotations from Pürvamœmå∫så, either in form of sütras or commentaries, no doubt play an important role in his reasoning, but they do not form a skeleton for his Introduction. They appear and are commented along with other sources of authority which seem no less important for Såyaña, mostly that of Yåska, Jaiminœyanyåyamålåviståra by Mådhava, Purußårthånu†asana, and †ruti, of course. The problem of ‘invisibility’ of dharma had been discussed on various occasions by the adherents of Sanskrit †åstra/dar†ana traditions. As it is not the place to review them, let me instead draw attention to na hi vaya∫ på™hakramanœyamådr¢ß™am nivårayåmaΔ (Mller 1983: 5, l. 13). mantrasyåvivakßitårthatve tu ki∫ nåma tåtparya∫ mantre vyåkhyåyeta Ù tasmådvivakßitårthå mantråΔ prayogakåle svårthaprakå†anåyaivoccårayitavyåΔ (Mller 1983: 6, l. 28-29). 26 See Jaiminœyanyåyamålåviståra, 1.2.4: “And now there’s a saying of two †lokas: Are mantras [like that beginning with words] ‘spread wide’ of unseen reason only? Or are they also making intelligible the ritual of spreading puroda†a and the like? If bråhmañas also make intelligible the same, are mantras of [pure] virtue reason only? No! For through their power of making things intelligible, their explicit value comes before the unseen one” (tatra sa∫graha†lokau Ù mantrå uru prathasveti kimadr¢ß™aikahetavaΔ Ù yågeßüta puro∂å†aprathanåde†ca bhåsakåΔ Ù bråhmañenåpi tadbhånånmantråΔ puñyaikahetavaΔ Ù na tadbhånasya dr¢ß™atvåddr¢ß™a∫ varamadr¢ß™ata iti ÙÙ). 27 Såyaña uses several different terms for ‘meaning’ in his introduction. What does he actually have in mind while using artha, mantrårtha, vedårtha, vedoktårtha, tatpårya, abhipråya and other terms in different contexts is an interesting question demanding a separate study yet. As for the ambiguity of the term artha it seems that a ‘core’ meaning of ‘purpose’ could be sustained in a number of examples. See Kahrs 1988; Clooney 1994: 283-284. 28 A few of interesting issues pertaining to the structure of argumentation in Såyaña’s Upodghåta are reflected upon in Oertel 1930. 24 25

332

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

a well known passage from Nirukta by Yåska –one of the authorities quoted most frequently by Såyaña. For Yåska (perhaps 6th B.C.), it seems clear that in his own times the age of the r¢ßis having direct access to dharma was long over. In Nirukta 1.20 he builds an opposition between såkßåtkr¢tadharmåñaΔ r¢ßayaΔ (original r¢ßayaΔ who had the power of direct experiencing of the dharma) and asåkßåtkr¢tadharmåñaΔ (later generations of teachers, deprived of that faculty). For Såyaña, a Mœmå∫saka as he was, it must have gone without saying (as was the case with Durga, commentator on Yåska, whose commentary on Nirukta 1.20 reads: “[…] na hi dharmasya dar†anam asty […]”) that whatever dharma actually is, it is first and foremost beyond human perception due to its (dharma’s) very nature (atœndriya) (Wezler 2001: 224; Halbfass 1988: 325). It is not my intention here to define or give historical survey of the idea of dharma as such but rather to get closer to that how Såyaña would make use of the concept in the introduction to the R¢gvedabhåßya. After all, we should not forget that “the concept of dharma is so difficult to define because it ignores or transcends differences which are essential or irreducible for Western understanding –differences between fact and norm, cosmos and society, physics and ethics, etc.” (Halbfass 1988: 312). A list of quotations could be offered in reference to what was taken as the meanings of the term dharma in various primary and secondary sources. While referring the reader to the works by Hacker, Lingat, Halbfass,29 let me recall here one stray passage from Såyaña’s commentary to R¢gveda, quoted also by Halbfass, who says that “in order to explain the ‘supportive’, ‘maintaining’ function of dharma, he [Såyaña] instead refers to the installation of such essential facilities as wells and water reservoirs” (Halbfass 1988: 316). While this rather short recourse to a passage from the commentary itself is somewhat reminiscent of the crucial role of the king for the practical maintenance of the social order, we cannot meet any of such explanations offered by Såyaña in his Upodghåta, except perhaps one. At a certain moment while discussing the object (vißaya) of the Veda as being constituted by dharma and brahma, he refers to the idea of dharma as something invisible which makes the king withhold his dañ∂a while executing due punishment, or something that helps a weaker of the two disputants win in a contest.30 Usually it is rather an idea of dharma as transcending the direct apprehension through senses that is stressed by the commen29 While admitting different contexts in which dharma might take different meanings, Halbfass pays attention to the primary meaning of the term derived from its meaning in the R¢gsa∫hitå and concludes: “The ritual dharma is the reactualization and earthly analogue of the original cosmogonic acts of ‘upholding’ and ‘holding apart’. Whatever the functions of the ritual in Indian history may have been, its fundamental, though forgotten connections with cosmogony, and its commitment to ‘upholding’ the space of the world, and to keeping the entities apart from each other and in their appropriate identities, is beyond question” (Halbfass 1988: 317). 30 uddañ∂asya råj∞o niyåmakatvådvivadamånayoΔ purußayormadhye durbalasyåpi råjasåhåyyajjayahetutvåcca dharmaΔ purußårthaΔ (Mller 1983: 18, ll. 11-13).

III. How to Produce, Construct and Legitimate a Tradition

333

tator in his rationale for the commentary. The dharma that remains essentially adr¢ß™a for any other means of cognition that the †abdapramåña, or verbal testimony, which is nothing else than the testimony of the Veda itself. However, it appears that this is not directly (textually) comprehensible to an adhikårœ (person eligible for the study of the Veda), and as such, needs an explanation (vyåkhyåna, or prakå†a). There would be nothing special to it, if not the characteristic attitude of Såyaña vis à vis the postulated injunction to apprehend the meaning of the Veda. This remains rather unclear along the reading of the R¢gbhåßya Upodghåta: at first Såyaña takes much effort to explain why the Vedic vidhi pertains primarily to the study (adhyayana) and perfection (sa∫skåra)31 of the Vedic text rather than to understanding (arthåvabodha), only to stress, later on, that mere recitation of the Veda does not constitute a situation for its full appreciation and is only a partial fulfillment of the bråhmaña’s duty vis à vis his Vedic tradition (Mller 1983: 15, ll. 22-25). Be it as it may, the general idea of dharma as transcending the means of human direct perception seems to be taken by Såyaña from Jaimini.32 For some reason, however, Såyaña feels obliged to support the idea of Jaimini (glossed by ‡abara) with an additional authority of Purußårthånu†asana: “[…] it is said in a sütra of Purußårthånu†asana: ‘Both dharma and brahma are to be known only through the Veda’”.33 One more recourse to Halbfass’ article seems of interest in reference to Såyaña’s treatment of the term in question: “All subsequent human dharma is perpetuation and renewal of the primeval upholding […]. In this sense, dharma is ‘upholding’ (dhåraña) as well as that which ‘has to be upheld’ (dhårya). Those who fulfill the dharma uphold the condition which upholds them” (Halbfass 1988: 318). Now, it is stressed once and again by Såyaña that the Veda (as expressing dharma) shall protect and favor one who is a vedavid (Mller 1983: 13, ll. 22-23). The idea of mutual support of the Veda and its practitioner is also visible in Såyaña’s commentary on the injunction to study the Veda, which however, puzzlingly, is not to lead primarily to understanding, but to a text appropriation (pråpti) and perfection (sa∫skåra). But to be a vedavid in the sense in which Såyaña wants his readers to understand means something quite different than only fulfilling an (important) injunction to study. Here we step into an important fundamental for Såyaña’s reasoning: different ‘lectures’ of the Veda have their own ready-made procedures and not much seems to be left for the ‘reader’ to decide –to study the Veda according to adhyayana procedure is to lead to a quite different goal than ‘under31 On the idea of ‘perfection’ of the Vedic text elaborated by Såyaña see Galewicz 2003. 32 Jaiminisütra, 1.1.4 (pratyakßam-animitta∫ vidyamånopala∫bhanatvåt): “Sense-perception is not the means (of knowing Dharma) because it apprehends only objects existing at the present time” (Jha 1916: 6). 33 tathå ca purußårthånu†åsane sütrita∫ dharmabrahmañœ vedaikavedye (Mller 1983: 17).

334

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

standing’. It is meant to perfect (refine) the Vedic text and the effects of this perfection are said to be directly visible to the onlookers in the time of reciting that accompanies the ritual (Mller 1983: 15, ll. 9-13). On the other hand, to become a vedavid, or vedaj∞a, which appears also to be enjoined by †ruti,34 is to embark on a totally different procedure. Såyaña states clearly that it is not enough to merely study the Veda and refrain from accepting its hidden message (which does not however mean indulging in free interpretation). The reasoning again is inscribed within the authoritative commentary by Yåska on the †ruti text of R¢gveda 10.71 which speaks about the flower and the fruit of the speech (våc) but is understood by Såyaña as referring to the knowledge of dharma and brahma.35 There are passages in the Upodghåta which create impression that it is an in-depth study of the Vedic text with the help of 14 ‘branches of science’ (vidyåsthånas) which is meant by Såyaña as a procedure leading directly to the vedårthaj∞åna. But from others, like the one just referred to, another picture emerges: to understand the meaning of the Veda and become a vedavid or vedårthaj∞a, in contradistinction to the mere Veda-reciter (på™hakamatra), an explanation, a vyåkhyåna, or a prakå†a, is a must. Careful reader will not fail to notice that such a necessary prakå†a is exactly what is being offered by Såyaña himself. After all, is it not with his bhåßya prepared according to those 14 ‘branches of science’, the most important of which remains Mœmå∫så understood as a sort of ‘investigation’ (vicåra), that one may expect to attain at the arthaj∞åna? To this effect the authority of Yåska is used (and re-cycled) once again, and his art of nirvacana36 is adopted. Extensive quotations from Nirukta on R¢gveda 10.71 are in fact re-interpreted by Såyaña with a view to showing the necessity of explanation in the form of a ‘proper knowledge’ (samyagj∞åna): The saying beginning with pada svam [… ] is an explanation of individual words, the saying beginning with jnåna [… ] is an explanation of the true meaning. By the saying of the third pada, the mantra addresses the right knowledge suitable for shedding light on the meaning of the Veda. (svamityådika∫ padavyåkhyåna∫ Ù j∞ånamityådika∫ tåtparyavyåkhyåna∫37 Ù vedårthaprakå†anakßama∫ samyagj∞ånamanayå tr¢tœyapådarüpayå våcå mantra åheti).38 34 It is again Yåska whose authority Såyaña is bringing for support when advocating for a kind of vidhi with respect to arthaj∞åna in the following passage: ittha∫ yåskena j∞ånastutyaj∞ånanindodåhara-ñasya prapa∞citatvådyacca stüyate tadvidhœyate iti nyåyenådhyayanavadarthaj∞ånasyåpi vidhirabhyupagantavyaΔ (Mller 1983: 17, ll. 16-17). 35 For the passage concerning the concept of the flower of dharma and the fruit of brahma see the section Flowery dharma and fruity brahma, below. 36 For a review of nirvacana method and tradition see Kahrs 1998. 37 Are there two ‘levels’ of meaning –padavyåkhyåna and tåtparyavyåkhyåna– to be elucidated? 38 Mller 1983: 16, ll. 34-35.

III. How to Produce, Construct and Legitimate a Tradition

335

In another passage Såyaña states bluntly that to understand the meaning of the Veda amounts to ‘properly see the mystery (rahasya) of the Vedic meaning/message with the help of fourteen basic sciences’ (and not to interpret it independently): […] this man here who has undertook a thorough study of fourteen sciences, he sees properly the mystery of the meaning of the Veda. ([…] tathåya∫ caturda†avidyåsthånapari†œlanopetaΔ purußo vedårtharahasya∫ samyak pa†yati).39

In this approach the meaning of the Veda appears to be as if located in some space outside the Vedic text itself, or in the relations it enters into with other Vedic and non-Vedic texts, relations to be detected, identified and explored with the help of someone who ‘knows the way’. Taking up again the metaphorical expression of the ‘invisible flower of the Veda’, a passage which refers to it explores the possibilities offered by Yåska’s nirvacana method while inscribing new meaning into the old text and its commentary. The rahasya of the Veda seems to have a connection with the dharma as well as with something which amounts to a ‘hegemony of the proper interpretation’. 2. The architecture of the Upodghåta Certain features of the construction of the Upodghåta as well as that of the commentary itself reflect, or, should we rather say, are made to look as reflecting, a royal authority behind it. The same could be noticed with regard to the authority of a Hindü deity (Gajånana) and that of a respected teacher (Vidyåtœrtha? and Madhåva): in the preambula to both Taittirœyasa∫hitå and R¢gbhåßya it is explicitly said that it was king Bukka Mahœpati himself who, inspired by the mysterious look of Gajånana ([Vidyåtœrtha]Mahe†vara?), ‘ordered’ the prakå†a of the vedårtha to be done. As we have an intermediary of Mådhava who (in the version of Taittirœyasa∫hitå Upodghåta) transfers the task to his brother Såyaña, there comes a charismatic figure of a sage and a sa∫nyasin to add more authority to the picture. It is interesting to have a quick look at the logic behind the construction of the arguments in the Upodghåta and behind the construction of the bhåßya. A simplified model of this architecture would look as the following: At the very beginning it is said that the Veda needs an artha- prakå†a which is ordered by the king Bukka Mahœpati himself induced by a mysterious ‘side-glance’ (ka™åkßa) of [Vidyåtœrtha]Mahe†vara. By the royal order it is compassionate Mådhava who is given the task;40 39

Mller 1983: 16, l. 40. yatka™åkßeña tadrüpa∫ dadhad bukkamahipatœΔ Ù ådi†anmådhavâcårya∫ vedârthasya prakå†ane [...] kr¢pålur mådhavâcåryo vedårtha∫ vaktum udyataΔ Ù (Mller 1983: 1, l. 3). Whom does the introductory formula actually belong to is another problem. It reiterates 40

336

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

The same prakå†a seems to be meant where, far later on in the Upodghåta, Såyaña says that it is not enough to study the Veda from the mouth of the teacher without getting knowledge of it, which however needs a prakå†a [...]; The Yåska’s explanation of R¢gveda 10.71 opens, culminates and ends the exposition of the rationale beyond the Updghåta, however cleverly reworked […]; arthaj∞åna can be known only after Yajurveda is commented first, as the knowledge of ritual is necessary for the knowledge of the vedårtha.

The reasoning of the Upodghåta is based on a cyclic backward reference built upon a våda discussion with an (imagined) opponent. Thus the idea of sa∫skåra, prominent to the passage concerning vidhi of studying and vidhi of understanding, is based on the previously mentioned idea of verbal refinement41 of the yaj∞a by Hotar, Udgatar, Adhvaryu and a mental refinement of the same by Brahman priest (a sort of exercise before actual ritual?), said to be practiced in order to avoid mistakes. The concept of textual body of the Veda and yaj∞a remains here triple-fold, like the old tråyividyå concept,42 and it is taken from Chåndogyopanißad 4.16.1. 3. The inner architecture of the prakå†a (R¢gvedabhåßya) On the other hand the whole body of Såyaña’s commentary appears to be quite intentionally distributed over textual units of the R¢gvedasa∫hitå. Clever use of the ‘artificial’ system of division in aß™akas, adhyåyas and vargas is visible in the placement of the metatextual introductory maõgala†lokas ending colophon-like formulas. The latter ones are inserted at the beginnings and ends, respectively, of every important unit of study.43 Borderlines of adhyåyas –as the most essential units– appear to be especially targeted: each and every adhyåya is made to start with a paraphrase of the benedictory formula beginning the whole bhåßya where explicit statements are made as for the divine, royal and religious authorities behind the Såyaña’s commentary. The ends are provided with a particularly elaborate in all Vedic commentaries ascribed to Såyaña, however in a slightly modified way. Perhaps it might have been authored by one of the panditas cooperating with Såyaña and commissioned with a sort of a final touch to the texts of the commentaries by supplying them with a similar benedictory formula giving also ‘historical’ circumstances of the text in question. 41 For an interesting interpretation of sa∫skåra as perfecting the person of the brahmacårin see Malamoud 1977: 52. 42 In most instances where the idea of the unity of the Veda is re-constructed the own wordings of Såyaña or quotations used by him avoid referring to Atharvaveda, thus (seemingly) excluding it from the body of the Veda which has one common artha. But on the other hand we should remember that Såyaña composed also a commentary to Atharvaveda with an introduction in which, among else, he comments on the concept of tråyœvidyå. 43 For units of Veda-study see, e.g., Scharfe 2002: 244-246.

III. How to Produce, Construct and Legitimate a Tradition

337

statement of the king Bukka’s royal legitimation behind Såyaña’s commentary (as well as, vice versa, religious legitimation of the kingly power by the act of Vedapravartana).44 A sort of colophon after each adhyåya looks as follows with an extraordinary way of paying respect to the patron and initiator of the commentary:45 By [this] prakå†a-explaining of the aim [artha] of the Veda one makes the darkness of tamas disperse. May Vidyåtœrthamahe†vara grant [him] the four aims (artha) of man. Here ends the eighth adhyåya-chapter of the first aß™aka of the commentary46 being an explanation of the Veda in honour of Mådhava and compiled by the noble Såyañåcårya most fit to bear the burden of the most sovereign [order of the] great Bukka, Lord of the Earth, the King of kings, the Suzerain and Propagator of the path of the Veda. (vedårthasya prakå†ena tamo hr¢da∫ nivårayan Ù pumarthå∫†caturo deyådvidyåtœrthamahe†varaΔ ÙÙ iti †rœmad råjådhiråja parame†vara vaidikamårgapravartaka †rœvœrabukkabhü-pålasåmråjyadhura∫dhareña såyañåcåryeña viracite mådhavœye vedårthaprakå†e r¢ksaμhitåbhåßye prathamåß™ake ß™amo dhyåyaΔ samåptaΔ ÙÙ)

Characteristically enough, Såyaña’s commentaries to other Vedic works bear similar ‘royal stamp’47 while the works on other subjects written by him, like vyåkaraña, ala∫kåra, artha, are conspicuously deprived of them.48 Moreover, a preliminary study of the extant man44

See Mller 1983: 109; 173; 235; 304; 365; 427; 495; 549.

45 Such a way of closure given to each unit of study (adhyåya, perhaps new division had

been introduced) entails a submission to the politically correct and religiously proper authority, that of Bukka, that of Vidyåtœrthamahe†vara and that of Mådhava Vidyårañya. 46 The maõgalavacana which opens the commentary as well as the division units of the sa∫hitå, gives impression of having been perhaps written by one of the collaborators to the project of Veda-commenting. The other problem is the opening verse to Gañe†a which occurs also in Jaiminœyanyåyamålåvistara, perhaps popular in use in the area and within the intellectual circles influenced by ‡r¢õgeri må™ha at that particular time. On the other hand, the name of Gajånana occurs also (I owe this remark to G. Oberhammer) in Våtsya Varadaguru’s Prapannapårijåtam (1.16,25-28) referring to a deity of Vißvakßena, a deity considered to be a Sena of Vißñu and said to destroy the obstacles (Oberhammer 2004: 68). 47 See MSS No 186 of Taittirœyasa∫hitåbhåßya in A Descriptive Catalogue of the Manuscripts in the Adyar Oriental Reaseach Library, p. 58 (iti †rœmad-divyayogœndra-vidyåtœrthamahe†varâparâvatårasya †rœmad-råja-adhiråja-råja-parame†varasya †rœvœrabukkamahåråjasyâj∞åparipålakena såyañâcåryeña viracite [...]); or MSS No 134 of Udgåtr¢tvaprayoga from A descriptive catalogue of the Manuscripts in the Library of the University of Bombay, 1953, p. 26: (the end reads: iti †rimadråjådhiråjaparame†vara [...] vaidika-mårga-sthåpanâcåryasya såyañâcårya-kr¢tau [...]). 48 As far as the preliminary stage of research in this respect permits for conclusions it should be pointed out that the evidence of manuscripts of Såyaña’s non-Vedic works features a totally different set of formulations in the extant colophons –characteristic formula of royal order with respect to the composition of the work in question is conspicuously absent (even if names of royal patrons are mentioned) either from the introductory stanzas or from the colophon. See e.g. colophon to the manuscript of Mådhavœyadhåtuvr¢ttiΔ, in particular MSS 402 (Descriptive catalogue of Sanskrit Manuscripts in the Adyar Library, vol. VI, p. 134: iti pürva-dakßiñapa†cima-samudrådhœ†vara-†rœkampa-råjasuta-saõgama-råja-mantriñå måyaña-putreña såyañena [...] sa∫pürñåΔ Ù), or MSS 403 (ibidem

338

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

uscripts in the manuscript libraries shows that Såyaña’s Vedic commentaries had been copied (and probably recopied) in most different scripts49 at use in South India (i.e. Grantha, Telugu, Devanågarœ, Malayalam), which might be an additional indication to their territorial circulation. 4. For the unity of the Veda and its artha(-prakå†a) Såyaña devotes a substantial part of the Upodghåta to a defense and a re-construction of the unity of the Veda. By ‘re-construction’ I mean reformulation of the idea of one single Veda with the help of the terminology of Mœmå∫så which reflects an initial de-construction of the Vedic text into mantra, vidhi and arthavåda portions done by Mœmå∫sakas. Såyaña’s idea of a unitary Veda is not only an apology of Mœmå∫så point of view –commencing from Mœmå∫så perspective that singles out vidhi-portion (injunctions) as the only authoritative part of the Veda, he re-establishes a place for both mantra and arthavåda portions of the Veda left almost aside by Pürvamœmå∫så thinkers. Both mantras and arthavådas are in the discourse of his argument supported by explanations and quotations and defended as authoritative enough to contribute to the unity of the Veda.50 Såyaña, however, escapes a definition of what really Veda is in the Upodghåta to R¢gvedabhåßya.51 In another work of his (Taittirœyasa∫hitåbhåßya Upodghåta), he rather unexpectedly appears to conceive the unitary Veda as a sort of ‘delimited and bounded text’ (grantha)52 made of three parts only (Atharvaveda apparently excluded). When it comes to definition within the introduction to commentary on R¢gsa∫hitå he sticks to and defends the old one, i.e. the one which says that the Veda p. 135), featuring unprecedented self-appraise on the part of Såyaña, or perhaps, respect on the part of the compiler or scribe (?) (iti pürva-dakßiña-pa†cima-samudrâdhî†vara†rœkampa-råja-suta-saõgama-mahåråja-mahå-mahå-mantriñå måyaña-suteña-mådhavasahodareña- såyañâcåryeña [...] sa∫pürñåΔ Ù). 49 See Catalogue of the Adyar Manuscript Library. Here the R¢gvedabhåßya is preserved in manuscripts written in Devanågarœ, Grantha and Telegu, while manuscripts of Taittœriyasa∫hitåbhåßya preserved are in Grantha and Telegu. Those preserved in Trivandrum are written in Grantha, Devanågarœ, Malayalam. 50 Mller edition ends up a long defense of the authoritativeness of arthavåda portions of the Veda with tasmåtsa∫bhåvitadoßåñå∫ parihr¢tatvådarthavådånåmasti pråmåñya∫ (Mller 1983: 10, l. 22). 51 Beside the traditional definition of a sum of mantras and bråhmañas, it is only when it comes to stating the object of the R¢gvedabhåßya itself that we meet with the formulation of so called catußtaya i.e. vißaya (the object), prayojana (the use of), sa∫bandha (the relation of the two) and adhikåra (the eligibility) with respect to the Veda as being itself the object (vißaya) of the (Såyaña’s) prakå†a-explanation. See Mller 1983: 17, l. 40 ff. 52 See Taittirœyasa∫hitåbhåßyabhümikå, p. 2, ll. 12-13 (in Vedabhåßyabhümikåsa∫graha 1934): “A book which informs about the divine ways to achieve the iß™a (the desired) and avoid the aniß™a (unwanted) is called ‘Veda’” (iß™apråptyaniß™aparihårayorlaukikamupåya∫ yo grantho vedayati sa vedaΔ). Somewhat later, line 20 reads: “therefore the definition of the Veda, that it is a souce of knowledge of the supernatural measures, is fully applicable” (tasmåd alaukikopåyabodhako vedo iti lakßañasya nåtivyåptiΔ) (tr. Bali 1999: 34).

III. How to Produce, Construct and Legitimate a Tradition

339

is a sum of mantras bråhmañas. One more instance of defending/reconstructing the unity of the Veda should not escape our attention either. This is the passage explaining the necessity of appropriating the Veda text (svådhyåya only, one’s own Veda) in its entirety. It is interesting also for a reference it makes to the conception of åpta –living men with authority comparable to that of scripture (another instance is that referring to paramaha∫sa)53: the mastery over the whole text has the purpose of japa-recitation.54 And it should not be doubted and said ‘if there be no indication [of the injunction to know], then the knowledge of the meaning would indeed not be attained’. For a means of cognition exists thanks to its own nature of indicating things to be cognized. And we perceive denotation of the sayings of living learned men [which are in force] without resorting to injunction. (kr¢tsnapråptirjapårtheti55 ÙÙ7ÙÙ na cåbodhakatve arthåvabodha eva na sidhyediti †aõkanœya∫56 pramåñasya prameyabodhakatvasvåbhåvyåt Ù laukikåptavåkyånåm57 –antareñaiva vidhi∫ bodhakatvadar†anådityåha ÙÙ).

However at the end he purposefully re-constitutes homology of the Veda as a subject for commentary, since if he would not do that, as he says quite logically, his commentary would be futile and useless.58 If we pay a closer attention to the construction of the Upodghåta it becomes clearer that Såyaña defends not so much the idea of the Veda’s unity as he does that of the usefulness of his own commentary. In the words put in the mouth of the opponent (pürvapåkßin) it is admitted that even if there existed such a thing as the Veda (in which the opponent strongly doubts), still the explanation thereof (vyåkhyåna) might not be authentic (apramåña), neither may it be of any use (anupayukta [Bali 1999: 70]) if that very thing be of no object or use itself: one cannot comment on something which is meaningless (and some mantras appear to look so!).59 The same idea is back with 53 See Mller 1983: 15, l. 31. The passage contains a †åstra quotation coming most probably from the work entitled Purußårthånu†åsana. 54 For another idea of a japa-recitation of the whole R¢gvedasa∫hitå and the benefactory results it is believed to procure see R¢gvidhåna, 1.67-71. 55 The quoted and commented sütra comes from Purußårthånu†åsana, a work highly respected and referred to as authoritative in many crucial moments of the introduction. 56 Which seems to indicate that the understanding of the Vedic text is still possible though it is not aimed by the injunction to study. 57 See Wezler’s remarks: “[…] conception of åpta, a person whose statements are a reliable means of valid cognition, a conception so important for the history of Indian philosophy and religion” (Wezler 2001: 223). See also Seyfort Reugg 1994: 303-320; Franco 1997: 29 ff., n. 38 ff. 58 See Mller 1983: 17, ll. 38-39: “if the Veda [itself, not the commentary] were without subject, use, etc., than the more so its explanation were without subject or use” (yadyapyetåvatprasiddha∫ tathåpi vedasya vißayådyabhåve vyåkhyånasyåpi paramavißayådika∫ na syåt Ù ato vedasya catuß™ayamucyate Ù). 59 See Mller 1983: 17, ll. 38-39: api vedasya vißayådyabhåve vyåkhyånasyåpi paramavißayådika∫ na syåt.

340

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Såyaña after a long discussion when he declares that he has to state the catuß™aya of his commentary. He says that he can do that only after stating it first a propos its very subject, i.e. the Veda. This creates another opportunity of declaring one Veda as the object of his commentary –the Veda that has its object, use, relation of one to the other and its ‘reader/user’ (adhikårin). In all instances of defending and re-constructing a unity to the Veda Såyaña makes use of different sources of authority, often rather freely, especially that of Yåska, whose nirvacana is reworked by him in order to support his own ideas. Establishing the commentary as a canonical and legitimized one seems, at least in my opinion, one of the main objectives behind the discourse of the Upodghåta aiming to show that there is one Veda which needs explanation. It constitutes an undercurrent of thought in the Upodghåta hidden behind the discussion on the Vedic unity and it’s expressiveness of the two subjects of dharma and brahma. Before actually stating that, however, Såyaña presents a spectacular defense of the unitary and authoritative Veda. In a series of impressively staged attacks the opponent tries to deconstruct the unity of the Veda and question its very existence on the grounds that: 1. No thing like the Veda exists, because there is neither definition (lakßaña) nor pramåña to it, and after admitting one offered by uttarapåkßin, he follows with the opinion, that 2. there is no authoritativeness of the constituent parts of the Veda defined in this way –the mantras are either unintelligible or they lack pråmåñya; and bråhmañas are made in part of arthavådas which are of no authority whatsoever and happen to be contradictory. 3. There happens to be noticed an inner contradiction within the Veda between mantras, arthavådas and vidhis. In defense60 of the of the unity61 of the Veda Såyaña resorts paradoxically to a logic that runs counter any idea of such unity, that is to the Mœmå∫så de-construction of the Vedic text into two kåñ∂as with the two distinct objects (dharma and brahma), and into mantras and bråhmañas. The latter ones are divided into injunctions (codana, vidhi –the Veda par excellance for Mœmå∫så, again divided into karmavidhi and j∞ånavidhi) and explanations/eulogies (arthavåda). He proves authoritativeness of each and every division, one by one. But eventually the same perspective allows him to reconstitute one Veda as the object of his commentary defined with the four elements: 60

For a systematically reconstructed apologetics of the Veda as pramåña see Oertel

1930. 61 Another, functional, definition of the Veda is given by Såyaña in the Upodghåta to Taittirœyasa∫hitåbhåßya (iß™apråptyaniß™aparihårayoralauki-kamupåya∫ yo grantho vedayati sa vedaΔ Ù).

III. How to Produce, Construct and Legitimate a Tradition

341

And here are the four [heads] as concerns the Veda: The subject matter of the two kåñ∂as –pürva- and uttara- –, in the Veda are respectively dharma and brahma. For you cannot acquire it from any other [source]. Therefore it is said in a sütra of Purußårthånu†åsana: ‘Both dharma and brahma are to be known only through the Veda’. And in the second sütra of Jaimini there is a twofold restriction also referred to by those who know the tradition: ‘codana, Vedic injunction, is the true means of cognition for the Veda. Indeed codana is the true means of cognition’.62 In order to establish the former one, in the fourth sütra it is refuted that the dharma is cognizable by sense organs: ‘Sense-perception is not the means (of knowing dharma) because it apprehends only objects existing at the present time’.63 (ato vedasya catuß™ayamucyate Ù vede purvottarakåñ∂ayoΔ krameña dharmabrahmañœ vißayaΔ Ù tayorananyalabhatvåt Ù tathå ca purußårthånu†åsane sütrita∫ Ù dharmabrahmañœ vedaikavedye iti Ù jaiminœye ca dvitœyasütre codanaiva dharme pramåña∫ Ù codanå pramå-ñameveti niyamadvaya∫ sa∫pradåyavidbhirabhihita∫ Ù codanaivetyamumarthamupa-pådayitu∫ caturthasütre pratyakßavißayatva∫ dharmasya niråkr¢ta∫).64

So finally Såyaña concludes asserting that he has established both the object (vißaya) and the use (prayojana) of the Veda in order to establish the same for his own commentary: Therefore because of impossibility of acquiring it [the knowledge] otherwise, the knowledge of dharma and brahma is the object of the Veda. And the knowledge of the two is the immediate use of the Veda [Veda is used for it]. (tasmådananyalabhyatvådasti dharmabrahmañorvedavißayatva∫ Ù tadubhayaj∞åna∫ vedasya såkßåtprayojana∫ Ù).65

5. Flowery dharma and fruity brahma A point of departure for the metaphor-like looking expressions of a ‘flower of dharma’ and a ‘fruit of brahma’ is R¢gveda 10.71 analyzed by Yåska, whose nirvacana explanation is quoted by Såyaña and eventually re-worked in order to suit the aims of his commentarial task. It is actually nothing else but three stanzas from R¢gveda 10.7166 that 62 In fact, this is not a quotation from Jaimini, but from a commentary (‡abara’s?). Såyaña uses abhihœta, not udåhr¢ta. 63 Translation from Jha 1916. 64 Mller 1983: 18. 65 Mller 1983: 18, l. 7. 66 R¢gveda, 10.71.4-6: “One who looked did not see speech, and another who listens does not hear it. It reveals itself to someone as a loving wife, beautifully dressed, reveals her body to her husband. One person has grown awkward and heavy in this friendship; they no longer urge him forward in the contests. He lives with falsehood like a milkless cow, for the speech that he has heard has no fruit no flower. A man that abandons a friend who has learned with him no longer has a share in speech. What he does hear, he hears in vain, for he does not know the path of good action” (tr. Doniger 1981: 61) (10.71.4a utá tvaΔ pá†yan ná dadar†a vÌcam utá tvaΔ †r¢ñván ná †r¢ñoty enàm Ù 10.71.4c utó tvasmai tanvà∫ ví sasre jåyéva pátya u†atÎ suvÌsåΔ Ù 10.71.5a utá tva∫ sakhyé sthirápœtam åhur naína∫ hinvanty ápi vÌjineßu Ù 10.71.5c ádhenvå carati måyáyaißá vÌca∫ †u†ruv̺

342

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

form a pretext and constitute a sort of framework67 in which Såyaña draws the rationale for an authoritative commentary of his own: And the other one, who is just a mere reciter, listens to the speech deprived of either the flower or the fruit. And the flower is the knowledge of dharma declared in the first kåñ∂a. And the fruit is the knowledge of brahma declared in the second kåñ∂a of the Veda68. As in the common world, a flower is a producer of its fruit, so the repetition of the Veda and other activities that constitute the knowledge of the dharma generate, through the gates of sacrificial rituals, a desire for the knowledge of the brahma having a nature of the fruit. (yastvanyaΔ på™hamåtraparaΔ pußpaphalarahitå∫ våca∫ †u†ruvånbhavati Ù pürva-kåñ∂oktasya dharmasya j∞åna∫ pußpa∫ Ù uttarakåñ∂oktasya brahmaño j∞åna∫ phala∫ Ù yathå loke pußpa∫ phalasyotpådaka∫ tathå vedånuvacanådidharmaj∞åna∫anuß™håna-dvårå phalåtmakabrahmaj∞ånecchå∫ janayati Ù).

By such formulation a relation among the object of the study, its nature and content, as well as procedure of attaining to its meaning is pre-defined. In the next stanza of R¢gveda 10.71 to be picked from Yåska, Såyaña under the guise of nirvacana resorts once again to rather metaphorical explanation of the passage and goes far beyond Yåska while attributing to it the meaning referring to the knowledge of 14 vidyåsthånas as characterizing the knower of the Veda. By this taking over of the authority of Yåska, Såyaña authorizes his own commentary in the shape he brings it to the reader: ‘About another one they say that he guards himself; and they do not stir him into the holy debates; He walks as a barren cow, with illusion; Though he sees and hears the speech it has neither flower no fruit’ –of which meaning is this: having quoted previously uta tvaΔ pa†yan, there is remembered a r¢k which is next to that one, and which brings better explanation of that previously spoken. It is capable of much better clarifying that meaning. How is it possible? It is said as follows: ‘And of the man who is versed in the fourteen branches of science they say also that standing firmly in the friendship of the Speech in the form of the Veda, he has come into touch with the drinking of the nectar of the meaning said in the Veda’.69 And those who know say so. aphalÌm apußpÌm Ù 10.71.6a yás tityÌja sacivída∫ sákhåya∫ ná tásya våcy ápi bhågó asti Ù 10.71.6c yád œ∫ †r¢ñóty álaka∫ †r¢ñoti nahí pravéda sukr¢tásya pánthåm Ù). 67 See how the same passage (R¢gveda 10.71.6) is commented upon with a view to stating the importance of svådhyåya study by the author of Taittirœyårañyaka, 2.15.6. 68 Såyaña never states what he actually means by the two kåñ∂as of the Veda. Does he, perhaps, refer to the previously composed commentary on Taittirœyasa∫hitå referred in the opening formula to the R¢gvedabhåßya Upodghåta or to Mœmå∫så in general? See Taittirœyasa∫hitåbhåßya Upodghåta (in Vedabhåßyabhümikåsa∫graha 1934: p. 5, ll. 28-29 [vedaståvat kåñ∂advayåtmakaΔ Ù tatra pürvasmin kåñ∂e [...] caturvidha∫ karma pratipådyam Ù]). Further Bali 1999: 43. A traditional notion of a division into karmakåñ∂a and brahmakåñ∂a is known perhaps from Taittirœyårañyaka. 69 It is not clear whether Såyaña presupposes any essential difference between vedårtha and vedoktårtha. See this brief remarks on the notion of artha: “Nor, finally, can artha be said to have an established referential role in Mœmå∫så. The Mœmå∫sakas do take artha very seri-

III. How to Produce, Construct and Legitimate a Tradition

343

(uta tva∫ sakhye sthirapœtamåhurneina∫ hinvantyapi våjineßu Ù adhenvå carati måyayå eßa våca∫ †u†ruvå∫ aphalåmapußpåm iti Ù Nirukta 1.19 Ù (R¢gveda 10.71.5) ayamarthaΔ Ù pürvodåhr¢tåyå uta tvaΔ pa†yannityådikåyå r¢co antaramevåmnåtå kåcidr¢k tasya pürvoktamantrasya bhüyase nirvacanåya sa∫padyate Ù tamartham-ati†ayena pratipådayitu∫ prabhavati Ù kathamiticet taducyate Ù api caika∫ caturda†åvidyå-sthånaku†ala∫ purußa∫ vedarüpåyå våcaΔ sakhye sthitvå sthairyeña vedoktårthåm-r¢tapånayuktamåhuΔ Ù).

Thus Såyaña affirms that the Veda should be known also as far as its meaning is concerned, though he does not seem to leave much freedom for anybody who would like to search for vedårtha on his own.70 The Vedic text should be rather received from the teacher along with its meaning.71 Where is then to be seen this dharma which is said to be one of the two objects of the Veda? Apparently, it is not to be seen: The dharma could be perceived only after the sacrifice is accomplished, and it does not exist beforehand, so there is no use here of the sense-perception cognition. But also afterwards, as it is deprived of form etc., it cannot be perceived by the senses. Therefore it is said by everyone to be invisible. As it is deprived of characteristic marks it cannot be an object to the inference. You might say that happiness and suffering are characteristic marks of dharma and adharma. Let it be so. But the existence of such characteristic mark and its bearer can ously, as the referent of words (†abdårtha), the purpose of actions (kriyårtha), and the motivating human goal (purußårtha). Artha’s rich multivalence enables the Mœmå∫saka to reinterpret texts, revise actions and rethink the role of performers, continually balancing the one against the other” (Clooney 1990: 283-284). 70 “And if the one who is studying the Veda does not understand the meaning, such a one indeed takes on a weight, i.e. he’s [merely] upholding it. By the word sthanu a dried root of a tree with cut off branches is meant. And as it [the sthanu] is used with the aim of [bringing] fuel and not with the aim of [bringing] flowers and fruits, so it is here with the one who is only reciting that he merely ‘does not fall into the state of an outcast’, and only this, nothing more. But his is not attainment of any fruit of the performing rites, as heaven and the like. And the word ‘kila’ expresses here a common sense. As much of honor with money and the like is a share of the reciter, indeed much more of that is seen in the case of the knower” (yastu vedamadhœtyåpyartha∫ na vijånåti so ’ya∫ pumånbhårameva harati dhårayati Ù sthånuriti dr¢ßñåntaΔ Ù chinna†åkha∫ †ußka∫ vr¢kßamüla∫ sthåñu†abdenocyate Ù sa ca yathendhanårthamevopayujyate na tu pußpa-phalårtha∫ Ù tathå kevalapå™hakasya vråtyatva∫ na bhavatœtyetåvadeva Ù na hyanuß™håna∫ svargådiphala-siddhirvåsti Ù kiletyanena lokaprasiddhirdyotyate Ù loke’pi på™hakasya yåvatœ dhanådipüjå tato ’pyadhikå vidußi dr¢†yate Ù) (Mller 1983: 16, ll. 12-16). 71 “No text received from the teacher without the knowledge of the meaning, and then only recited again and again will ever aflame, ever enlighten its own meaning. Just as in the place with no fire a dry piece of wood thrown in will never blaze up. Thus the Veda-nature of such texts would not come into the fore! ‘One gets to know an otherwordly means of attaining the object of human desire’ –this is the explanation of the words of the Veda. And it is said: the means which is not known through direct perception or through inference they learn through Veda. There from comes the vedatva of the Veda [therefore the Veda is called the Veda, i.e. ‘knowledge’]. So to attain at the essence of the Veda its meaning should be learned” (ki∫ca yadvedavåkyamåcåryådgr¢hœtamarthaj∞ånarahita∫ på™harüpeñaiva punaΔ punar-uccåryate tatkadåcidapi na jvalati svårtha∫ na prakå†ayati Ù yathågnirahita-prade†e prakßipta∫ †ußkakåß™ha∫ na jvalati tadvat Ù tathå sati tasya våkyasya vedatvameva mukhya∫ na syåt Ù alaukika∫ purußårthopåya∫ vettyaneneti veda-†abdanirvacana∫ Ù tathåcokta∫ Ù pratyakßeñånumityå vå yastüpåyo na budhyate Ù eta∫ vidanti vedena tasmådvedasya vedatå Ù iti Ù ato mukhyavedatvasiddhaye j∞åtavya eva tadarthaΔ Ù) (Mller 1983: 16, ll. 16-18).

344

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

be known only through the Veda. Therefore the Vedic injunction (codana) is the only valid means of cognition of dharma. (anuß™hånådürdhvamutpatsyabhånasya dharmasya pürvamavidyamånatvånna pratyakßa-yogyatåsti Ù uttarakåle ’pi rüpådiråhityånnendriyairavagamyate Ù ata eva adr¢ß™amiti sarvair-abhidhœyate Ù liõgaråhityånnånumånavißayatvamapyasti Ù suΔkhaduΔke dharmådharmayor liõgamiti cet vå∂ha∫ Ù ayamapi liõgaliõgibhåvo vedena-eva gamyate Ù tata†codanaiva72 dharme pramåña∫ ÙÙ).73

Eventually, a qualified Veda-adept is left no choice but the Veda itself as the only source of knowledge of the otherwise ‘invisible’ dharma. Though this is exactly the point of Mœmå∫så, the purpose of this final conclusion serves also, at least in my opinion, to support Såyaña’s own explanatory exposition of the Veda in the form of a prakå†a which had been ordered by the great Bukkamahœpati and sanctioned by the authority of his charismatic brother (probably referred to indirectly by the term of paramaha∫sa [Mller 1983: 15, l. 31], meaning here someone with an exclusive right and obligation to research into the vedårthåvabodha of his own). The complex argumentation concerning the actual contents of the vidhi-injunction to study the Veda (adhyayanavidhi) has a final message, which could be resumed as follows: One has to study the Veda, because this is enjoined by the †ruti itself. But this does not mean that one is necessarily to study it (according to the procedure of adhyayana) with the aim of understanding.74 Rather not. The vidhi-injunction to study is meant to be an independent one. Studying has independent value from that of understanding. On the other hand, a few passages later, Såyaña strongly advocates against ‘mere reciters of the Veda’ in contradistinction to the knowers of the Veda. And has in utmost contempt those bråhmañas whose ignorance does not allow the Veda they use to have a flower or a fruit. Their Veda, as he argues after Yåska, is ‘like a dry fuel put on the ashes –it will never blaze up’. This sort of argumentation suits not only the Mœmå∫så vision of the Veda, it serves well the purpose of establishing Såyaña’s commentary as a canonical one. For it is not adhyayana which is the right way to tap on the rahasya75 of the Veda (as rather unexpectedly voiced once by Såyaña). Only persons deserving a title of paramaha∫sa may and should do so. The right way of apprehending the meaning of the Veda is to get it from the teacher along with the help of 14 branches of knowledge through another procedure 72 There seems to be a shortcut here –in place of Veda the codana is substituted: perhaps having in mind the commentary of the Jaiminisütra making recourse to this deconstructive application of this substitution. 73 Mller 1983: 18, ll. 1-4. 74 A different approach to the problem of the actual content of the adhyayanavidhi can be seen in the argument with the opponent and perhaps it exemplifies two different positions in that matter of the schools of Kumårila and Prabhåkara. See Kurata 1994: 363-364. 75 Mller 1983: 16, l. 40: caturda†avidyåsthånapari†œlanpetaΔ purußo vedårtharahasya∫ samyak pa†yati. See also Mller 1983: 13, ll. 22-23.

III. How to Produce, Construct and Legitimate a Tradition

345

called vicåra.76 The Veda should be explained by the knowing one: it needs explanation (prakå†a). And here we are told by Såyaña that we are offered one –a true vedårthaprakå†a supported by the thorough knowledge of 14 branches of knowledge and the triple authority of Mådhava (a paramaha∫sa sa∫nyasin), king Bukka Mahœpati and (Gajånana)77 Vidyåtœrtha Mahe†vara. To support this enterprise Såyaña needs authority of the tradition to be reworked and re-apprehended. To this aid he uses the testimony of Yåska’s exegesis of one rgvedic hymn, i.e. that of R¢gveda 10.71, or rather just a few stanzas of it. And certainly the remarks of Såyaña directed against the arguments of nonintelligibility of mantras concern a procedure of establishing meaning which is dramatically different and far-removed from studying the Veda referred to by the term adhyayana: It is and being there yet it is not known. The meaning which indeed exists happens to be not understood due to errors, laziness and the like. The meaning of those [mantras like mentioned above] should be determined from the root with the help of parallel passages, art of etymology and grammar. (sataΔ paramavij∞ånamiti ÙÙ20ÙÙ [Jaiminisütra 1.2.49] vidyamåna evårthaΔ pramådålasyådibhir na vij∞ayate78 Ù teßåm nigamaniruktavyåkarañava†ena dhåtuto arthaΔ parikalpayitavyaΔ Ù).

Thus Såyaña seems to sharply differentiate between reciting/training/perfecting and reading/listening/establishing the meaning/interpreting. Both procedures have their distinctive and different aim and goals to achieve. In his most valuable study Oertel does not seem to take into account this situational aspect of Såyaña’s discourse treating his text as referring to generally accepted common sense concerning textual reading anytime and anywhere (Oertel 1930: 67). An important thing while reconstructing the train of thought of Såyaña is to become aware of his horizon of understanding. This is made by supposition that it is Yajurveda, or rather this what constitutes Yajurveda tradition revolving around yaj∞a, which is crucial for understanding R¢gveda as part of the Veda in its textual, ideological and performative unity. This horizon makes him refer in the Upodghåta to R¢gbhåßya not only to r¢k mantras or rgvedic textual tradition but to †ruti in general. This very perspective is actually the first and foremost thing he strives to establish while setting off to expose the ideology of his vedårthaprakå†a in his introduction. In a passage at the beginning of the Upodghåta, Såyaña takes pains to prove the priority of the task of understanding the purpose of Vedic 76

Probably meaning a sort of Mœmå∫så reasoning here. The elephant-faced one –perhaps Gañe†a. 78 See ‡abara’s commentary on this sütra (vidyamåno ’py arthaΔ pramådålasyådbhir no ’palabhyate Ù) and a slightly different impact of ’py (special case: even if) and evam (all cases). apalabh- connotes rather passive meaning while vij∞å a more active one on the part of the enquiring subject. 77

346

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

†rauta rituals first before the understanding of the meaning of the Veda (vedårthaj∞åna) in general, and that of R¢gveda in particular: Because arthaj∞åna is not only the knowledge of the meaning of the Veda sayings but also the knowledge of the purpose (artha) of these sayings within the context of the sacrifice.79 (astu eva∫ sarvavedådhyayanatatpåråyañabrahmayaj∞ajapådau r¢gvedasyaiva pråthamyam Ù arthaj∞ånasya tu yaj∞ånuß™hånårthatvåt tatra tu yajurvedasyaiva pradhånatvåd tadvyåkhyanam eva ådau yuktam).

We must remember that the exact reference of the term artha was a matter of discussion among traditional vaiyåkarañas as well as nairvacanas.80 6 In quest for the legitimation of power and authority Perhaps H. Kulke was right while saying that “[…] Vidyåtœrtha, Bhåratitœrtha81 and the two brothers Vidyårañya82 and Såyaña formed a most fascinating group of religious reformers and creators of a new religious institution. Due to their immense philosophical and literary activities their work had a tremendous impact on Hinduism that lasts till today”. More intriguing however are lines by the end of Kulke’s article: “Not Sringeri and its jagadguru Vidyårañya established Vijayanagara, but the kings of Vijayanagara established Sr¢õgeri with its new ‘‡ankara tradition’ for their own political ends”.83 Kulke is aware of the possibility of understanding such inter79 See also his discussion about the necessity of the knowledge of the concept of quotation/indication by mantrapratœka: “But there is a reference to a characteristic mark [of a deity] which serves that purpose. There is a verse of the †ruti ‘let him approach the Agnidhra-priest with an Agneyi’. Here is the meaning of it: Whichever hymn has Agni as its deity it is [called] Agneyi. With this let him approach the place of the Agnidhra-priest. Here the bråhmaña which indicates this approaching does not indicate ‘With that of agne naya let him approach’ but, having read [having quoted] the sign of the mantra, it indicates this with the characteristic mark of the Agneyi. Whenever in a stanza the mention of Agni is made in a prevalent way, then Agnirdevata is the deity of that stanza. Being so, with the word ‘Agneyi’ there takes place an indication of taddhita suffix expressing a devatå. This indication makes one understand that the ‘saying of this mantra is meaningful’.” (liõgopade†a†ca tadarthavaditi Ù22Ù ågneyyågnœdhramupatiß™heteti †rüyate Ù tasyåyamarthaΔ Ù agnirdevatå yasyå r¢caΔ seyamågneyœ Ù tayågnœdhrasthånam upa-tiß™heteti Ù tatra hyupasthånamupadi†adbråhmañamagne naya (R¢gveda 1.189.1) ity ana-yopatiß™heteti mantrapratœka∫ pa™hitvå nopadi†ati ki∫tu ågneyœtvaliõgenopa-di†ati Ù yadå tasyåmr¢cyagniΔ prådhånyena pratipådyate tadå tasyå r¢co ’gnirdevatå bhavati Ù tathå satyågneyyeti devatåvåcitaddhitåntanirde†a upapadyate Ù tasmåd-ayamupade†astanmantravåkyam arthavaditi bodhayati Ù). 80 For details see Kahrs 1998: 39-50. Kahrs often resorts to illustrating the position of Yåska through the words of Såyaña, but does not count him as an important representative of nirvacana school. 81 Names of respectively the guru of Såyaña and Mådhava and Mådhava’s predecessor on the throne of the Sr¢õgeri Vidyå Pœ™ha. 82 Another name of Mådhava which he might have acquired after his attaining the sa∫nyasa stage. Mådhava-Vidyårañya should however most probably be distinguished from Mådhava-mantrin with whom he is often erroneously identified. See Kulke 1993: 128-129. 83 Kulke 1993: 238.

III. How to Produce, Construct and Legitimate a Tradition

347

pretation as “an example of a gross manipulation of religious institutions”. So I am aware of a danger of ‘over-interpretation’ of the motives behind Såyaña’s commentary. The subtitle of this short study points to something which might look a bit bizarre for a representative of Mœmå∫så school of thought as probably Såyaña was. Has not Ch. Malamoud noticed already that “[…] le Veda sur lequel se fonde cette écolle pour édifier son système est partielle: ne sont vraiment à prendre en considération que le portions du Veda qui sont de structure vidhi”?84 The metaphor of the ‘flower of dharma’ is referred to here for the primary purpose of showing the working of tradition in its resilience and adaptiveness. To adapt to the changing socio-cultural context a tradition has to develop certain attitude towards itself, an image of self-reflection and the self-reflection as a topos embarked upon by its elite intellectuals. For the Vedic exegetical tradition turned political, the flower of the dharma has to remain invisible in order to sustain and re-use the traditional notion of Veda as propounding dharma. Only if the dharma is invisible remains it beyond the scope of means of cognition other than that of †abdapramåñya, or valid testimony, of the †ruti. This being so, the power of deciding about dharma remains in the hands of those who hold control over the Vedic pråmåñya. And the control means here securing the right exegesis and the use of particular specified commentary. The deconstruction of the Vedic corpus done by Mœmå∫sakås is taken over by Såyaña (inspired by his brother Mådhava) as a powerful tool in the task of establishing a ‘canonical’ text, i.e. a commentary that would be able to carry and implement a double authority and superiority in religious matters –that of the Vijayanagara kingly rule and that of Mådhava’s ma™ha (Sr¢õgeri Vidyå Pi™ha). Both pertain to religious and political domination, i.e. to the sphere of power. On the other hand, the same commentary is presented in a way that strongly suggest a royal and religious ‘seal’ to its authority. We might ask how could this kind of executing power through powerful text be implemented in practical terms? The key to understanding the working of such implementation lies perhaps in identifying possible addressees of such a commentary in the historical context of late 14th century South India. The project of producing such extensive commentarial set of works must have some other aim than only a general revitalization of Vedic culture. The structure of political power and religious authority in the late medieval South India was conspicuously characterized by multiple centres of power and authority. The centres of authority must have remained within the hands of priestly (mostly brahmanic) collective or family bodies presiding over big temples and other religious institutions, like ma™has, which often constituted not only religious but also economic cores of the land. And perhaps this plurality of authority centers is what should be taken into account while 84

Malamoud 1977: 69.

348

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

explaining the early Vijayanagara rulers’ need of some unifying principle that could appeal to most of them. This plurality of authority centers in South India has been referred to by modern historical conceptions, one of them being that of ‘segmentary state’85 by Burton Stein. Whatever contemporary to date Hindü religious affiliation of the presiding temple bodies could be, an at least outward respect for the Veda as embodying dharma must have been of certain appeal to most of them. Though I am rather far from dismissing the question of the beginnings of the Vijayanagara by subscribing to a simplistic statement about its mission for a rescue of the Hindü world from the challenge of the Muslim power, I also think that such an idea could be cleverly used by the early rulers and their supporters as a pretext accounting for their political ambitions of dominance over the religiously and ethnically divergent South. In this respect we should not forget that dharma remained, to use Halbfass’s words “[…] the main concept of traditional Hindu xenology, as the standard used to demarcate the Aryan from the mleccha” (Halbfass 1988: 311). However a confirmation of such a hypothesis is awaited to come from more sustained study concerning manuscript dissemination in South India in late medieval period. It is beyond doubt that Vijayanagara rulers had at their disposal resources needed to have a piece of a scholarly work like that of Såyaña’s commentary distributed over a large area.86 Thinking about the actual range of such text-distribution I would tend rather to assume that it must have been a something like targeting carefully chosen intellectual centres87 and centres of religious authority –rather than blindly augmenting the volume of copies. Those centres must have comprised courts, temples, and, last but not least, educational institution and libraries. We know from South Indian medieval inscriptions recording donations about huge endowments granted to education (sometimes as an act of vidyådåna) which in late medieval times of the Vijayanagara empire already went through much transformation and developed institutional forms (Altekar 1934: 284-293). These were of several types (i.e. ma™has, devasthånas, gha™ikåsthånas) located often by rich and influential tem85 The concept itself is taken over and re-worked by Stein from Southall (Alur Society) and the field of African studies. 86 There’s not much evidence (at least not known to me) concerning the reality of medieval manuscript circulation and knowledge transmission or, what Pollock calls “morphology of manuscript culture in its material and economic aspect” (Pollock 2004: 2). We should perhaps draw more conclusions from the material available while paying attention to recently re-formulated theories of cultural and knowledge transmission (vide the formulation of Pollock’s about ‘script- mercantilism’). Among scanty information concerning economic aspect of manuscript distribution there’s relation of Al-Bœrünœ about Ugrabhüti, the teacher of king Anangapåla (ca. 11th c. AD) who happened to write a book on grammar called ‡ißyahitåvr¢tti and sent it to Kashmir, but the Kashmiri scholars refused to adopt it. To support the project, his royal disciple “[…] sent a sum of 2000.000 dirhams (Rs 60.000) to Kashmir for being distributed among those who would study the book of his master” (Altekar 1934: 143). 87 For the idea of multiplied civilizational centres see Cohn 1983: 78-87.

III. How to Produce, Construct and Legitimate a Tradition

349

ples, they must have been important and influential intellectual centres. In great number of them Vedas and †åstras were studied and commented upon (Altekar 1934: 148). We hear about the libraries by those institutions –one such existing at least from 13th c. in Cidambaram was called Sarasvatœ Bhañ∂åra.88 All those ‘temple colleges’89 must have been targets of the manuscripts of Såyaña’s Vedabhåßya (preceded by the introduction and furnished with the proper colophon as a form of ‘royal seal’) ordered to be produced and circulated by kings of Vijayanagara desiring to support relations with and gain recognition from sometimes far-removed ma™has of the territories they claimed. We do not know much about the nature and technical means of such manuscript circulation praxis. We do not know particularly whether in this specific case it was the whole sa∫hitå with the bhåßya attached in a manner of commentaries to †åstric works (at least Upodghåta has actually strongly †åstric character, its construction following the rule of våda discussion with an imagined opponent) or rather separately to it. The evidence of the extant manuscripts attests rather to the latter case. Some of the original traditional schooling system of vedapå™ha†ålås survived even in modern times, or has been revived after a century or more break, much along the traditional lines developed in medieval period. It shows strong affiliation of the local schools with powerful må™has. At least two of them in contemporary South India run or support a veritable independent network of local på™ha†ålås or smaller ma™has. They set curriculum, decide on the textual basis of the teaching of the Vedic texts, as well as on their interpretation. The two most influential ma™has of the South, that of Kå∞cœ Kåmakotœ and Sr¢õgeri developed their own separate institutions, carefully planning and administering Sanskrit and Vedic schooling network under their jurisdiction. This picture may not be very far removed from that of Medieval South India,90 when both centres formed their identity boasting of royal patronage. The rulers of Vijayanagara must have been aware (especially in the formative period of the empire) of the powerful authority of such institutions and perhaps helped creating the image of one in order to keep this influential structure on their side and use it as a powerful tool of legitimation for their imperial hegemony. That a temple and a †ålå institution could actually and actively influence political life in medieval South India is evidenced for instance by the Kå∞cœ Vaikuñ™haperumål inscription telling the story about a senior staff member (gha™ikaiyår) of the temple ghañikå who took an extraordinary initiative towards securing a coronation of a new king to the empty throne after the 88 Scharfe 2002: 183. Another library –Sarasvatœ Bhavana– in Tirunevelli district is mentioned in South Indian Inscription n. 695 (Altekar 1934: 293). 89 For a review of such ‘temple colleges’ and education see Scharfe 2002: 166-193. 90 “[…] down to the 18th century every religious center in south India used to maintain a Sanskrit På™ha-†ålå” (Altekar 1934: 293).

350

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

death of Pallava Parame†varavarman II (730 AD) (Sankaranarayanan 1993: 9). Now, it could be argued that Vedic education used to (and to some extent still does) do well for centuries, if not millennia, without resorting to writing and manuscripts, even refraining from doing so. Is it not the authority of Påñini†ikßå which warns against resorting to writing,91 and that of Mahåbhårata which condemns to hell those who profess by writing the Vedas (does it mean that there were actually some who did it?)?92 Well into the era when writing was widely used for literary purposes in India we still hear authoritative condemnation of writing used for Vedic studies.93 Even long after Vasukra’s bold enterprise (9th or 10th cent. AD), in the second half of 14th century, it is no one else than Såyaña himself who strongly advocates against reading from written version of Vedic text while studying the Veda during svådhyåya.94 The same means, however, that the written versions must have been at hand and, no doubt, consulted when needed during other occasions and other type of textual lecture than that of svådhyåya. Characteristically we don’t hear about explicit prohibiting of japa-type recitation being done with the help of written text (common contemporary practice among bråhmaña gr¢hasthas). The double (or even multiple) representation of Vedic study and text-lecture puts Såyaña somewhere in between Pürva and Uttara Mœmå∫så.95 Does it mean that he was eclectic? Do we have any right of depriving him of his own choice which, motivated by somewhat different forces than exposition of this or that doctrine made him think in his own way within the frames of the socio-cultural conditions of his 91 gœtœ †œghrœ †iraΔkampœ tathå likhitapå™hakaΔ Ù anarthaj∞o ’lpakañ™ha† ca ßa∂ ete på™hakådhamåΔ Ù (Påñini†ikßå, 32). 92 ņvamedhikaparvan, 106.92: “Those who make money on the Vedas are wicked and sinful, those who commit the Vedas to writing are bound for hell” (vedavikrayiña†caiva vedånå∫ caiva düßakåΔ Ù vedånå∫ lekhakå†ca te vai nirayagåminaΔ Ù). 93 See Kumårila Bha™™a, Tantra-vårttika, 1.3, p. 86 (p. 123, l. 20 in K.V. Abhyankar’s ed.). It is Kumårila, who in VII c. “[…] reasserted (in writing, of course) that learning the Veda from a concrete text-artefact –‘by means contrary to reason, such as from written text’– could never achieve the efficacy of the Veda learned in the authorized way, ‘by repeating precisely what has been pronounced in the mouth of the teacher’” (Pollock 2004: 4). See also Scharfe 2002: 8, n. 2, referring to Al-Bœrünœ, who “[…] reported in the eleventh century: ‘They do not allow the Veda to be transmitted to writing’”. 94 See Upodghåta to R¢gvedabhåßya (Mller 1983: 14, l. 15): “As the injunctions to perform sacrifice do indeed require the knowledge of their dominion [of application], for this knowledge they apply svådhyåya. And as the injunction to study excludes reading from written words, they understand ‘perfection of study’ as the svådhyåyana-study. So in appropriation of both of them lies the attainment of this [result of ritual perfection]” (kratuvidhayo hi vißayâvabodham apekßamåñås tadavabodhe svâdhyåya∫ viniyu∞jate Ù adhyayana-vidhi† ca likhita-på™hâdi vyåvr¢tyâdhyayana-sa∫skr¢tatva∫ svâdhyåyasya gamayanti Ù ata ubhayôpådånåt tatsiddhiΔ Ù). 95 See, for instance, his attitude towards the result assigned to the Vedic knowledge of the procedure of ritual. He voices an opinion, and provides relevant quotations, that sound knowledge is as good as performing rituals per se. On the other hand he advocates that knowing and performing results in more merit than mere possessing knowledge of the rituals (appropriating the Veda to that extant), which brings him much closer to Pürvamœmå∫så positions. Yet he refers to the division of the Veda into dharmakåñ∂a and brahmakåñ∂a, hold-

III. How to Produce, Construct and Legitimate a Tradition

351

historical situation. After all, is it only his single instance of being determined by history and particularity of situation he found himself in? The success of Såyaña’s commentary in modern Indian Vedic exegesis is somewhat difficult to account for.96 We do not know much continuity to his thought. Except an abbreviated version of R¢gvedabhåßya by Mudgala –a rather unsuccessful enterprise judging from the extant copies of its manuscripts–, we know only a name of a commentator Ravana whose work has not survived. Some sort of Vedic exegesis remained in fashion among intellectuals in pre-colonial era, however differently motivated it must have been. An example of Nœlakañ™ha, who not uncritically refers sometimes to Såyaña’s work, might be given as a representative of another use made of an exegetical strategy. The nature of relation of the latter to Såyaña has been summarized thus by Minkowski: While Nœlakañ™ha makes regular use of the glosses of the commentator Såyaña, he denies Såyana’s hermeneutic assumptions about the Veda’s ritual application, indeed even as he denies the centrality of the commentator’s elucidations. While Nœlakañ™ha invokes Yåska as representative of the Nighañ™u-nirvacana tradition to pen up the possibility of reading the Vedas on several layers of meaning simultaneously, he never limits himself to the particular meanings Yåska has assigned to the verses. (Minkowski 2003: 238)

7. Orality of the Veda and literality of the †astric commentary The aporia of memorizing-perfecting/reading-understanding that seems rather puzzling in Såyaña might refer also to the duality of oral-writing in Sanskrit på™ha†ålås of his time: on the one hand timeperfected, and protected by authority of tradition, oral system for ing the former one to be inferior and elevating the other one (with a notable indication towards Upanißads) as the higher one, what makes him an advocate of Vedånta position in this respect. The same pertains to his treatment of the meaning of svådhyåya. For discussion of Pürvamœmå∫så and Uttaramœmå∫så views on svådhyåya see Malamoud 1977: 44-67. It seems not so easy to determine whether Såyaña is more inclined to take the stance of the Prabhåkara school of Mœmå∫så or that of Kumårila. Several instance point to the former case. For instance, while enumerating five prayojanas of vyåkaraña (Mller 1983: 19, l. 24) he seems to accept all the five instead of explicit reluctance to do the same on the part of Kumårila (see Ramachandrudu 1994: 244) who is ready to accept their role only as one among other vedåõgas. Accepting six aõgas with vyåkaraña (of the Påñini school) as necessary in studying the Veda and its artha by Såyaña runs also against the opinion by Kumårila. But other scholars take Såyaña’s views as supporting Kumårila school especially in his discussing the problem of the extent of the injunction to study the Veda (adhyayanavidhi), which according to Såyaña cannot extend primarily to the knowledge of meaning on the account that a single vidhi cannot serve two purposes (Kurata 1994: 364). 96 It might be argued, however, that the old archetype-model of vedåõgas as developed for preserving the Veda has been re-established with new vitality by Såyaña and the Vedic tradition in a way re-defined. With this holistic vision the Veda after Såyaña was no longer the same as it used to be before. Of course not in the textual sense but as a type of social and cultural experience regulated by methodological norms legitimized by the form of a canonical commentary.

352

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Veda learning and textual perfection, and a strong presence of vedåñga-studies with prominent vyåkaraña and Mœmå∫så studies97 referring widely to writing on the other. The commentary by Såyaña belongs to the world of writing and manuscripts but refers to double world of the Veda itself (oral) and its explanatory studies (using sources committed to writing). Historical evidence concerning the schools of Vedic training that since medieval time seemed to have taken over the education, shows certain characteristic traits. First, combining Vedic studies with those of the †åstras (as further initiation after completing Vedic curriculum),98 and, second, the presence of written manuscripts and often their library-like collections99 in the same places, as †åstra-studies required it. Writing a commentary with a rationale for the same in the way that Såyaña did it, places him not only within the history of Vedic commentators but also in the tradition of several †åstras, like that of nirvacana to which in practical terms he seems to be mostly indebted (Kahrs 1998: 28-33; 46-49).100 The extensive quotations from Yåska in his Upodghåta as well as the nirvacana method being time and again made recourse to within the body of the commentary itself, testify to Såyaña’s close affinity or predilection to nirvacana tradition at one hand and a clever use of the method as well as ideology of Yåska for his own enterprise. The point of departure for the two intellectual enterprises distanced by, no less than, two millennia, is somewhat similar: a conquest for regaining the lost vedårtha. For Yåska it meant initializing new tradition through a regular exposition of the method of nirvacana (we know however that Yåska had a chain of predecessors behind him). The method was needed as –in the formulation of Wezler– an Ersatz for the discontinued tradition of upåde†a which used to accompany the 97 For the evidence concerning the curriculum of medieval Sanskrit schooling system with the institutions like på™ha†ålås, ghå™ikas, etc., see Keay 1918: 54; Sankaranarayanan 1993: 22. 98 Institutions like that, named gha™ikå, vidyåsthåna, ma™ha, på™ha†ålå, etc., must have been in existence in South India at least from 4th-5th c. AD and continued to function well into late medieval period. See Guñ™upalli Pillar Inscription of ‡ålaõkåyana Nandivarman recording a journey by a bråhmaña to Kå∞cœ for examination to an institution of that kind controlled by Pallava kings (Sankaranarayanan 1993: 7). For later instances like that of XI c. AD Gaõgaikoñ∂a†olan by the temple of Narasi∫ha Perumål at Eññåyiram (South Arcot district) founded by Råjendra Col¢a I to teach several †åkhås of the Veda along with vyåkaraña, Pråbhåkara Mœmå∫så and vedånta. See also later 13th-14th c. AD examples in Sankanarayanan 1993: 8-32. 99 See the famous Sarasvatœ Bhañ∂årå at Cidambaram, referred to in Sankaranarayanan 1993: 28, 32. The custom of maintaining libraries by temples continued also in later medieval period: the Kå∞cœpuram Varadaråja temple inscription of 1359 AD (years of Vijayanagara Bukka I reign) records how the temples authorities provided for the proper preservation of a collection of manuscripts stored in a ma™ha by the temple. See Inscription No. 574 in Annual Report on Indian/South Indian Epigraphy, Dept. of Epigraphy, Archeological Survey of India, 1919; Epigraphya Indica, vol. 25, pp. 325-26 (referred to by Sankaranarayanan 1993: 32, n. 2). Another example of an inscription referring to a grant for a library (Sarasvatœ Bhavana) in Tinevelli district is in South Indian Inscription, No. 695, 1916 (referred to by Altekar 1934: 293). 100 Kahrs, however, does not include Såyaña explicitly within the nirvacana tradition.

III. How to Produce, Construct and Legitimate a Tradition

353

transmission of the mantras in the times of såkßåtkr¢tadharmånaΔ (r¢ßayaΔ who had perception of the dharma) (Wezler 2001: 237).101 The tradition of upade†a (oral technique of explaining the mantras, or “[…] a method of teaching students” [Wezler 2001: 237]) seems to have ceased to continue since the later generations of teachers, deprived of the faculty of their predecessors, “[…] unfortunately and for unknown reasons became averse of delivering on this instruction” (Wezler 2001: 238). Whatever historical probability of representing the actual situation of Vedic hermeneutical tradition in the Yåska’s times could this statement have, the interest of the present short study lies rather in the logic behind the intellectual enterprise of a commentarial work. The point of departure for Såyaña is somewhat similar at first: he is going to present a vedårthaprakå†a in order to cope with the situation of the lost meaning of the Veda. However he does not mean to initialize a method (however ‘crossing the boundaries of †astric disciplines’ would sound much more positive and innovative than the epithet ‘eclectic’ stuck to him might suggest), but rather integrate existing methods in a way that might be accepted as a standard and canonical commentary able to protect the Veda from free- and over-interpretation. On the other hand the prakå†a is to supply a qualified practitioner in what is lacking in the system of Veda transmission –in itself a powerful tool of the working of tradition. Such pratictioner should not be allowed, or moreover induced, into (as implied by pürvapåkßin), offering a random and haphazardous interpretation. What the system itself is lacking (and in reality supplied differently by different religious traditions of medieval Hinduism) is the upade†a. The latter one is needed to account for all the profound meaning of the mantras in particular and the Veda in general (here Såyaña takes a decidedly different stance from that of the Yåska) and the whole wide realm of dharma, which, after all, is what the mantras are somehow to embody and express. There is one more thing to this logic behind the vedårthaprakå†a and R¢gvedabhåßya: after the very last lines of the introduction, when a lengthy quotation of four stanzas describing the goddess’ of knowledge advise to a Veda teacher, the gloss by Såyaña ends with the following words: ittha∫ vidyådevatayå prårthitatvåd åcåryeña mukhya†ißyåya vedavidyopadeß™avyå Ù tadartha∫ r¢gvedo ’småbhiΔ ßa∂aõgânusåreña vyåkhyåyate Ù. None of the three existing translations seems to take into account the compound tadartham as referring to the preceding line, i.e. to the situation of supplying atide†a by a teacher to a student who deserves it.102 At least to my mind the aim of the compound is just that –‘the 101 Nirukta, 1.20: såkßåtkr¢tadharmåna r¢ßayo babhüvus te ’varebhyo ’såkßåtkr¢tadharmabhya upade†ena mantrån sa∫prådur upade†åya glåyanto ’jare bilmagrahañåyema∫ grantha∫ samåmnåsißur veda∫ ca vedåõgåni ca Ù. 102 See Peterson 1974: 88 (“Therefore we will now explain the ¥gveda in accordance with its six Aõgas”); Bali 1999: 206 (“We here give the commentary of ¥gveda in accordance with the six Vedangas to maintain the Vedic tradition”); Ray 1961: 322 (“And with

354

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

R¢gveda is here explained by us according to six aõgas, in order to make it possible for a teacher to transmit this explanation to his preferred student’. Isn’t it the upadeß™avyå an echo of Yåska’s upade†a? I can’t help the feeling that the line sounds technically didactic, however uninteresting it might seem. Was the enterprise of commenting the Veda a part of setting standards of the Veda teaching regionally along the lines accepted by the Sr¢õgeri ma™ha, and under the guise of complacency with 14 vidyåsthånas? This ‘educational project’ (aimed at authoritative centres of Vedic educations located often by big influential temples), would easily fit into something what Hacker termed ‘intentional act of cultural politics on the part of the rulers of Vijayanagara’. It should be noted however, that an overemphasis on the role of Vedic commentary project as the most influential enterprise carried out by the Vijayanagara rulers with a view to securing legitimation for their imperial power runs a risk of aberration inflicted to a general historical reality of the times. Much more cautious attitude would allow seeing this intellectual, educational and propaganda project as just one part of a multilayered strategy of Vijayanagara rulers striving for acknowledgment for their political hegemony within a much broader cultural context. We should not forget about all those elements of courtly etiquette adopted steadily by Vijayanaga rulers in order to appeal to the standards of legitimation that must have had in late 14th centre a much broader sphere of reference than regional Hindu one. These must have drawn widely from Islamicate103 standards and in this way might have appeared abusive in the eyes of Hindü traditionalists connected to big temple centres. Viewed from this perspective the enterprise of extensive Veda commentary with inscribed royal legitimation in it might well have been meant as a way of doing away with such hesitations on the part of Hindü religious authorities. In the vision of the Veda drawn by Såyaña in his R¢gvedabhåßya Upodghåta, the old model of vedåõgas, once developed for the preservation of the Vedic ‘scripture’ and later turned into independent †åstras, is reestablished for the purpose of vedårthaprakå†a. With a new commentary the Vedic tradition is re-defined again in order to suit the ideas of influential religious thinkers and political aspirations of the kingly patrons. It is a holistic vision and the Veda the object of knowing in view, and for such deserving pupils Rigveda is now being explained by us in the line of or in accordance with the six angas of the Veda”). 103 Those other elements of such a conscious cultural and state policy that might appeal also to the Islamicate modes of representing legitimation are referred to in Wagoner 1996. According to Wagoner a most important characteristic of Islamicization of Vijayanagara’s courtly life can be seen in adopting new system for men’s courtly dress (long tunic and conical high cap) which according to him “was a deliberately calculated act on the part of Vijayanagara’s courtly elite […] paralleled by the appropriation of Islamicate modes of political language, […] adoption of the title Hindu-råya-suratråña, literally Sultan among Hindu Kings”, seen by Wagoner as “integral part of Islamicization as a broader process of cultural change” (Wagoner 1996: 853).

III. How to Produce, Construct and Legitimate a Tradition

355

after Såyaña is no longer the same as it used to be before. Not in the textual shape, which is to remain intact, but as a type of social and cultural experience regulated by methodological norms legitimized by the form of a canonical commentary.

BIBLIOGRAPHY Aithal, K.P. (1991), Veda-lakßaña. Vedic Ancillary Literature. A Descriptive Bibliography (Beiträge zur Sdasienforschung, Sdasien Institut, Universität Heilderberg), Franz Steiner Verlag, Stuttgart. Altekar, A.S. (1934), Education in Ancient India, Benares. Baladeva, U. (ed.) (1934), Vedabhåßyabhümikåsa∫graha, Krishnadas Academy, Varanasi. Bali, S. (1999), Såyaña’s Upodghåta to the Taittirœya Sa∫hitå and the R¢gveda Sa∫hitå, Pratibha Prakashan, Delhi. Bronkhorst, J. (1997), Philosophy and Vedic Exegesis in the Mœmåμså, in «Poznan Studies in the Philosophy of the Sciences and the Humanities», vol. 59, pp. 359-371. ______ (1998), Does the Veda have an Author?, in «Asiatische Studien», vol. 52, n. 1, pp. 5-14. Certeau de, M. (1975), L’écriture de l’histoire, Gallimard, Paris. ______ (1990), Invention du quotidien, Gallimard, Paris. Clooney, F. (1990), Thinking Ritually. Rediscovering the Pürva Mœmå∫så of Jaimini, Publications of the De Nobili Research Library, Wien. ______ (1994), The principle of upasa∫håra and the development of Vedånta as an Uttara Mœmå∫så, in Dwiwedi, R.C. (ed.), Studies in Mœmå∫så, Motilal Banarsidass, Delhi. Cohn, B. (1983) [1958], Networks and Centres in the Integration of Indian Civilisation, in Id., Anthropologist among Historians. Collected Essays, Oxford University Press, Delhi. Doniger, O’Flaherty W. (trans.), The Rig Veda: an anthology, Penguin Books, New York 1981. Deutsch, E. (1989), Knowledge and the Tradition Text in Indian Philosophy, in Deutsch, E., Larson, G. (eds), Interpreting Across Boundaries: New Essays in Comparative Philosophy, Delhi. Dhadphale, M.G. (1982), Nirukta: its precise purpose, in Dharmadhikari, T.N. (ed.), Golden Jubilee Volume, Poona, pp. 94-104. Dsa, F.X. (1980), ‡abdapråmåñyam in ‡abara and Kumårila. Towards a study of the Mœmå∫så experience of language, Publications of the De Nobili Research Library, Wien. Figueira, D. (1994), The Authority of an Absent Text, in Patton, L. (ed.), Authority, Anxiety and Canon, State University of New York Press, Albany, pp. 201-234. Filliozat, V. (1977), L’epigraphie de Vijayanagara du debut.

356

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Franco, E. (1997), Dharmakœrti on Compassion and Rebirth, Wien. Galewicz, C. (2003), A keen eye on details: Reviving ritual perfection in Trichur Somayaga 2003, in «BEI», vol. 21, n. 1, pp. 239-253. Gonda, J. (1975), Vedic Literature, Harrassowitz, Wiesbaden 1975. Gopala Bhatta, R. (1910), Mœmå∫sådar†ana with the commentary of Sabara swami, Vidya Vilasa Press, Benares. Hacker, P. (1978) [1965], Dharma im Hindismus, in Id., Kleine Schriften, edited by L. Schmithausen, Wiesbaden, pp. 496-509. Halbfass, W. (1988), Dharma in the Self-Understanding of Traditional Hinduism in Id., India and Europe. An Essay in Understanding, State University of New York Press, Albany, pp. 310-333. Jha, G. (1916), The Pürva Mœmå∫så Sütras of Jaimini. Translated with an original commentary, Allahabad. Kahrs, E. (1998), Indian semantic analysis. The ‘nirvacana’ tradition, Cambridge University Press, Cambridge. Keay, F.E. (1918), Ancient Indian Education. An inquiry into its origin, development, and ideals, Oxford University Press, Oxford. Kripacharyulu, M. (1986), Sayana and Madhava-Vidyaranya. A Study of their Lives and Letters, Guntur. Krishnamacharya, P. (ed.) (1947), Descriptive Catalogue of Sanskrit Manuscripts in the Adyar Library, Adyar Library, Madras, vol. 6. Kulke, H. (1985), Mahåråjas, Mahants and Historians. Reflections on the historiography of Early Vijayanagara and Sringeri, in Dallapiccola, A.L. (ed.), Vijayanagara-City and Empire, Stuttgart. ______ (1997), The State in India 1000-1700, Oxford University Press, Delhi. ______ (2001) [1993], Kings and Cults. State Formation and Legitimation in India and Southeast Asia, Manohar, Delhi. Kurata, H. (1994), Mådhava and Mœmå∫så, in Dwivedi, R.C. (ed.), Studies in Mœmå∫så, Motilal Banarsidass, Delhi, pp. 341-369. Malamoud, C. (1977), Le svådhyåya. Récitation personelle du veda. Taittirœya-Årañyaka Livre II., Institut de Civilisation Indienne, Paris. Minkowski, C. (2002), Nœlakañ™ha Caturdhara and the Genre of Mantrarahasya-Prakå†ikå, in «Journal of American Oriental Society», vol. 122, n. 2, pp. 213-223. Modak, S. (1995), Såyaña, Sahitya Academy, Delhi. Mookerji, R.K. (1969) [1947], Ancient Indian Education, Motilal Banarsidass, Delhi. Mller, M. (ed.) (1983) [1890], R¢gvedasa∫hitå ‡rœsåyañåcåryaviracitamådhavœyavedårtha-praka†asahitå, Varanasi. Narasimhachar, R.B. (1916), Madhavacharya and his younger brothers, in «The Indian Antiquary», vol. 14, pp. 1-24. Oberhammer, G. (2004), Materialen zur Geschichte der Råmånujaschule VII. Zur spirituallen Praxis des Zufluchtnehmens bei Gott (†arañågatiΔ) vor Vienka™anå™ha, Verlag der ÖAW, Wien.

III. How to Produce, Construct and Legitimate a Tradition

357

Oertel, H. (1930), Zur Indischen Apologetik, Stuttgart. Oliver, C.F. (1979), Some aspects of literacy in Ancient India, in «The Quarterly Newsletter of the Laboratory of Comparative Human Cognition», n. 1, pp. 57-62. Patton, L. (1994), Poets and fishes, in Patton, L. (ed.), Authority, Anxiety and Canon, State University of New York Press, Albany. Peterson, P. (1974) [1890], Såyaña’s Preface to the R¢gvedabhåßya, Bhandarkar Oriental Research Institute, Poona. Pollock, S. (2003), Sanskrit Literary Culture from the Inside Out, in S. Pollock (ed.), Literary Cultures in History. Reconstructions from South Asia, University of California Press, Berkeley. ______ (2004), Literary Culture and Manuscript Culture in Precolonial India, unpublished version. Ramachandrudu, P. (1994), Kumårila Bha™™a on the prayojana of vyåkaraña, in Dwivedi, R.C. (ed.), Studies in Mœmå∫så , Motilal Banarsidass, Delhi, pp. 243-255. Ray, K. (1961), Sayana’s Introduction to his Vedic Bhashya, Calcutta. Sankaranarayanan, S. (1993), Sanskrit education in ancient Tamil country, in «The Adyar Library Bulletin», vol. 57. Sarup, L. (1967), The Nighañ™u and the Nirukta, The Oldest Indian Treatise on Etymology, Philology, and Semantics, Motilal Banarsidas, Delhi. Sawai, Y. (1992), The Faith of Ascetics and Lay Smårtas. A Study of the ‡ankaran Tradition of ‡r¢õgeri, Wien. Scharfe, H. (2002), Education in Ancient India, E.J. Brill, Leiden. Seyfort Reugg, D. (1994), Pramånabhüta, *pramåña(bhüta)-purußa, pratyakßadharman and sakßatkr¢tadharman as epithets of the r¢ßi, åcårya and tathågata in grammatical, epistemological and Madhyamaka Texts, in «Bulletin of the School of Oriental and African Studies», n. 57, pp. 303-320. Staal, F. (1979), The Meaninglessness of the Ritual, in «Numen», vol. 26. ______ (1996), Ritual and Mantras: Rules Without Meaning, Delhi. Stein, B. (1989), The Cambridge History of India. Vijayanagara, Cambridge. Stietencron von, H. (2001), Charisma and Canon. The Dynamics of Legitimization and Innovation in Indian Religions, in Dalmia,V., Malinar, A., Christof, M., Charisma and Canon, Oxford University Press. Taber, J. (1989), Are Mantras Speech Acts? Mœmå∫så Point of View, in Alper, H.P., The Mantra, State University of New York Press, Albany. Wagoner, P.B. (1996), Sultan among Hindu Kings: Dress, and the Islamicization of Hindu Culture at Vijayanagara, in «The Journal of Asian Studies», vol. 55, n. 4, pp. 851-880. Wezler, A. (2001), Some remarks on Nirukta 1.20 såkßåtkr¢tadharmåña r¢ßayo, in Michaels, A. (ed.), The Pandit. Traditional Scholarship in India, Manohar, Delhi.

IV. Experiencing Boundaries within Tradition: The Case of the Sankrit Grammarians

MADHAV M. DESHPANDE Ultimate Source of Validation for the Sanskrit Grammatical Tradition: Elite Usage versus Rules of Grammar In his forthcoming article Where do lakßañaika-cakßußka and lakßyaika-cakßußka apply?,1 Ashok Aklujkar takes up a very important question, namely a proper interpretation of the terms lakßañaika-cakßußka and lakßyaika-cakßußka as they appear in the works of Någe†abha™™a. In the course of arriving at his own understanding of these terms, Aklujkar reviews the previous work of Thieme, S.D. Joshi, and Deshpande, and in a richly documented argument, develops an understanding that differs from all the previous authors, but most markedly from that offered by Deshpande. At the outset, the general issue may be presented briefly in the following way. In my previous work (see the bibliography of this article), I have argued on many different occasions that by the time of Bhartr¢hari, a r¢ßification and mythologization of the founding fathers of Sanskrit grammar, namely Påñini, Kåtyåyana, and Pata∞jali had taken place. I have argued that Bhartr¢hari especially highlights Pata∞jali as the ådi†iß™a ‘first, foremost among the †iß™as’ and that the notion of †iß™a itself seems to have shifted from an idealized contemporary elite Brahmin community of Åryåvartta, as seen in Pata∞jali’s Mahåbhåßya, to an identification in Bhartr¢hari of †iß™as with r¢ßis with extraordinary powers of perception, who have a direct insight into dharma (pratyakßadharmåñaΔ), and who are superior to the contemporaries of Bhartr¢hari (asmadvi†iß™a). This change reflected in Bhartr¢hari, in my opinion, allowed grammarians like Kaiya™a to formulate the famous doctrine of the hierarchical authority of the first three sages of Sanskrit grammar, culminating in the unsurpassed prominence of Pata∞jali 1 Professor Ashok Aklujkar informs me that this article is included in the Kelkar Festschrift (Aklujkar forthcoming). I am thankful to him for sending me a pre-publication copy of his article. Even as I may need to disagree with some of his opinions from time to time, I admire his contributions to our understanding of the history of Sanskrit and Sanskrit grammar.

362

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

over his predecessors (uttarottaram munœnåm pråmåñyam). The logical conclusion of this development was that a view developed among Sanskrit grammarians that grammarians up to Pata∞jali had direct access to grammar-independent Sanskrit usage and therefore had the authority to modify the rules of grammar to make grammar better reflect the usage. On the other hand, in this view, grammarians who came after Pata∞jali could focus only on the rules formulated by the founding grammarians, and could not modify the rules. Their sole access to usage, according to this view, was through grammar, and that they did not have independent access to living usage, and hence no authority to modify grammar to account for such grammar-independent usage. As I have argued, this feeling is crystallized in Någe†a’s terms lakßañaika-cakßußka and lakßyaika-cakßußka, which in my understanding refer to ‘those whose eyes are solely fixed on rules of grammar’ and ‘those whose eyes are solely fixed on the usage’. Referring to my views, as well as to the views of Thieme and S.D. Joshi,2 Aklujkar begins his discussion with the following statement: In the discussion of some of these questions and in the discussion of how the acceptability of a word form was determined in the Sanskrit tradition, two textual pieces have gained some prominence. They are Pata∞jali’s discussion under Påñini 6.3.109 (pr¢ßodarådœni yathopadiß™am) and Någe†abha™™a’s terms lakßañaika-cakßußka and lakßyaika-cakßußka. The purpose of this article is to argue that, contrary to the assumption made by at least three excellent specialists of Påñinian grammar in our times, the latter textual piece, consisting of two terms, has no bearing on the issue of determining change in the condition of Sanskrit, on the issue of how the Sanskrit grammarians perceived the history of Sanskrit and, consequently, also on the issue of how they determined if certain words etc. were permissible usages. (Aklujkar forthcoming: §1.4)

I agree with Aklujkar on the first of the three goals of his article. An interpretation of the terms lakßañaika-cakßußka and lakßyaikacakßußka has no direct bearing on determining objective change in the condition of Sanskrit. An objective assessment of change or continuity in Sanskrit can and must be made on the basis of an examination of objective evidence of Sanskrit usage itself. On the remaining two goals of Aklujkar’s article, however, I respectfully beg to differ. Contrary to his assertion, I firmly believe that these two terms do indeed have a central relevance in judging how the Sanskrit grammarians perceived the history of Sanskrit, and especially how they perceived their own role with regard to how to validate or invalidate Sanskrit usage as they knew in their contemporary world. In what follows, I would like to examine the textual evidence in the works of the late grammarians regarding the usage of and understanding of the terms lakßañaika-cakßußka and lakßyaika-cakßußka, keeping in view 2 For the bibliographical details of the works of Thieme and S.D. Joshi cited and examined by Aklujkar, please consult the bibliography of his forthcoming article.

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 363

some of the issues raised by Aklujkar. I would like to thank him for his detailed textual analysis of certain passages, though my own understanding substantially differs from his. Let us start with my understanding of these terms, namely that lakßañaika-cakßußka refers to someone whose eyes are solely fixed on the rules of grammar and lakßyaika-cakßußka refers to someone whose eyes are solely fixed on the usage. Do these terms have a historical reference? None of the authors reviewed by Aklujkar, namely Thieme, Joshi, and Deshpande, have asserted that Sanskrit grammarians were historical linguists in a modern sense. However, do they have a general sense of chronology and change? Consider Pata∞jali’s words: puråkalpe etad åsœt [...] tad adyatve na tathå (Mahåbhåßya [Kielhorn], vol. 1, p. 5). Pata∞jali is saying that in the previous age Brahmins used to study grammar after their thread-ceremony (upanayanasa∫skåra). After they had already studied the concepts of points of articulation, manner of articulation, and types of phonation, they were then taught the Vedic texts. Such is not the case in the contemporary (adyatve) world any longer. One may disagree with Pata∞jali’s understanding of how things used to be at a given point in time, and yet it is clear that he has a certain sense of history. Coming down to Bhartr¢hari, Kaiya™a, and Någe†a, we need to accept that each of these grammarians had his own sense of the history of his tradition and his own place in relation to that perceived history. Kaiya™a is clearly aware that the three founding sages of Sanskrit grammar are from three successive ages, and this sense is embedded in his expression uttarottaram. Bha™™oji Dœkßita not only cites Kaiya™a’s principle in his own works dozens of times,3 he repeatedly says that Påñini does not have Kåtyåyana’s Vårttikas in front of him while he was composing his own sütras.4 Bha™™oji Dœkßita demonstrates an elaborate understanding of the history of the grammatical tradition and develops complex ways to deal with resolving myriad conflicting interpretations found across the entire span that extends from Påñini and his predecessors on one hand and Bha™™oji’s contemporaries on the other. Någe†a, following in the footsteps of Bha™™oji Dœkßita and Haridœkßita, indeed has his own perception of history of the tradition and of his own place within that tradition. This perception of history is not free from what would appear to be mythical elements in our eyes, and yet these are defining perceptions for him. He knows that the three sages of Sanskrit grammar were munis, and not mere åcåryas. He knows the story of Påñini receiving his grammar, and 3 yathottaram munœnåm pråmåñyåt (‡abdakaustubha, vol. 2, p. 159); yathottaram munœnåm pråmåñyam iti sthånivadbhåva evocitaΔ (Prau∂hamanoramå, vol. 1, p. 211); yathottaram iti (Prau∂hamanoramå, vol. 1, p. 139); yathottaram pråmåñyam (Prau∂Ûhamanoramå, vol. 3, p. 646). 4 na tåvad vårttika∫ dr¢ß™vå sütrakr¢taΔ pravr¢ttiΔ (‡abdakaustubha, vol. 1, p. 39); vårttika∫ dr¢ß™vå sütrasyåpravr¢tteΔ (‡abdakaustubha, vol. 2, p. 235); vårttika∫ dr¢ßtvå sütrakr¢to ’pravr¢tteΔ (‡abdakaustubha, p. 430; Prau∂hamanoramå, vol. 3, p. 719).

364

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

especially the akßarasamåmnåya, from ‡iva. He also knows the story of Pata∞jali being an incarnation of ‡eßa. Introductory verses to many of his works refer to Pata∞jali in his appearance as a serpent (see Någe†abha™™o Någe†abhåßitårthavicakßañaΔ, verse 2ab, at the beginning of Uddyota [Mahåbhåßya, Motilal Banarsidass, vol. 1, sec. 1, p. 1]). This is a post-Bhartr¢hari element in the history of the Påñinian tradition. He is indeed aware of Kaiya™a’s maxim: ‘authority of the three sages is successive, namely the later the sage, the greater his authority’ (yathottara∫ hi munitrayasya pråmåñyam). While commenting on this maxim, Någe†abha™™a says that the later sage is likely to have known a wider usage of language (uttarottarasya bahulakßyadar†itvåt) (Uddyota, [Mahåbhåßya, Motilal Banarsidass, vol. 1, sec. 1, p. 217]; also see Uddyota on P. 3.1.87 [dhinvikr¢ñvyor ...] [Mahåbhåßya, Motilal Banarsidass, vol. 2, p. 100]). It is clear that among the three grammarian-munis Någe†a shows the highest respect for Pata∞jali. At the very beginning of his Pradœpa, Kaiya™a takes up the question of why the Bhåßyakåra (=Pata∞jali) should explain the purposes of studying Sanskrit grammar (bhåßyakåro vivarañakåratvåt vyåkarañasya såkßåt prayojanam åha) (Pradœpa [Mahåbhåßya, Motilal Banarsidass, vol. 1, sec. 1, p. 3]). Commenting on this passage of Pradœpa, Någe†abha™™a refers to all the three grammarians. While he refers to Påñini by his name, and to Kåtyåyana as vårttikakr¢t, he refers to Pata∞jali by the honorific bhagavån (Påñininå [...], vårttikakr¢tå [...], bhagavå∫s tu [...]) (Uddyota [Mahåbhåßya, Motilal Banarsidass, vol. 1, sec. 1, p. 3]).5 All these factors, and especially the divine origin of Påñini’s grammar, are taken up by Någe†a and woven into a powerful argument to explain the status and the authority of the rules of Sanskrit grammar. Någe†abha™™a, more than most other grammarians, uses the reason of ‘being contrary to the Bhåßya’ to reject the views of Kaiya™a and others (e.g. yat tu [...] iti abhyastasa∫j∞åsütre Kaiya™aΔ, tan na, na samprasårañe iti sütrastha-bhåßyavirodhåt) (Laghu†abdendu†ekhara (A), p. 48). This shows that there is a limit to the doctrine of uttarottara∫ munœnå∫ pråmåñyam for Någe†a. After Pata∞jali, no other later grammarian is r¢ßified. The doctrine clearly stops with Pata∞jali. Respected predecessors like Bhartr¢hari, Kaiya™a, Helåråja, Haradatta etc. are not given the same status as Pata∞jali, and if their views are contrary to the statements of the Bhåßya, they are rejected. This indicates that for Någe†a, grammarians after Pata∞jali do not have the same authority. The age of the r¢ßis of grammar ends with the big bang of Pata∞jali. For Någe†a, the age of Pata∞jali and his predecessors is markedly different from the age of Pata∞jali’s successors. This is his sense of history, and his statements need to be understood in reference to this sense of history. 5 The term bhagavån refers to Pata∞jali even in the works of Vaidyanåtha Påyaguñ∂e, who is a disciple of Någe†a. See aster bhüΔ ity atra sandihånena bhagavataiva tadabhyupagamåt (Bhåvaprakå†a on ‡abdaratna, p. 19).

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 365

The change in perception is even more vivid if we compare Någe†abha™™a with Kaiya™a. On the Mahåbhåßya on the ‡ivasütra (a-iu-˜), Kaiya™a says: ‘is it the case that the author of the sütra himself used an open (vowel a), and the author of the Vårttikas is merely explaining the purpose, or is it that the author of the sütras did not use an open (vowel a), but the author of the Vårttikas is advocating such a use?’ (ki∫ sütrakåreñaiva vivr¢topade†aΔ kr¢to vårttikakåreña tu tasya prayojanam uktam, athavå akr¢ta eva vivr¢topade†o vårttikakr¢tå kartavya-tvenopanyasta iti pra†naΔ) (Pradœpa [Mahåbhåßya, Motilal Banarsidass, vol. 1, sec. 1, p. 64]). It is clear that Kaiya™a believes that the author of a-i-u-˜ is the Sütrakåra, i.e. Påñini. On the other hand, Någe†abha™™a’s Uddyota identifies this Sütrakåra unquestionably with Mahe†vara (sütrakåraΔ - mahe†varaΔ vedapurußo vå, yenåkßarasamåmnåyam ity ådy aitihyåd ity åhuΔ) (Uddyota [Mahåbhåßya, Motilal Banarsidass, vol. 1, sec. 1, p. 64]; also åcåryeña, Pradœpa, but contrast †ivo vedapurußo våtråcåryaΔ, Uddyota, [Mahåbhåßya, Motilal Banarsidass, vol. 1, sec. 1, p. 74]). It may also be pointed out that what we may call a story or myth today, Någe†abha™™a refers to by the word aitihya ‘historical account’. Clearly, Någe†a is no modern historian, but we cannot deny that he has his own perception of history. We can indeed attempt to comprehend Någe†a’s perception of history, and place that perception of history within a history of changing perceptions of history. Consider the following statement from Någe†a: On P. 1.4.8 (patiΔ samåsa eva). The word eva ‘only’ is meant to restrict as desirable [certain declensions of the word pati to its occurrence in the compounds]. Forms like pataye and sakhinå occurring in Smr¢tis and Puråñas are incorrect, since they cannot be derived by the rules of grammar. Like officiating for unworthy individuals like Tri†aõku, the usage of incorrect words by the r¢ßis does not generate any fault, due to the eminence of their penance.6 They incur a fault only if they use incorrect expressions during a sacrificial performance. By the power of the religious prescription for the study of Smr¢tis and Puråñas, and due to their [authors’] power of penance, even for us the use of citations from such texts even in ritual performance does not lead to a fault. Folks like us, using such forms independently do incur a fault. Therefore in the Bhåßya on P. 1.4.3 (yüstryåkhyau nadœ), having cited the saying ‘Poets compose [freely] like Vedic texts’, it is said that this is not a statement of desired usage (patiΔ samå [...] Ù evakåra iß™ato ’vadhårañårthaΔ Ù smr¢tipuråñeßu pataye sakhinetyådi vyåkarañånißpanna-tvenåsådhv eva Ù tri†aõkvådyayåjyayåjanådivat tapomåhåtmyenåsådhuprayoge ’pi r¢ßœñå∫ doßåbhåvåt Ù yåj∞e karmañy evåsådhuprayoge doßåc ca Ù asmåkam api smr¢tipuråñådhyayanavidhibalåt teßå∫ tapobalåc ca yaj∞amadhye ’pi tadantar gatatatpå™he doßåbhå6 Någe†a here echoes the statements of the Åpastambadharmasütra, 2.6.13.8-10 (dr¢ß™o dharmavyatikramaΔ såhasa∫ ca pürveßåm Ù teßå∫ tejovi†eßeña pratyavåyo na vidyate Ù tad anvœkßya prayu∞jånaΔ sœdaty avaraΔ Ù). Notice the relative chronology expressed with the expressions pürveßåm and avaraΔ. The folks referred to as pürva belong to the golden age of r¢ßis, while the avara belongs to the latter-day Kaliyuga-like period. Also compare Åpastambadharmasütra, 1.1.5.4 (tasmåd r¢ßayo ’vareßu na jåyante niyamåtikramåt Ù).

366

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

vaΔ Ù svåtantryeñedr¢†a∫ prayu∞jånå asmadådayaΔ pratyavayanty eva Ù ata eva nadœsa∞j∞åsütre bhåßye chandovat kavayaΔ kurvantœty uktvå na hy eßeß™ir astœty uktam Ù). (Laghu†abdendu†ekhara (A), p. 241)

This passage from Någe†a clearly shows awareness of relative chronology of different speakers of Sanskrit and their special status in relation to Sanskrit grammar.7 A certain amount of understanding of relative chronology did exist among the traditional grammarians in India, though various domains of this chronology may be mythological from a modern point of view. However, on occasion, the traditional grammarians come up with a startlingly clear statement of interpretive ideology. Consider the following comment from Någe†a:8 By this [previous discussion] it is suggested that [Pata∞jali’s] view rejecting [the necessity of] the sütra is stronger than the view expressed by the sütra [itself], [and hence] as far as possible, somehow one should try to derive the example [as] intended by the upholder of the rejection of the sütra even from the point of view of the sütra, otherwise, the sütra [rejected by Pata∞jali] is certainly [to be considered as being] invalidated, and that such a rejection of the sütra may indeed be acceptable to [the author of] the sütra. Like the notion of invalidity of the statements of ancient sages rejected by the author of the sütras [=Påñini], it is appropriate to accept a notion of invalidity of those portions of Påñini’s rules in view of the statements from the Vårttikas and the Bhåßya that aim at rejecting the words of the author of the sütras [=Påñini] (anena sütramatåt pratyåkhyånavådimatam prabalam iti pratyåkhyånavådisammatalakßyam eva katha∫cit sütramate ’pi sådhya∫ na tu viparœta∫ sati sambhave anyathå sütram apramåñam eveti ca pratyåkhyåna∫ ca sütrasammatam iti dhvanitam Ù sütrakr¢tkhañ∂itapråcœnar¢ßivåkyånåm apråmåñyavat sütrakåravacaΔ khañ∂anapravr¢ttabhåßyavårttikavacanais tada∫†e sütrakåravacaso ’py apråmåñyakalpanå evociteti dik). (Laghu†abdaratna, pp. 260-261)

This statement combines a sense of relative chronology with the logical conclusion of Kaiya™a’s doctrine of the successively rising authority of the three founding sages. There is a passage identical with these words above to be found in Någe†a’s Br¢hacchabdendu†ekhara (pt. 1, p. 425). Issues of relative chronology did engage the attention of Någe†a and his predecessors. The Laghu†abdendu†ekhara 7 Also see: kaviprayogåñåm bahu†aΔ pråmådikånå∫ dar†anena teßåm årßavacanasa∫koce na månatvam (Laghu†abdaratna, in Prau∂hamanoramå (B), p. 473). 8 I should note that there is a dispute over the authorship of [Laghu]†abdaratna. Some ascribe it to Haridœkßita and others to Någe†a. The ‡abdaratna is available in different versions, and this whole issue has been discussed by V.L. Joshi (1966: pp. 309ff) who argues that the Laghu†abdaratna of Någe†a occasionally criticizes the ‡abdaratna of Haridœkßita, who is referred to by the term månya by Vaidyanåtha Påyaguñ∂Ûe in his commentary Bhåvaprakå†a. The inclusion of the Br¢hacchabdaratna by Haridœkßita and Laghu†abdaratna in the edition of the Prau∂hamanoramå (B), and the discussion in its introduction by Sitaram Shastri also makes this evident. For the purposes of this paper, I shall consider the Laghu†abdaratna as being authored by Någe†abha™™a, even though the editions I have used often ascribe it to Haridœkßita.

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 367

and the Br¢hacchabdendu†ekhara of Någe†a contain discussions about relative chronology of different sages.9 The pre-modern ideas of these grammarians may not always match ours, but they were indeed guided by their understanding of chronology. In interpreting the terms lakßaña-eka-cakßußka and lakßya-ekacakßußka, the problem may be narrowed down to how to understand the value of the expression -eka- in these compounds. To be fair to Aklujkar, he does not deny that literally these terms are likely to mean ‘those whose eyes are focused exclusively on the rules of grammar’ and ‘those whose eyes are focused exclusively on the usage of language’. However, Aklujkar objects to such a literal interpretation by pointing out that if lakßyaika-cakßußka is applied to Pata∞jali, the assertion would be that Pata∞jali does not direct his eye to anything other than usage, and if lakßañaika-cakßußka is applied to post-Pata∞jali Påñinœyas, it would follow that they do not view anything other than the (Aß™ådhyåyœ etc.) rule. Such clearly is not the case in the Mahåbhåßya or the works of Pata∞jali’s successors. Thus, if one is to apply the terms in the proposed way, one must take them in not too literal a way and understand the references made to lakßya and lakßaña as containing a ‘pinch of salt’. (Aklujkar forthcoming: §2.6)

What Aklujkar is referring to, I suppose, is the implications as he can imagine of a literalist rendering of these terms. However, I am not aware that any of the scholars Aklujkar examines have at any time asserted that Pata∞jali knew only the usage, and not the rules, or that post-Pata∞jali grammarians knew only the rules, and not the usage. At least on my own part, I have clarified what these terms suggest to me: The term lakßañaika-cakßußka refers to those grammarians who have no independent access to the correct usage of Sanskrit and whose sole attention is fixed on the inherited rules of Sanskrit grammar. Thus, while for the lakßyaika-cakßußka grammarians, the correct usage of Sanskrit, independently known, is the final authority, the reverse is the case for the lakßañaika-cakßußka grammarians. For them, the authority of the inherited rules is the most decisive principle of authority. It is within this lakßañaika-cakßußka phase of Sanskrit grammar, that one seeks the authority of the ancient grammarian-Munis, rather than of any true contemporary ‡iß™as. (Deshpande 1998: 24-25)

To my knowledge, no one has claimed that these terms mean that one set of grammarians knew only grammar or that the other knew 9 yåj∞avalkyådayo hi påñinyapekßayå ådhunikå iti prasiddhir iti haradattaΔ Ù bhåßye tu eßåm api †å™yåyanåditulyakålatvåt [...] Ù (Br¢hacchabdendu†ekhara, pt. 2, p. 1360); etad bhåßyapråmåñyåt vårttikånåm api sütrasammatatvam åhuΔ Ù tad uktam abhiyuktaiΔ - sütreßv eva hi tat sarva∫ yad vr¢ttau yac ca vårttike iti Ù [...] na hi påñineΔ sütrakarañakåle kåtyåyanavacaso bhåva iti kaiya™asyåpi tadvacaso niyamena buddhau bhåva ity arthaΔ Ù [...] påñini† ca tajj∞åne ’py anyabhyå∫ tasya våcyatvåt svaya∫ vårttikabhåßyoktårthån noktavån iti såmpradåyikåΔ Ù (Laghu†abdendu†ekhara (B), pt. 1, p. 494).

368

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

only usage. The terms are understood in the context of how one locates authoritative value. What is suggested is that the individuals described as being lakßyaika-cakßußka place the final authority in the usage that they know independently of the inherited grammatical system, and are for that reason deemed to be entitled to adjust the rules of grammar to fit to that usage. The second term, lakßañaika-cakßußka, is taken to refer to those individuals who place the final authority in the inherited rules of grammar, and not in the known usage of the language, which it is felt is no longer independent of the received rules of grammar.10 Having presented a prima-facie literalist interpretation of these terms and having pointed out problems in the presumed implications of such literalist interpretations, Aklujkar concludes that the literalist interpretations are therefore not acceptable, and he then offers another tentative interpretation that moves away from the normal sense of eka as ‘only, solely, exclusively’ to ‘almost, predominantly’: But, can we justifiably go beyond the ‘almost’ or ‘predominantly’ adjustment? And if we cannot, would a rendering of lakßañaika-cakßußka as ‘almost always or predominantly concerned with the Aß™ådhyåyœ rules’ and of lakßyaika-cakßußka as ‘almost always or predominantly concerned with the usage to be derived’ be true to what we find in Pata∞jali’s work and what we find in the works of the later Påñinœyas, respectively? Even if we were to replace ‘concerned with’ by ‘dependent on’ in the preceding renderings, or allow us to infer in a logically unassailable way that Pata∞jali was thought of as having access to a means (in the form of usage verification) which was not at all to be predicated in the case of the later Påñinœyas? (Aklujkar forthcoming: §2.6)

This suggests that Aklujkar is proposing to find a looser way to interpret the -eka- in these two compounds, a way that would preserve this new sense of eka and yet fit the history of Sanskrit and Sanskrit grammar. Thus, both the types of grammarians would know both usage as well as grammar. It is just that one set would predominantly or mostly or almost always depend on the usage, while the other set would predominantly or mostly or almost always depend on the rules of grammar. In this provisional view, the difference between these two types is not categorical, but only that of proportion, and neither is particularly privileged over the other. The above modified understanding offered in Aklujkar’s article seems only provisional, and ultimately not satisfactory to him, and, after a detailed analysis of five passages from Någe†a’s works, Aklujkar concludes as follows: 10 In this respect, one may note that Bha™™oji and his successors are rather conservative, as compared to other contemporary Påñinians like Nåråyañabha™™a, Apåñinœyapramåñatå, verse 2: påñinyuktam pramåña∫, na tu punar apara∫ candrabhojådisütram Ù ke ’py åhus tallaghiß™ha∫, na khalu bahuvidåm asti nirmülavåkyam ÙÙ bahvaõgœkårabhedo bhavati guñava†åt, påñineΔ pråk katha∫ vå Ù pürvokta∫ påñini† cåpy anuvadati, virodhe ’pi kalpyo vikalpaΔ ÙÙ

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 369

Någe†a should be understood as speaking of two Aß™ådhyåyœ user personae. One persona allows the rules to derive whatever they can. He first tests how far the automation feature of the sütras will take him, without worrying at the same time about whether the forms generated are acceptable as means of standard linguistic communication. The other user persona represents those users of the Aß™ådhyåyœ who are primarily interested in checking how the rule mechanism will derive the forms which already exist in the object language as far as they know. To state the same point differently, one persona tests the rules for possible over-application, while the other tests them for adequacy, and that, in this sense, their thoughts are completely given to the rules and to the forms-to-be-derived, respectively, in the beginning stages of their effort. (Aklujkar forthcoming: §2.10)

Such an achronological or synchronic interpretation of the distinction between lakßañaika-cakßußka and lakßyaika-cakßußka as if these were two competing sections within a contemporary meeting of grammarians, as I hope to demonstrate in the present article, is in direct contradiction with the textual evidence from the works of Någe†a, his predecessors and successors. With this framework, Aklujkar then examines five occurrences of these terms in Någe†a’s works in some details. While doing this, he makes the following statements: a) “In all, Någe†a seems to have used the terms only five times” (§ 2.7). b) “As far as my study of secondary literature on the Påñinœyas goes, no pre-Någe†a occurrences of lakßañaika-cakßußka and lakßyaika-cakßußka have been noticed. This indicates that the terms could be important in judging the extent of Någe†a’s original contribution to the Påñinian tradition” (§ 2.7). As I hope to demonstrate below, on both of these counts, Aklujkar’s statements need to be adjusted in view of the available wider textual evidence. The terms occur more commonly in Någe†a’s works than ‘only five times’. Secondly, the terms occur in the Ratnaprakå†a commentary of ‡ivaråmendrasarasvatœ on Kaiya™a’s Mahåbhåßyapradœpa. ‡ivaråmendrasarasvatœ lived between Bha™™ojœ Dœkßita and Någe†abha™™a. ‡ivaråmendrasarasvatœ criticizes Bha™™ojœ’s ‡abdakaustubha, while his views are in turn criticized by Någe†a.11 This takes the usage of these terms to a period at least a generation before Någe†a. Since Någe†a is known to criticize ‡ivaråmendrasarasvatœ’s views, we can trace Någe†a’s usage of these terms to his acquaintance with the works of his predecessors like ‡ivaråmendrasarasvatœ. Thirdly, these terms are used by Någe†a’s teacher Haridœkßita in his commentary Br¢hacchabdaratna on Bha™™oji Dœkßita’s Prau∂ha11 For a discussion of the relationship of ‡ivaråmendrasarasvatœ to Bha™™oji Dœkßita and Någe†abha™™a, see Upodghåta, Mahåbhåßya-Pradœpa-Vyåkhyånåni, vol. 1, pp. xviii-xix.

370

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

manoramå. This provides us a direct connection between Någe†a and his teacher Haridœkßita. Fourthly, the best way for us to understand what Någe†a meant by these terms is to look at the commentaries on Någe†a’s work by his immediate disciple Vaidyanåtha Påyaguñ∂e. I should note here that Påyaguñ∂e not only explains these terms where they appear in Någe†a’s works, he independently uses these terms elsewhere in his own explanations. Thus the terms originated at least a generation before Någe†a and continued to remain influential for at least a generation after Någe†a. In order to get a fuller sense of how these terms were used and what the authors intended, we need to begin with ‡ivaråmendrasarasvatœ. In what follows, I shall first examine five passages from ‡ivaråmendrasarasvatœ’s Ratnaprakå†a, a commentary on Kaiya™a’s Mahåbhåßyapradœpa. Ratna 1: ‘Since, for us who are lakßañaika-cakßußka, it is impossible to determine the absence of the usage of that form [...]’ (lakßañaikacakßußkåñåm asmåka∫ tatprayogåbhåvanirñayåyogena [...]) (Ratnaprakå†a, vol. 2, p. 43). Ratna 2: ‘Moreover, this †åstra is useful for the lakßañaikacakßußkas, since they are unable to determine the meaning of a rule of grammar by consulting the usage’ (ki∫ ca lakßañaikacakßußkåñåm upayogœda∫ †åstram Ù tathå ca te prayoga∫ dr¢ß™vå sütrårtha∫ nirñetum asamarthåΔ) (Ratnaprakå†a, vol. 6, p. 18). Ratna 3: ‘In reality, such an argument [rejecting a certain substitute as prescribed in Påñini’s rule] is merely a demonstration of one’s false scholarly confidence. It is inappropriate to reject [the prescription of] the substitute, a prescription that is designed for a lakßañaika-cakßußka person, by assuming the point of view of a lakßyaika-cakßußka person. Otherwise, [from the point of view of a lakßyaika-cakßußka person] one may end up rejecting the entire †åstra of grammar, and that is undesirable’ (vastutas tv idam prau∂himåtra∫, lakßañaikacakßußkårtha∫ kr¢tasyåde†asya lakßyaikacakßußkatå†rayeña pratyåkhyånånaucityåt Ù anyathå samastavyåkaraña†åstrasyaiva pratyåkhyeyatåpatteΔ Ù) (Ratnaprakå†a, vol. 5, p. 286). Ratna 4: ‘Since it is possible for the r¢ßis to determine the [significance of the] marker-sounds [attached to grammatical elements] simply by relying on the usage [of the language] without any expectation of an exegetical explanation, and since it is appropriate for people like us to make such a decision solely based on a [received] exegetical explanation’ (r¢ßœñå∫ vyåkhyånanairapekßyeña prayogeñaivånubandha nirñayasambhavenåsmadådœnå∫ kevalavyåkhyånenaiva tannirñayau cityena ca) (Ratnaprakå†a, vol. 4, p. 165). Ratna 5: ‘It means to say that for the lakßañaika-cakßußkas like us, it is not possible to have the knowledge of proximity even in terms

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 371

of meaning. [Kaiya™a] offers an explanation from the point of view of a lakßyaika-cakßußka just by assuming scholarly bravado’ (lakßañaikacakßußkåñåm asmadådœnåm arthato ’py åntarya j∞ånåsambhavåt iti bhåvaΔ Ù prau∂him å†ritya lakßyaikacakßußka para∫ vyåkhyånam åha Ù) (Ratnaprakå†a, vol. 4, p. 168). If we cumulatively look at these five passages, several features of the classes designated by the terms lakßañaika-cakßußka and lakßyaikacakßußka become transparent: a) The lakßañaika-cakßußka group is often clarified by adding terms like asmadådœnå∫ ‘people like us’. b) The lakßañaika-cakßußka folks are unable to decide the meaning of a rule of grammar by looking at the usage (te prayoga∫ dr¢ß™vå sütrårtha∫ nirñetum asamarthåΔ), and it is appropriate for folks like us to make decisions solely based on received exegetical explanations (asmadådœnåm kevalavyåkhyånenaiva tannirñayaucityena). c) Without explicitly using the term lakßyaika-cakßußka, passage 4 cited above says that the r¢ßis do not need exegetical explanations (vyåkhyånanairapekßyeña) and they can make decisions solely based on the usage of language (prayogeñaiva [...] nirñayasambhavena). This can only be a reference to the three founding munis of Sanskrit grammar. d) If one fully resorts to the point of view of a lakßyaika-cakßußka, the entire grammatical †åstra can be rejected as being unnecessary. A true lakßyaika-cakßußka does not need any grammar. He already has full access to the usage. e) In commentarial literature, if someone argues from the point of view of a lakßyaika-cakßußka, it only reflects prau∂hi ‘assumed stance, pride in oneself’, and such an assumption of the lakßyaika-cakßußka point of view is counter-productive for a commentator. The lakßyaika-cakßußka point of view is appropriate only for a r¢ßi like Pata∞jali.12 Having examined the usage of the terms lakßyaika-cakßußka and lakßañaika-cakßußka in ‡ivaråmendrasarasvatœ’s Ratnaprakå†a, let us note the usage of these terms by Haridœkßita in his commentary Br¢hacchabdaratna on Bha™™oji Dœkßita’s Prau∂hamanoramå. Haridœkßita, 12 Bhairavami†ra in his commentary Bhairavœ on the ‡abdaratna (p. 40) makes an interesting comment on a passage where an explanation in the Bhåßya is called prau∂hivåda. He says that the normal meaning of the term prau∂hivåda is that a statement is made with great assertiveness and yet it is inconsistent in reality. However, such a meaning of the term is not applicable to a statement of Pata∞jali. It is inappropriate to label a statement in the Bhåßya as being inconsistent (prau∂hivådam ity asya mahattvenokta∫ vastuto ’saõgatam ity arthaΔ pratœyate, sa tu na yuktaΔ bhåßyakåroktåv asaõgatatvakathanasyåyuktatvåt Ù).

372

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

the grandson of Bha™™oji, was a near contemporary of ‡ivaråmendrasarasvatœ and most importantly he was the teacher of Någe†abha™™a. This work has now been clearly established as an independent work by Haridœkßita, and as being distinct from the Laghu†abdaratna of Någe†abha™™a, which is ascribed in many older editions to Haridœkßita himself. In order to examine the views of Haridœkßita, we shall exclusively examine passages from the Br¢hacchabdaratna: Hari 1: ‘The linguistic usage of people like Påñini is, however, not dependent upon grammar, because they are lakßyaika-cakßußka’ (påñiniprabhr¢tœnå∫ †abdaprayogas tu na †åstrasåpekßaΔ, lakßyaikacakßußkatvåt) (Br¢hacchabdaratna, p. 5). They have an independent access to the living usage of the language. Hari 2: ‘Referring to all of this, it has been said by Bhagavån, the author of the Bhåßya’ (etad eva sakalam abhisandhåya bhagavatå bhåßyakåreñoktam) (Br¢hacchabdaratna, p. 104). Reference to Pata∞jali with this distinctive title is shared by most authors of this period. Hari 3: ‘It has been conclusively established by the author of the Bhåßya himself that it is not possible for people like us to determine the absence of a usage’ (asmadådibhiΔ prayogåbhåvasya nirñetum a†akyatvåd iti bhåßyakårair eva siddhåntitatvåt) (Br¢hacchabdaratna, p. 327). This passage is parallel to the usage of the Ratnaprakå†a where the term lakßañaika-cakßußka is used to refer to such people like us who lack this ability. Hari 4: ‘Even though a rule is split into two in order to achieve a desired result, and even though this is understood to mean that, with different target examples in view, it sometimes applies and sometimes does not, it is not possible for the lakßañaika-cakßußka people like us to make that decision without an indication as to the intention [behind applying or not applying the rule]’ (nanu yogavibhågasyeß™asiddhyarthatayå lakßyabhedena kvacit pravr¢ttiΔ kvacid apravr¢ttir ityarthalåbhe ’pi tayor lakßañaika-cakßußkair asmadådibhis tåtparyagråhaka∫ vinå nirñayo ’†akyaΔ) (Br¢hacchabdaratna, p. 330). Hari 5: ‘The decision [regarding application and non-application of a rule], with desired target examples in mind, cannot be made in the case of a statement coming from a modern ignorant lakßañaika-cakßußka person, the way it can be made in the case of a statement coming from a lakßyaika-cakßußka sage (muni)’ (na hi lakßyaikacakßußkamuniprañœtavåkyasyeva lakßañaika-cakßußkådhunikabålakalpitavåkyasya lakßyånurodhena vyavasthå sådhyå) (Br¢hacchabdaratna, p. 511). The statement clearly distinguishes between the r¢ßi/munis who alone are lakßyaikacakßußka and the latter-day modern grammarians who are ignorant of the living usage of the language, independently of the rules of grammar.

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 373

Hari 6: ‘Thus [this particular view] is rejected, since, as indicated earlier, it does not agree with [the views of] the three munis’ (etena [...] ity apåstam Ù uktarœtyå munitrayåsammatatvåt) (Br¢hacchabdaratna, p. 619). Hari 7: ‘This verse of Bhartr¢hari, however, is in accordance with the opinion expressed in the Bhåßya’ (hari†lokas tu bhåßyoktamatåbhipråyeña) (Br¢hacchabdaratna, p. 660). This points out that Bhartr¢hari is to be considered valid only to the extent his words express opinions that are in agreement with the words of Pata∞jali’s Bhåßya. Hari 8: ‘Like officiating at the sacrifices of unworthy individuals like Tri†aõku, an incorrect usage by the sages who are endowed with great austerities does not generate that much of a fault. On the other hand, either on account of the religious prescription for the study of Smr¢tis and Puråñas [composed by the ancient sages] or on account of their great power of austerity, for people like us there is no fault in using [incorrect] readings in [the recitation or ritual application of] such [texts]. However, folks like us, using such [incorrect forms] independently do incur a fault’ (tri†aõkvådyayåjyayåjanådivat tapomåhåtmya†ålinåm munœnåm asådhuprayogo ’pi nåtœva bådhakaΔ Ù asmadådœn prati tu smr¢tipuråñådhyayanavidhibalåt teßå∫ tapomåhåtmyabalåd eva vå tadantargatatatpå™ho na bådhakaΔ Ù tathå ca svåtantryeñedr¢†a∫ prayu∞jånå asmadådayaΔ pratyavayanty eva) (Br¢hacchabdaratna, p. 387). Hari 9: ‘Here, the Bhåßya, rejecting the sütra [of Påñini] and the Vårttika [of Kåtyåyana] is the [validating] authority’ (sütravårttikapratyåkhyånapara∫ bhåßya∫ cåtra pramåñam) (Br¢hacchabdaratna, p. 682). This affirms the superior authority of Pata∞jali in relation to Påñini and Kåtyåyana. Hari 10: ‘Indeed by the rejection of the sütra [of Påñini, in this case] we achieve the desired result’ (sütrasyaiva pratyåkhyåneneß™åpatteΔ) (Br¢hacchabdaratna, p. 704). The rejection of the sütra is proposed by the Bhåßya, and Haridœkßita asserts that we achieve the desired derivational result by rejecting the sütra, as suggested by Pata∞jali, rather than by keeping it and following the opinion of Påñini. Thus, among the three founding sages of grammar, it is Pata∞jali who stands taller than the other two. It is his opinion that decides what is ‘desired’ for us. We should note that these passages from Haridœkßita’s Br¢hacchabdaratna confirm the general understanding of the terms lakßañaika-cakßußka and lakßyaika-cakßußka that we gathered from passages cited from ‡ivaråmendrasarasvatœ’s Ratnaprakå†a earlier. I should also note that only a small portion of Haridœkßita’s work is available in print, and that these terms may indeed appear in the as-yet

374

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

unavailable portion of his work. While these terms thus appear in the works of these two contemporary writers, they are not found in the works of Bha™™oji Dœkßita, and yet, as I intend to demonstrate below, his works fully support the conceptual framework that is expressed through these terms. Below, I shall examine a sample of relevant passages from Bha™™oji Dœkßita’s ‡abdakaustubha and Prau∂hamanoramå. Bha™™oji 1: ‘Even though folks like us are unable to observe such a large field of linguistic usage [as described by Pata∞jali in his Bhåßya], we should infer that the usage of words derived by the rules of grammar exists’ (etåvantam prayogavißayam paryålocayitum a†aktair apy asmadådibhiΔ lakßañånugatånåm prayogo ’stœty anumeyam) (‡abdakaustubha, vol. 1, p. 24). This statement distinguishes the abilities of munis like Pata∞jali from the limited capacities of folks like us. While they can directly look into the living usage, we must follow the rules of grammar. This distinction is analogous to the conceptions found in the works of ‡ivaråmendrasarasvatœ and Haridœkßita discussed earlier. Bha™™oji 2: ‘The Bhagavån [=Pata∞jali], rejecting [a certain] Vårttika [of Kåtyåyana], says’ (vårttikam pratyåcakßåño bhagavån åha) (‡abdakaustubha, vol. 1, p. 154). Here, Bha™™oji not only acknowledges Pata∞jali’s authority to reject the Vårttikas of Kåtyåyana, the term bhagavån applied to him shows a rare degree of high respect. For the use of bhagavån for Pata∞jali, also see Prau∂hamanoramå, vol. 2, p. 221; vol. 3, p. 647. Bha™™oji 3: ‘Is there anything that an unrestrained person like you, describing things contrary to what is acceptable to the Bhåßya, would not do?’ (bhåßyådisammatåd viparœtam eva varñayan bhavådr¢†o niraõku†aΔ ki∫ ki∫ na kuryåt) (‡abdakaustubha, vol. 1, p. 156). The passage draws a distinction between an ignorant person who speaks contrary to the Bhåßya and the non-ignorant who speaks in accordance with the Bhåßya. Bha™™oji 4: ‘Bhagavån [=Pata∞jali], while rejecting the entire single-remainder operation for words with identical and nonidentical forms’ (sarüpåñåm virüpåñå∫ ca sarvam eka†eßam pratyåcakßåñena bhagavatå) (‡abdakaustubha, vol. 2, p. 41). Bha™™oji acknowledges the right of bhagavån Pata∞jali to reject Påñini’s rules. Bha™™oji 5: ‘A [certain] Vårttika [of Kåtyåyana] has been rejected by bhagavån [Pata∞jali], the author of the Bhåßya’ (iti vårttikam pratyåkhyåtam [...] bhagavatå bhåßyakåreña) (‡abdakaustubha, vol. 2, p. 65). Bha™™oji 6: ‘For this reason, for bhagavån [Pata∞jali], while rejecting [a particular] sütra [of Påñini]’ (tasmåt sütram pratyåcakßåñasya bhagavataΔ) (‡abdakaustubha, vol. 2, p. 122).

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 375

Bha™™oji 7: ‘Without [an express] statement from the three munis, the assumption that [a word is restricted to] Vedic [usage] on the basis of our own desires leads to an over-extension’ (munitrayokti∫ vinå asmadicchayå chåndasatvakalpane ’tiprasaõgåt) (‡abdakaustubha, vol. 1, p. 140). Bha™™oji 8: ‘The division of correct versus incorrect words [is made] these days on the basis of the views of the three munis, because only that [tradition of the three munis] is accepted as being ancillary to the Veda by our contemporary elites’ (munitrayamatena idånœ∫ sådhvasådhuvibhågas tasyaiva idånœ∫tana†iß™air vedåõgatayå parigr¢hœtatvåt) (‡abdakaustubha, vol. 1, p. 165). Bha™™oji 9: ‘What was said in the Kå†ikå and the explanation offered in the Nyåsa, both of these are to be set aside because they are contrary to the Bhåßya’ (yat tu kå†ikåyåm [...] ity uktam, tatraiva nyåse ’pi [...] iti vyåkhyåtam, tad ubhayam bhåßyaviruddhatvåd upekßyam) (‡abdakaustubha, vol. 1, p. 251). The Bhåßya of Pata∞jali, for Bha™™oji Dœkßita, overrides the authority of the Kå†ikåvr¢tti and its commentaries. Bha™™oji 10: ‘What the author of the Nyåsa, Haradatta, and Mådhava have said [in this particular instance], that is nothing but an expression of false confidence. Because of the conflict with the words of the three munis, one should not accept the propriety as imagined by oneself. So it should be understood by the good-hearted [scholars]’ (nyåsakåraharadattamådhavås tu [...] ity åhuΔ, tat tu prau∂hivådamåtram, munivacanavirodhena svotprekßitasvårasyå nådarttavyatvåd iti suhr¢dayair vibhåvyatåm) (‡abdakaustubha, vol. 2, p. 320). Bha™™oji 11: ‘The [particular statement from] Kalpataru, though it may seem possible by the method of the [Kå†ikå]vr¢tti, is merely an expression of false confidence, since it conflicts with the Bhåßya’ (iti kalpatarugrantho vr¢ttikårarœtyå sambhavann api bhåßyaviruddhatvåt prau∂hivådamåtram) (‡abdakaustubha, vol. 2, p. 333). Bha™™oji 12: ‘Since he is opposed to the views of the three munis, he is not authoritative. [Further], your lack of desire is inconsequential’ (tasya ca munivacanavirodhenåpråmåñikatvåt Ù tvadœyå-nicchåmåtrasyåki∞citkaratvåt) (Prau∂hamanoramå, vol. 1, p. 282). While the ancient munis are the bearers of authority, someone who contradicts their views is indeed not to be accepted as authoritative, and the lack of support from such a modern person [like ‘you’] is of no great consequence. Bha™™oji 13: ‘Here, the author of the [Kå†ikå]vr¢tti offered this example. Kaiya™a, Haradatta, and others say that that is invalid because it is not stated by the three munis’ (atra vr¢ttikåraΔ [...] ity udåjahåra Ù tac ca munitrayånuktatvåd apramåñam iti kaiya™aharadattådayaΔ) (Prau∂hamanoramå, vol. 1, p. 69). This

376

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

passage indicates that, for Bha™™ojœ Dœkßita, the overriding authority of the three munis is accepted by older scholars like Kaiya™a and Haradatta. Bha™™oji 14: ‘This means to say that the decision [regarding the application of the grammatical rule] is in accordance with the [known] usage. The authority of the munis is successively higher. Thus, these days, the decisions regarding application of rules is made exclusively by following the statements of the author of the Bhåßya’ (lakßyånurodhåd vyavastheti bhåvaΔ Ù yathottaram iti Ù tathå ca adyatve bhåßyakåroktyaiva vyavastheti bhåvaΔ) (Prau∂hamanoramå, vol. 1, p. 139). Bha™™oji 15: ‘[In a particular rule] the word ca ‘and’ is meant for the inclusion of what has not been explicitly stated. That is to say that what has not been explicitly stated should be determined on the basis of the Bhåßya and the Vårttika’ (iti cakåro ’nuktasamuccayårthaΔ Ù anukta∫ ca bhåßya-vårttikabalån nirñeyam iti bhåvaΔ) (Prau∂hamanoramå, vol. 1, p. 171). Bha™™oji 16: ‘If then the author of Prasåda broke apart the original [rule of Påñini] by following the opinions of Jayåditya, Jinendrabuddhi, Haradatta etc., what is such an offence? Therefore, [one should note that] the situation is such that all the works in this regard, from Kaiya™a to [our] contemporaries, are rather loose [in their interpretation]. Conclusions have already been drawn in agreement with the three munis’ (prasådakåro ’pi tarhi jayådityanyåsakåraharadattådy anurodhena mülagrantham abhåõkßœd iti ki∫ tenåparåddham Ù tasmåt kaiya™aprabhr¢tyarvåcœnaparyanta∫ sarveßå∫ granthå iha †ithilå eveti sthitam Ù nißkarßas tu munitrayåvirodhena kr¢ta eva) (Prau∂hamanoramå, vol. 3, p. 684). Bha™™oji 17: ‘Moreover, these days, the opinions of Kå†akr¢tsna etc., in the face of a conflict with the three munis, are certainly not accepted. It is a fact that, even among the three munis, by following their successively rising authority, decisions are made only in accordance with the Bhåßya. There is agreement on this among everyone including Bhartr¢hari, Kaiya™a, Haradatta etc. Therefore, one should not cause distress to one’s mind by imagined arguments of various sorts by supporting one or the other statements of recent [scholars, as against the ancient munis]’ (kim bahunå kå†akr¢tsnådimatånåm api munitrayavirodhe idånœm agrahañam eva Ù [...] munitrayamadhye ’pi yathottara∫ pråmåñyam å†ritya bhåßyånurodhenaiva vyavastheti tattvam Ù etac ca bhartr¢harikaiya™aharadattådisakalasammatam Ù ato ’rvåcœna∫ yat ki∞cid vacanam puraskr¢tyånekaüpotprekßayå mano na khedanœyam iti dik) (Prau∂hamanoramå, vol. 3, p. 646). This statement clearly lays out the ‘historical’ understanding of the grammatical tradition that forms the basis of Bha™™oji Dœkßita’s performance seen in his works.

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 377

The passages cited above from Bha™™oji Dœkßita’s ‡abdakaustubha and Prau∂hamanoramå are only a small selection from a massive number of similar statements, but even this small sample is sufficient to indicate the ideological foundations of Bha™™oji Dœkßita’s exposition of Sanskrit grammar. It is clear that he has heavily drawn upon Kaiya™a’s statement that the authority of the three founding munis of Sanskrit grammar successively rises, and finally culminates in the towering personality of Pata∞jali. Bha™™oji says that from Bhartr¢hari to his own contemporary scholars, everyone accepts this towering authority of Pata∞jali, and those who do not accept it should be rejected. The reason for the authority of the three munis is that the elites today accept only these three munis as representing the grammar as a vedåõga, and hence the authority of other munis of ancient times is not to be independently accepted. Bha™™oji rejects the statements of all subsequent authorities, if they conflict with the statements of Pata∞jali. This sets apart the world of the three munis from the world of the post-Pata∞jali grammarians in terms of their abilities and hence their mode of operation. It also sets apart the world of the three munis apart from the spheres of other munis, especially if their views do not agree with the views of the three munis of the Påñinian tradition. For Bha™™oji, the reason to prefer the three munis of the Påñinian tradition over the other ancient munis is that by his time it was the Påñinian tradition that was exclusively given the status of vedåõgavyåkaraña. This assertion of a form of neo-Vedicism is rooted in a historical context different from that of Bhartr¢hari.13 For Bha™™oji Dœkßita, the most immediate context is that of the popularity of non-Påñinian grammars like Kåtantra and Mugdhabodha, as well as the claims on the part of contemporary Mœmå∫sakas and Naiyåyikas to represent the true understanding of language. Within the world of the munis, it is finally the opinion of Pata∞jali that carries the day. While he retains the right to reject the formulations of Påñini and Kåtyåyana, the post-Pata∞jali grammarians do not have a similar right to override the opinions of Pata∞jali. This defensiveness and the insistence upon exclusively following the authority of the three munis is taking place in the works of Bha™™oji Dœkßita particularly in the context of a demonstrable willingness on the part of some of his rivals in the tradition of the Prakriyåkaumudœ to accept other authorities besides the Påñinian tradition. This appears to be Bha™™oji’s response to his immediate contemporary rivals, like Vi™™hala the author of the commentary Prasåda on the Prakriyåkaumudœ, who are willing to base their statements on non-Påñinian authorities like Cåkravarmaña or Bopadeva.14 It was within this context that Bha™™oji developed his interpretive principles that he partly inherited from the 13

See Deshpande forthcoming. See ‡abdakaustubha, vol. 1, p. 165: yat tu ka†cid åha cåkravarmañavyåkarañe dvaya†abdasyåpi sarvanåmatåbhyupagamåd tadrœtyåya∫ prayoga iti tad api na, munitrayamatenedånœ∫ sådhvasådhuvibhågas tasyaivedånœ∫tana†iß™air vedåõgatayå parigr¢hœtatvåt. Also 14

378

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

tradition going back to Bhartr¢hari and Kaiya™a, but which he tweaked to perfection and directed against his most immediate rivals like Vi™™hala and others. Thus, even though we have not found the terms lakßyaika-cakßußka and lakßañaika-cakßußka in the available works of Bha™™oji Dœkßita, it is safe to say that most of the core conceptions associated with these terms in the works of ‡ivaråmendrasarasvatœ and Haridœkßita are already present in the statements of Bha™™oji Dœkßita. Now we can turn to the usage of the terms lakßañaika-cakßußka and lakßyaika-cakßußka in the works of Någe†abha™™a. But before discussing specific instances of these terms in his works, we need to understand Någe†a’s general conceptions. The Laghu†abdendu†ekhara of Någe†a makes a clear distinction between his contemporary users of Sanskrit and the standard-setting ancient †iß™as. Någe†a says: The vivakßå ‘desire to speak’ is two-fold, that belonging to [particular] users, and that determined by [the elites of] the world. The word ‘users’ refers to users like us who use the language for transactions. The word loka ‘world’ refers to those elites (†iß™a) who know everything that should be known, this and the yonder.15 In the [linguistic] world, this is the notion of linguistic elites as explained in the Bhåßya on P. 6.3.109 etc. Here the first type of desire to speak is not intended, since the users [like us] are mostly engaged in linguistic usage that is incorrect (apabhra∫†aprayoga). Therefore, it is the second type [that is intended]. Words those elites use, assuming them to be meritorious, become subject of the grammatical †åstra. [...] This is clear in the Bhåßya (vivakßå dvividhå pråyoktrœ laukikœ ca Ù prayoktr¢†abdena vyavahårårtha∫ prayogakartåro ’smadådayaΔ Ù loka†abdena viditavedyaΔ paråvaraj∞aΔ †iß™aΔ Ù loke etad eva †iß™atva∫ pr¢ßodarådibhåßyoktam Ù tathåpy atra nådyå vivakßitå Ù teßåm apabhra∫†aprayoge båhulyena pravr¢tteΔ Ù tasmåd antyaiva Ù te hi tatrårthe ya∫ †abda∫ dharmabuddhyå prayu∞jate teßå∫ †åstravißayateti [...] bhåßye spaß™am Ù) (Laghu†abdendu†ekhara (B), pt. 1, p. 479).

Our understanding of the terms lakßyaika-cakßußka and lakßañaikacakßußka must be grounded in our understanding of the general conceptual structures reflected the works of Någe†a and others. We can now turn to the use of these terms in the works of Någe†a. Aklujkar comments: “another fact that should be noted is that the number of occurrences in which Någe†a uses lakßañaika-cakßußka see Prau∂hamanoramå, vol. 2, pp. 454-455: tat sarva∫ bopadevagrantha†raddhåjå∂yapravr¢ttaprasådagranthamülakam iti mülå†uddher evopekßañœyam Ù [...] våmanådayas tu mahåbhåßyavisa∫vådåd upekßyå iti Ù etac ca sarvam bopadevagranthaparyålocanayå paryavasyatœti satyam eva, kintu bopadevasyaiveha mahån pramådaΔ Ù. Consider the willingness of ‡rœkr¢ßña, the author of the commentary Prakå†a on the Prakriyåkaumudi (vol. 2, p. 421), to admit authorities beyond the Påñinian munis: ekavacanam iti tu bhåßyådyanuktam api durgapari†iß™akañ™håbharañådibahu granthasammatyoktam. It is this tendency among some of his contemporaries that Bha™™oji is reacting against in his works. 15 Also see: †iß™å† ca †abdatattvasåkßåtkåravanto yogina iti bhåßye spaß™am Ù (Laghu†abdendu†ekhara, pt. 1, p. 559). For details of how this view of †iß™a is inherited from Bhartr¢hari, see Deshpande, forthcoming. The notion of vågyoga-vid itself already occurs in the Mahåbhåßya (Kielhorn, vol. 1, p. 2): so ’nantam åpnoti jayam paratra vågyogavid [...] Ù.

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 379

and lakßyaika-cakßußka is far smaller than what we would expect the case to be if, through these terms, he was expressing a fundamental understanding of his about the history of Sanskrit language or of Påñinian grammar. In all, Någe†a seems to have used the terms only five times […]” (Aklujkar forthcoming: §2.7). Of the five occurrences cited and analyzed by Aklujkar, two are from Någe†a’s Paribhåßendu†ekhara (on Paribhåßå 50; 62-63), and three are from his Uddyota (on P. 2.1.1, P. 2.3.46, P. 8.3.15). However, the terms also occur in Någe†a’s Laghu†abdendu†ekhara, Br¢hacchabdendu†ekhara and Laghu†abdaratna, which has been traditionally ascribed often to Någe†a’s teacher Haridœkßita. In contrast with Aklujkar’s statements cited above, these terms have a wider usage in Någe†a’s own works. They have an ancestry in the works of ‡ivaråmendrasarasvatœ and Någe†a’s own teacher Haridœkßita, and the terms remain significant in the works of Vaidyanåtha Påyaguñ∂e, a direct disciple of Någe†a. Någe†a’s Laghu†abdendu†ekhara (A) (pp. 427-428) says: ‘Since the Bhåßyakåra is lakßyaika-cakßußka, this interpretation alone is appropriate’ (ki∞ ca bhåßyakårasya lakßyaikacakßußkatayå iyam eva vyåkhyocitå). This sort of explicit identification of Pata∞jali as being lakßyaika-cakßußka by ‡ivaråmendrasarasvatœ, Haridœkßita and Någe†a needs to be contrasted with the rather dismissive comments by Aklujkar: “the context needed for the use of lakßañaika-cakßußka and lakßyaika-cakßußka is not one in which the specialists of Påñini like Pata∞jali, Bhartr¢hari, Kaiya™a or Bha™™oji Dœkßita must be presupposed. In no passage, any of these ‘professional’ or first-rank Påñinœyas, whose works are commonly studied, are mentioned” (Aklujkar forthcoming: §2.8). Not only is Pata∞jali clearly identified as being lakßyaika-cakßußka, and sometimes as being the only lakßyaika-cakßußka, by ‡ivaråmendrasarasvatœ, Haridœkßita and Någe†a, this is precisely the reason why ‡ivaråmendrasarasvatœ labels Kaiya™a’s occasional lakßyaika-cakßußka explanations as being prau∂himåtra ‘conceited’ since Kaiya™a does not fall within the group of r¢ßis, who alone may be characterized as being lakßyaika-cakßußka. In what follows, I shall briefly analyze the relevant passages in Någe†a’s works, without going into an extensive analysis of the over-all grammatical arguments in each case. Nage†a 1: ‘Even those who say, as demanded by the form to be derived, that the directly available rule of Påñini (8.2.1) should be overridden by an inferred metarule (i.e. antaraõga-paribhåßå), should not be respected by those whose eyes are solely fixed on the grammatical rules’ (ye ’pi lakßyånurodhåd ånumånikyåpy antaraõgaparibhåßayå pratyakßasiddhasya pürvatrety asya (P. 8.2.1) bådha∫ vadanti te ’pi lakßañaikacakßurbhir nådartavyåΔ Ù) (Paribhåßendußekhara, Paribhåßå 50). Aklujkar (forthcoming: §2.7) translates lakßyånurodhåt by “in their pursuit of the forms to be derived”. The passage seems to be saying

380

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

that only the lakßyaika-cakßußka grammarians like Pata∞jali have the direct unmediated access to the usage, and hence only they can make lakßyaika-cakßußka arguments. This intent of Någe†a is clearly brought out by Vaidyanåtha Påyaguñ∂e in his commentary Gadå on the Paribhåßendu†ekhara (p. 92): ‘[Here Någe†a] rejects the statements of Kaiya™a etc. Those who [...] by this argument the claim that the lakßyaika-cakßußkas are justified in accepting that view is rejected. [...] That characteristic [of being lakßyaika-cakßußka] does not occur in anyone other than the bhagavat [=Pata∞jali]’ (kaiya™ådy-ukti∫ khañ∂yati –ye ’pœti’– Ù [...] etena lakßyaikacakßußå∫ tadådaro yukta ity apåstam Ù [...] Ù bhagavadatirikte tattvåbhåvåt). Någe†a 2: ‘The lakßañaika-cakßußka person surveys the domain of exception-rules, and only after determining that an example does not fall within the domain of the exception-rules, he processes it with the generic rule. [...] [On the other hand], the lakßyaika-cakßußka person [naturally] excludes the domains of the exception-rules even without reviewing the exception-rules, and processes [appropriate] examples with the generic rules. [Even a lakßyaika-cakßußka finally processes his examples with rules of grammar], since, even for him, generation of religious merit takes place only with a [proper] usage backed by the conscious awareness of the grammatical processes’ (lakßañaikacakßußko hy apavådavißayam paryålocya tadvißayatvåbhåvani†caya utsargeña tattallakßya∫ sa∫skaroti Ù [...] lakßyaikacakßußkas tu tattacchåstraparyålocana∫ vinåpy apavåda vißayam parityajyotsargeña lakßya∫ sa∫skaroti Ù tasyåpi †åstra prakriyåsmarañapürvakaprayoga eva dharmotpatteΔ Ù) (Paribhåßendu†ekhara, paribhåßå 62-63). In the initial distinction between the two classes, it is made apparent that the lakßañaika-cakßußka person has no ability to skip any rules of grammar. On the other hand, the lakßyaika-cakßußka person can afford to skip all the exception-rules because of his natural familiarity with the usage, and go directly to the generic rule. In an interesting twist, a question is raised as to why does he need to go to any rules of grammar, if he knows the usage directly. The tradition of grammar grounded in the notion of dharma does not allow even a true lakßyaika-cakßußka to escape the rules of grammar completely. He does not need to use grammar, if his practical purpose is merely to be able to speak correctly. However, if he hopes to acquire religious merit, he too must ground his usage on the prior knowledge of the grammatical †åstra.16 Någe†a 3: On a passage almost identical with the one above (Någe†a 2) found in Någe†a’s Laghu†abdendu†ekhara (A) (p. 4416 See Vårttika 9 on Aß™ådhyåyœ (†åstrapürvake prayoge ’bhyudayas, tat tulya∫ veda†abdena).

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 381

47), Vaidyanåtha Påyaguñ∂e, in his commentary Cidasthimålå, unpacks the significance of the last remark of Någe†a: ‘Now a [true] lakßyaika-cakßußka has no need for rules of grammar, and hence why say that he too processes his examples with rules of grammar? To explicate this, [Någe†a] says this. Even a lakßyaika-cakßußka [must process his usage with rules of grammar]. The word ‘only’ rejects the possibility of generation of religious merit in any other way’ (nanu lakßyaikacakßußkasya lakßañåpekßaiva neti katha∫ tena tatsa∫skåro ’ta åha - tasyåpœti lakßyaikacakßußkasyåpœty arthaΔ Ù [...] Ù evena tadanyathå dharmotpattiniråsaΔ Ù). Here, Påyaguñ∂e is bringing out the dilemma that he and his teacher are caught in. For practical purposes, a true lakßyaika-cakßußka does not need to have his usage backed by grammar, but for religious purposes he needs to do so. This answers one of the concerns of Aklujkar that if Pata∞jali is to be regarded as a true lakßyaika-cakßußka, as he indeed is regarded by Haridœkßita, Någe†a, and Vaidyanåtha, why should the same Pata∞jali be still involved with justifying the use of grammar. Pata∞jali himself is involved in contradictory pressures. While describing his †iß™a, he says that this person does not study the Aß™ådhyåyœ and yet speaks proper Sanskrit (see na cåß™ådhyåyœm adhœte, ye cåsyå∫ vihitåΔ †abdås tå∫† ca prayuõkte [Mahåbhåßya on P. 6.3.109]). But the same Pata∞jali cites Kåtyåyana’s introductory Vårttika 9 (on Påñini’s Aß™ådhyåyœ) that asserts: †åstrapürvake prayoge ’bhyudayaΔ. Thus, the dilemma regarding how to imagine the function of a true lakßyaika-cakßußka person has continued to persist. Någe†a 4: On Pata∞jali’s words ‘Even if the word samartha is explicitly included in the rule’ (atha kriyamåñe ’pi samarthagrahañe [on P. 2.1.1]), Kaiya™a remarks that such an explicit inclusion may be needed to inform the unlearned (abudhabodhanårtham). Någe†a in his Uddyota on this passage (NSP ed., vol. 2, p. 321) renders Kaiya™a’s expression ‘unlearned’ (abudha) with lakßañaika-cakßußka: ‘For those whose eyes are solely fixed on rules of grammar, it is difficult to figure out in all cases what is assumed and what is not’ (abudheti sarvatra gamakatvågamakatve lakßañaikacakßußkair durj∞eye iti bhåvaΔ). The lakßañaika-cakßußkas not only don’t have unmediated access to usage, they can evidently only know the rules of grammar as they are explicitly stated. They don’t always have the ability to figure out what is and what is not implied. Någe†a 5: ‘The lakßyaika-cakßußka person, even before seeing the words to be derived, determines the domain of the generic rule, excluded by the domains of the exception-rules’ (lakßyaikacakßußko hi lakßyadar†anåt pråg evotsargasyåpavådavißayåtirikta∫ vißaya∫ nirñayati) (Uddyota [Mahåbhåßya on P. 2.3.46, NSP ed., vol. 2, p. 516]).

382

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Någe†a 6: ‘It is the conceit [of those commentators assuming the lakßyaika-cakßußka point of view] that, by the force of the target usage, in some places a direct rule is overridden by an inferred metarule. One should know that for the lakßañaikacakßußka persons, explicit statements are necessary’ (lakßyava†åd kvacid ånumånikenåpi pratyakßabådha ity abhi-månaΔ Ù lakßañaikacakßußå∫ tu vacanåny evåva†yakånœti bodhyam Ù) (Uddyota [Mahåbhåßya on P. 8.3.15]). Någe†a 7: yat tu lakßyånurodhåt kåryakålapakße upasthitayå antaraõga-paribhåßayånumånikyå pratyakßasiddhasyåpi pürvatrety asya bådha iti, tanna lakßañaikacakßusåm ådarañœyam (Mahåbhåßya NSP ed., vol. 2, p. 297). This argument is identical with the one in Nage†a 1 given above. On this passage, the Cidasthimålå of Vaidyanåtha Påyaguñ∂e makes a statement that is even more explicit than the one cited earlier: ‘Now, if one argues that even then a lakßyaika-cakßußka must accept this argument, then that is not so. That characteristic [of being a lakßyaikacakßußka] is not found in anyone other than the bhagavat [=Pata∞jali], and the grammatical †åstra is meant for people like us’ (nanv evam api lakßyaikacakßußå tadådara eveti cen na Ù bhagavadatirikte tattvåbhåvåc chåstrasyåsmadådœn prati pravr¢tte† ca Ù) (ibidem, p. 297). This comment from Vaidyanåtha offers an explanation of the distinction between these two terms that is in line with the tradition. We, the contemporary grammarians, are not in the same league as Pata∞jali, who alone, in the words of Haridœkßita, Någe†a, and Vaidyanåtha, is a true lakßyaika-cakßußka. Någe†a 8: ‘Thus rules prohibiting certain compounds and the specific mention of ‘obligatory’ (nityam) in P. 2.2.17 (nitya∫ krœ∂åjœvikayoΔ) are purposeless. [...] However, all of these statements are necessary for the dimwitted lakßañaika-cakßußka persons to inform them about where compounding is obligatory, where it is optional, and where it does not take place. [...] From the point of view of a lakßyaika-cakßußka, this is rejected in the Bhåßya on P. 2.1.1 (samarthaΔ padavidhiΔ)’ (eva∫ samåsanißedhasütråñi ‘nityam krœ∂e’ ty atra nitya-grahaña∫ ca vyartham [...] tathåpi lakßañaikacakßußkån mandån prati kvaikårthœbhåvo nityaΔ kva påkßikaΔ kva nåsty evety artha bodhanå rtha∫ sarvam åva†yakam Ù [...] lakßyaikacakßußka∫ prati ‘samartha’ iti sütre bhåsye tat pratyåkhyåtam Ù) (Laghu†abdendu†ekhara, p. 807). Någe†a 9: Br¢hacchabdendu†ekhara, vol. 1, p. 14. This is identical with Någe†a 7 passage cited above. Någe†a 10: ‘[This particular] explanation should be disregarded by the lakßañaika-cakßußka folks, since it is difficult to comprehend’ ([...] samådhåna∫ tu lakßañaikacakßurbhir durj∞eyatvåd upekßyam) (Br¢hacchabdendu†ekhara, vol. 1, p. 374).

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 383

Någe†a 11: lakßañaikacakßußkair lokasvabhåvam ajånadbhiΔ kvaikårthœ bhåvo nityaΔ kva påkßikaΔ kva nåsty evety arthasya j∞åtum a†akyatvena tadbodhanåya teßåm åva†yakatvåt Ù [...] bhagavatå tu arthåbhidhånasya svåbhåvikatvapakße lakßyaikacakßußkadr¢ß™yå vaiyarthyam uktam (Br¢hacchabdendu†ekhara, vol. 2, pp. 976-977). This argument is identical with the one found in passage Någe†a 8 above. What is additionally evident here is that it is the Bhagavån Pata∞jali who is entitled to offer an explanation from the lakßyaika-cakßußka point of view, and Någe†a will not grant that privilege to anyone else. Någe†a 12: ‘In its absence, a lakßañaika-cakßußka may get confused’ (tadabhåve hi lakßañaikacakßußko bhråmyed iti) (Br¢hacchabdendu†ekhara, vol. 2, p. 1315). Någe†a 13: ‘Since the Bhåßyakåra (=Pata∞jali) is lakßyaika-cakßußka, only this interpretation is appropriate’ (bhåßyakårasya laksyaikacakßußkatayaißaiva vyåkhyocitå) (Laghu†abdaratna, pp. 360-361). Någe†a 14: ‘Since the Bhåßyakåra is lakßyaika-cakßußka, this interpretation alone is appropriate’ (ki∞ ca bhåßyakårasya lakßyaikacakßußkatayå iyam eva vyåkhyocitå) (Laghu†abdendu†ekhara (A), pp. 427-428). From these fourteen passages from Någe†a’s works cited above, one can draw several clear conclusions. The most important conclusions are that for Någe†a, Bhagavån Pata∞jali is the only true lakßyaika-cakßußka, and only he is entitled to offer arguments from the laksyaika-cakßußka point of view. Time and again, Någe†a clarifies himself, and his intention is clarified by his disciple Vaidyanåtha, that the lakßañaika-cakßußka are contemporary [and especially post-Pata∞jali] scholars and users of Sanskrit like ourselves (asmadådi) who do not have the ability to check the usage directly. The †iß™as like Pata∞jali are superior to us. They are r¢ßis not just of another era, but they are deemed to possess great power of penance and superior knowledge, including yogic knowledge. In most of these respects, Någe†a’s conceptions seem to closely match those of Bhartr¢hari, as I have discussed at length in some previous as well as forthcoming publications.17 The conceptions of lakßañaika-cakßußka and lakßyaika-cakßußka that were inherited by Någe†a from ‡ivaråmendrasarasvatœ and Haridœkßita were significantly used by himself fairly widely, and were transmitted to his disciples like Vaidyanåtha Påyaguñ∂e. I have already cited several passages from Vaidyanåtha’s commentary Gadå on Någe†a’s Paribhåßendu†ekhara and from his commentary Cidasthimålå on Någe†a’s Laghu†abdendu†ekhara. Vaidyanåtha adds 17

I especially would like to refer to my Deshpande forthcoming.

384

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

unusual clarity to these terms. Consider the following passages from Vaidyanåtha’s commentary Chåyå on Någe†a’s Uddyota:18 Chåyå 1: ‘In that context, a lakßañaika-cakßußka desires to derive words only by the power of grammar’ (tatra hi lakßañaikacakßußko vyåkarañabalenaiva †abda vyutpattim icchati) (Mahåbhåßya, NSP ed., vol. 1, p. 149). Chåyå 2: ‘Since the grammarian-sages are lakßyaika-cakßußka’ (munœnå∫ lakßyaikacakßußkatvåt) (Mahåbhåßya, NSP ed., vol. 1, p. 290). It is thus clear that these terms are not purely incidental and rare in Någe†a’s works. These terms and the concepts they represent are fundamental in the interpretive techniques of authors like Någe†a, and they are conceptually grounded in the interpretive principles laid out by Bha™™oji Dœkßita in his works, even though these terms are themselves not found in his works. By using these categories, like the distinctions of upåya and upeya or nœtårtha and neyårtha in Buddhist works, authors like Någe†a are able to critique the received texts in a particular way.19 An explanation labeled as coming from a lakßyaika-cakßußka point of view, in effect, amounts to saying that a particular explanation bypasses several grammatical procedures and directly relies upon the knowledge of forms to be derived ahead of their grammatical derivation, but in the process makes many rules of grammar, or perhaps the whole grammar, rather vacuous. Such a privilege is granted by Någe†a only to Pata∞jali, the r¢ßi, and not to others. Other such explanations found in the works of authors like Kaiya™a are rejected as being prau∂hivåda ‘conceited statements, though not true’. The lakßañaikacakßußka characterization of a particular explanation or procedure seems to suggest that this particular procedure follows every explicitly stipulated grammatical rule, and does not, and cannot, take for granted the form to be derived ahead of its derivation. But beyond this, the terms have far wider ideological, mythological, and chronological associations, and I have tried to explicate these associations on the basis of ample textual evidence. The terms have an ancestry going back at least to ‡ivaråmendrasarasvatœ and Haridœkßita and they are inherited by Någe†a’s disciples like Vaidyanåtha Påyaguñ∂e among others. Though the terms themselves do not seem to go to a period before ‡ivaråmendrasarasvatœ, I have argued here that the underlying conceptions are fully manifest in the works of Bha™™oji Dœkßita. Most of these conceptions, except for the divinity of Pata∞jali and the divine origin of the 18 Other occurrences of these terms in Vaidyanåtha’s Chåyå are to be found on pp. 52; 198 (Mahåbhåßya, NSP ed., vol. 1). 19 Also compare the way the commentators divide up the Bhåßya into statements that are then labeled variously as representing the point of view of the Siddhåntin, Pürvapakßin, Ekade†in etc. If one disagrees with a particular statement in the Bhåßya, it was easier to describe that passage a statement of the Ekade†in, rather than calling it wrong.

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 385

‡ivasütras, are at least as old as Bhartr¢hari. The most important of these conceptions is the notion that the ancient world of the founding grammarians belonged to the age of r¢ßis, culminating in the towering personality of Bhagavån Pata∞jali, to use the reverential expression from the works of Bha™™oji Dœkßita, ‡ivaråmendrasarasvatœ, Haridœkßita, Någe†a, and Vaidyanåtha. Vaidyanåtha could not be more explicit: ‘since that characteristic [of being a lakßyaika-cakßußka] is not found in anyone other than Bhagavån Pata∞jali’ (bhagavadatirikte tattvåbhåvåt). With the textual evidence and analysis provided above, it should be evident that Någe†a’s conceptions of lakßañaika-cakßußka and lakßyaika-cakßußka are thus critical in our understanding of how the grammarians perceived the history of Sanskrit and the history of the Sanskrit grammatical tradition.

BIBLIOGRAPHY Sources Åpastambœyadharmasütram, edited by Georg Bhler, Government of Bombay, Mumbai 1868, pt. 1. ņvalåyanagr¢hyasütra, (with a commentary by Gårgyanåråyaña), edited by Vasudev Panashikar, Nirnayasagara Press, Mumbai 1894. Bhairavami†ra, Bhairavœ, commentary on ‡abdaratna, in Prau∂hamanoramå (with the commentaries ‡abdaratna, Bhairavœ, Bhåvaprakå†a, and Saralå), edited by Gopal Shastri Nene, Chowkhamba Sanskrit Series Office, Banaras, 1939, pt. 2, Kashi Sanskrit Series, Haridåsa Sanskrit Granthamålå, n. 125, Vyåkaraña Section, n. 16. Bha™™oji Dœkßita, Prau∂hamanoramå (A), with Laghu†abdaratna by Någe†abha™™a (ascribed to Haridœkßita on the title-page), pt. 1 Avyayœbhåvånta, edited by Madhav Shastri, 1920; pt. 2 From Tatpurußa to Yaõantaprakriya [no date on my copy]; pt. 3 Uttarårdha, edited by Ratnagopal Bhatta, Chowkhamba Sanskrit Series Office, Banaras 1907. Bha™™oji Dœkßita, Prau∂hamanoramå (B), with Br¢hacchabdaratna by Haridœkßita and Laghu†abdaratna by Någe†abha™™a, vol. 1, Avyayœbhåvånta, edited by Sitaram Shastri, Hindu Vishwavidyalaya Nepal-Rajya Sanskrit Series, n. 8, Banaras 1964. Bha™™oji Dœkßita, ‡abdakaustubha, edited by Gopal Shastri Nene and Mukund Shastri Puntamkar, vol. 1, fasc. 1 to I4, 1933; vol. 2, fasc. 5-10, 1929. Haridœkßita, Br¢hacchabdaratna. See Prau∂hamanoramå (B). Joshi, V.L., Prau∂hamanoramå, with commentary [Br¢hat]‡abdaratna,

386

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

vol. 1, Deccan College Monograph Series, n. 31, Deccan College, Pune 1966. Melputtür Nåråyañabha™™a, Apåñinœyapramåñatå, edited by E.V. Raman Namputiri, V.V. Press, Trivandrum 1942, Kerala Sanskrit Målå 1. Någe†a, Br¢hacchabdendu†ekhara, edited by Sitarama Shastri, Saraswati Bhavan Granthamålå, n. 87, Våråñaseya Sanskrit Vi†vavidyålaya, Banaras 1960, 3 voll. Någe†a, Laghu†abdendu†ekhara (A), Avyayœbhåvånta portion, with seven commentaries, edited by Guruprasad Shastri and Nand Kishore Shastri, ‡rœ Råjasthåna Sanskrit College Granthamålå, vol. 14, Bharatiya Pustakalaya, Banaras 1936. Någe†a, Laghu†abdendu†ekhara (B), in Vaiyåkaraña Siddhånta Kaumudœ, with the commentaries Tattvabodhinœ, Bålamanoramå, Laghu†abdendu†ekhara, and Subodhinœ, edited by Guru Prasad Shastri, Rajasthan Sanskrit College Granthamålå 42, Rajasthan Sanskrit College, Banaras 1940, 2 voll. Någe†a, Paribhåßendu†ekhara, with the commentary Gadå by Vaidyanåtha Påyaguñ∂e, edited by Ganesha Shastri Gokhale, Ånandå†rama Sanskrit Series, n. 72, Ånandå†rama, Pune 1913. Någe†abha™™a, Laghu†abdaratna, (ascribed to Haridœkßita on the titlepage), in Bha™™oji Dœkßita, Prau∂hamanoramå, (with the commentaries Laghu†abdaratna, Kucamardinœ, Jyotsnå, Prabhå, Ratnadœpikå, and Vibhå), edited by Sadashiva Sharma Shastri, Haridåsa Sanskrit Series, n. 23, Chowkhamba Sanskrit Series Office, Banaras 1934. Någe†abha™™a, Paribhåßendu†ekhara, critically edited with the commentary Tattvådar†a of Våsudev Shastri Abhyankar, by K.V. Abhyankar, pt. 1, Bhandarkar Oriental Research Institute, Pune. Pata∞jali, Mahåbhåßya, edited by Franz Kielhorn, 3rd revised edition by K.V. Abhyankar, Bhandarkar Oriental Research Institute, Pune 1962-1972 [1880-1885], 3 voll. Pata∞jali, Mahåbhåßya, with the commentaries Pradœpa by Kaiya™a and Uddyota by Någe†abha™™a, Motilal Banarsidass, Delhi 1967, 3 voll. Pata∞jali, Mahåbhåßya, with the commentaries Pradœpa, Uddyota, and Chåyå, vol. 1, edited by Shivadatta D. Kuddala, 1917; vol. 2, Nirnaya Sagara Press, Mumbai 1912. Råmacandra, Prakriyåkaumudœ, with the commentary Prakå†a by ‡rœkr¢ßña, edited by Muralidhara Mishra, Sarasvati Bhavana Granthamala, n. 112, Sampurnanand Sanskrit Vishvavidyalaya, Banaras, vol. 1, 1977; vol. 2, 1977; vol. 3, 1980. ‡ivaråmendrasarasvatœ, Ratnaprakå†a, in Mahåbhåßya-PradœpaVyåkhyånåni, edited by M.S. Narasimhacharya, Publications

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 387

de l’Institut Français d’Indologie, n. 51, vol. 2, 1975; vol. 4, 1977; vol. 5, 1978; vol. 6, 1979. Vaidyanåtha Påyaguñ∂e, Bhåvaprakå†a, (commentary on [Laghu]†abdaratna), in Prau∂hamanoramå, (with the commentaries ‡abdaratna, Bhairavœ, Bhåvaprakå†a and Saralå), edited by Gopal Shastri Nene, Chowkhamba Sanskrit Series Office, Banaras 1939, pt. 2, Kashi Sanskrit Series, Haridåsa Sanskrit Granthamålå, n. 125, Vyåkaraña Section, n. 16. Vaidyanåtha Påyaguñ∂e, Cidasthimålå. See Laghu†abdendu†ekhara (A). Vaidyanåtha Påyaguñ∂e, Gadå. See Paribhåßendu†ekhara (Ånandå†rama ed.). Studies Aklujkar, A. (forthcoming), Where do lakßañaika-cakßußka and lakßyaika-cakßußka apply?, in Kelkar Festschrift, in «Bulletin of the Deccan College Research Institute», voll. 62-63(2002-2003), pp. 167-188. Deshpande, M.M. (1985), Historical Change and the Theology of Eternal Sanskrit, in «Zeitschrift fr vergleichende Sprachforschung», vol. 98, n. 1, pp. 122-149. ______ (1993), The Changing Notion of ‡iß™as from Pata∞jali to Bhartr¢hari, in «Asiatische Studien», vol. 47, n. 1, pp. 95-115. ______ (1996), The Vedic Traditions and Origins of Grammatical Thought in Ancient India, in ‘Langue, style et structure dans le monde indien’, Bibliothèque de l’École des hautes Études, Sciences Historiques et Philologiques, H. Champion, Paris, pp. 145-170. ______ (1997), Who inspired Påñini? Reconstructing the Hindu and Buddhist Counter-Claims, in «Journal of the American Oriental Society», vol. 117, n. 3, pp. 444-465. ______ (1998), Evolution of the Notion of Authority in the Påñinian Tradition, in Histoire Epistemologie Langage, Tome XX, Fasc. 1, Les Grammaires Indiennes, pp. 5-28. ______ (2001), The Vedic Context of Påñini’s Grammar, in Kniffka, H. (ed.), Indigenous Grammar Across Cultures, Peter Lang, Frankfurt, pp. 33-51. ____ (forthcoming), Revisiting the Notion of ‡iß™as in Bhartr¢hari (To appear in the proceedings of the ‘International Seminar on Bhartr¢hari’, held in Delhi in December 2003, to be published by Motilal Banarsidass).

MARIA PIERA CANDOTTI Loke, vede, †åstre: Grammarians’ Partition of Tradition and Related Linguistic Domains

Indian grammarians knew of linguistic variation: in Påñini’s Aß™ådhyåyœ we already find rules to describe Vedic usages, marginal and preferred usages, even regionally restricted usages. Still, this awareness conflicts with a likewise deeply rooted belief in the intrinsic inalterability of language that Dehspande calls the ‘theology of Eternal Sanskrit’.1 This leads, as far as Påñini is concerned, to a description of Sanskrit language as a panchronistic flatland, i.e. as a totality which includes “all known diachronic and synchronic facts of Sanskrit” (Deshpande 1985: 124). Påñinian grammar is based on a common set of rules which represents all the shared facts of Sanskrit, including what we would consider Vedic facts, as long as they are not exclusively Vedic.2 Of course Påñini and his commentators distinguish restricted domains inside this totality; these are generally accounted for in rules limited by what later commentators call ‘a locative of domain’ (vißayasaptamœ), i.e. rules which apply exclusively in the domain of mantras (mantre), in 1 It is by no means possible to account here for the huge amount of literature devoted to the notion of Sanskrit and Vedic language among grammarians. Nevertheless, some of these materials (Cardona 1990; Deshpande 1985 and 1993) may be used as an excellent introduction to this topic. 2 Pata∞jali’s description (M v. I, p. 9, l. 20-23 ad vt. 5) of this huge domain of Sanskrit usage is almost bewilderingly vivid: ‘The domain of the usage of words is very extensive indeed. The earth has seven continents, there are three words and four Veda-collections alongside with their secret teachings, variously divided: one hundred branches of the Yajurveda, one thousand ways of the Såmaveda, twenty-one times fold is the R¢gveda and nine times fold the Atharvaveda; debates, historical narrative (itihåsa), mythological narrative (puråña) and medical texts: this is the domain of the usage of words’ (mahån hi †abdasya prayogavißayaΔ Ù saptadvœpå vasumatœ trayo lokå† catvåro vedåΔ såõgåΔ sarahasyå bahudhå vibhinnå eka†atam adhvaryu†åkhåΔ sahasravartmå såmaveda ekavi∫†atidhå båhvr¢cya∫ navadhåtharvaño vedo våkovåkyam itihåsaΔ puråña∫ vaidyakam ity etåvå∞ †abdasya prayogavißayaΔ).

390

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

the domain of the Vedic hymns (chandasi), in the domain of ordinary language (bhåßåyåm) and so on.3 Nevertheless these domains are not identified –through historical, regional or social criteria– as different linguistic systems, but simply as sub-sets of rules deviating from the set of common rules. Already Renou pointed out that “the chandas [...] do not represent in any way a state of language as distinct from the normal; it is, in the interior of a common language, a fringe of archaism or ‘diversities’ which, for some reason or other mark the margin of the system” (Renou 1941: 248-249). Still, this subdivision of language facts in different restricted domains of usage is much more than a mere device to facilitate grammatical description. Grammarians, despite their ideological commitment to the idea of ‘Eternal Sanskrit’, were fully aware of the fact that actual linguistic usage is characterized by the above mentioned kind of internal variation. In other words, restricted linguistic usages, in grammarians’ conception, are not only useful devices to keep linguistic exceptions out of the bulk of commonly shared rules, but have some reality in everyday linguistic practice as well. Some of these variation domains are, as we have already seen, merely domains of diachronic or diatopic variation. Others, which we will concentrate upon in this paper, are so to say linked with specific human activities: they are what we would call domains of specialized languages, scholastic languages. These domains are somewhat different from the preceding ones. They entail both consciousness and voluntariness in the process of modifying the language and in the transmission of the modified language itself; they are characterized by the creation of technical terms and other conventions and by the transmission of these features inside one’s own school. In a sense, specialized languages are the linguistic outward sign of the construction and transmission of different ‘traditions’ which the grammarians tried to account for within their ideological framework. One of the fundamental elements in this ideological construction of the concept of Sanskrit was the reference to the †iß™as, i.e. to the cultural élites that were considered the unique depositary of correct linguistic usage. This reminds us that, at least as far as explicit declarations of intents are concerned, the aim of grammar was supposed to be that of merely describing the linguistic knowledge and practice of cultural élites, and not to create a standard by its own strength. Grammar was considered to be grounded on usage (prayoga) and to depend on it.4 The élites granted, in some sense, a synchronic dimension in the panchronistic flatland described by grammar: all the his3 Few lines below, in the same context already quoted (M v. I, p. 9, l. 24 - p. 10, l. 1), Pata∞jali asserts that in this extended domain of usage there are words which are found only in some places and have a restricted domain (te te †abdås tatra tatra niyatavißayå dr¢†yante); the examples he gives are examples of regional variants. 4 In the well-known Pata∞jali’s passage concerning the notion of †iß™as (M v. III, p. 174, l. 4-10 ad A 6.3.109), the author maintains that reference to an extra-grammatical author-

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 391

torical stages and socio-linguistic levels of Sanskrit grammar accounts for were so to say actualized “in the contemporary standard usage of the language and the contemporary knowledge of the past linguistic usage” (Deshpande 1985: 132) of those élites. Nevertheless, the notion of †iß™a and its adequacy to the actual socio-linguistic position and usage of Sanskrit varied deeply through the ages. Deshpande treats with great detail the evolution of this concept from Pata∞jali’s to Bhartr¢hari’s times (Deshpande 1993). The main tenet of the author is that the †iß™as became slowly identified with the grammarians themselves –despite the risk of circularity which Pata∞jali had already pointed out– or rather with the most ancient authorities in grammar. Bhartr¢hari at times calls these ancient authorities r¢ßis or munis, shifting them even further away, in a kind of mythic past. In the author’s opinion this hints at the fact that by Bhartr¢hari’s times there was no contemporary usage of Sanskrit, independent from grammar, which could play the role of authoritative source to test grammarians’ Sanskrit. In the present paper I would like to present some further reflections on how grammarians worked out this concept of extragrammatical sources of authority, and on the transmission of this authoritativeness, more specifically in the restricted ‘scholastic’ domains discussed above. I will principally concentrate on the most ancient data available among Påñini’s commentators, especially on Kåtyåyana (3rd century B.C.), Pata∞jali (2nd century B.C.) Bhartr¢hari (5th century A.D.) and, only occasionally, on Kaiya™a (11th-12th century A.D.). One point I hope to argue is that in the way grammarians conceptualize these restricted domains of activities and their source of authority we may trace the same shift towards a purely textual dimension Deshpande has pointed out in the case of the notion of †iß™a itself. A second point of concern of the present paper will be the idea of transmission, inside restricted domains, of knowledge concerning specific linguistic features and specific linguistic behaviours: grammarians’ way of interpreting transmission of knowledge on the meaning of some technical terms inside a specific school, for example, gives us interesting clues on how they considered the building up of specific traditions. 1. On the meaning of the words ‘laukika’ and ‘vaidika’ Some vårttikas of Kåtyåyana corroborate various statements concerning the technical language or the practices of grammar by pointing out that similar practices may be found in a mundane (laukika) or in a Vedic (vaidika) context. We find an example of this discursive pattern in the very first vårttika where Kåtyåyana states ity on language is necessary to avoid the circularity that would arouse from asserting that grammar accounts for correct linguistic usage and, at the same time, that correct usage is grounded on the language taught by grammar.

392

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

that grammar’s aim is to limit linguistic usage in view of merit, just as it happens in things concerning the world and the Veda (yathå laukikavaidikeßu).5 Pata∞jali generally gives examples to support Kåtyåyana’s statement. In the above mentioned vårttika, for instance, Pata∞jali mentions the existence of food restrictions and marriage restrictions in everyday life, while the Vedic examples are restrictions on food which may be object of religious vows, on the material to construct the sacrificial post and on the mantras one must recite while performing some sacrificial action. These vårttikas seem to presuppose a tripartition, loka, veda and, implicitly, †åstra (or, more specifically, vyåkaraña), that structures human experience. Simple as it may seem this statement yathå laukikavaidikeßu presents nevertheless some problems. This may be quite easily seen, for instance, in Filliozat’s translation “comme il est fait dans des textes du monde et du veda” (Filliozat 1975: 101): the reference to a domain of texts is by no means justified by the wording itself but, as we will see, it follows Kaiya™a’s interpretation of the passage. The shortcomings of this translation appear even more clearly few lines below where Filliozat, on the assumption that Pata∞jali himself asserts that the two expressions, yathå laukikavaidikeßu and yathå loke vede ca are equivalent, is obliged to translate “comme dans des textes du monde et du veda” even the sentence yathå loke vede ca. Still, if we question a purely textual interpretation of loka and even of veda, it is necessary to try to get a better understanding of how these domains are construed and what is their relationship with things qualified as being laukika (mundane) or vaidika (Vedic). Is it a relationship between the whole and its parts (a laukika- or vaidikathing is a part of loka, a part of veda), or a relationship between a place and a thing located in it, or something else? An answer to these questions will give us a hint on how to interpret also the third domain, and on the criterion used to qualify something as †åstric. Pata∞jali, while commenting the first occurrence of the expression yathå laukikavaidikeßu, namely in vt. 1 quoted above, tackles the problem of the formation and meaning of these two words: he wonders why Kåtyåyana chose this formulation, with a secondary suffix, instead of the simpler ‘like (it happens) in the world and in the Veda’ (yathå loke vede ca). Pata∞jali offers two possible answers to this question.6 He first proposes to consider that the secondary suffix doesn’t signify any additional meaning and that its use is only justified by the love that people of the south nourish for secondary suffixes: the two formulae yathå laukikavaidikeßu and yathå loke vede ca are, in this view, almost 5 M v. I, p. 6, l. 16 - p. 8, l. 7 ad vt. 1: (siddhe †abdårthasa∫bandhe [...] lokato ’rthaprayukte †abdaprayoge †åstreña dharmaniyamaΔ [...] yathå laukikavaidikeßu). 6 M v. I, p. 8, l. 8-10, ad vt. 1: (priyataddhitå dåkßiñåtyå yathå loke vede ceti prayoktavye yathå laukikavaidikeßv iti prayu∞jate Ù atha vå yukta eva taddhitårthaΔ Ù yathå laukikeßu vaidikeßu ca kr¢tånteßu).

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 393

perfectly equivalent.7 The second answer is that the secondary suffix should on the contrary be considered as conveying an adjectival meaning, and the whole formula be read as ‘like in case of mundane or Vedic kr¢tåntas’ (yathå laukikavaidikeßu kr¢tånteßu). Unfortunately the world supplied, kr¢tånta, is not easily interpreted. In Pata∞jali we find it exclusively in the context of the discussions on yathå laukikavaidikeßu. Outside the domain of grammatical literature, it means the result of past actions –therefore destiny, fate– and the result of a reasoning, therefore a conclusion or a well-established opinion. Bhartr¢hari (imitated by Kaiya™a), glosses this term with ‘authoritative opinion’ (siddhånta) and then with ‘the collection of prescriptive treatises’ (samayagranthasa∫darbha) (D1 p. 26, l. 5-6, ad vt. 1), but there is no reason to accept such a restrictive interpretation. The analysis of Pata∞jali’s examples will show that a krtånta is a well-founded habit or belief, more than an authoritative statement strictly bound to some textual source. In Bhartr¢hari’s (and even more in Kaiya™a’s) view, on the other hand, the interpretation of kr¢tånta as siddhånta leads to a much more text-oriented interpretation of the mundane and Vedic domain. Bhartr¢hari’s comment on this passage offers some further elements to reflection. First of all, Bhartr¢hari does not accept, not even within the first view, that the two formulae yathå laukikavaidikeßu and yathå loke vede ca should be considered as equivalent. In fact –he points out– the first being plural and the second singular, they are set off also by their grammatical number. The usage of plural instead of singular is a means to signify the parts as something different from the totality. Similarly, one will use the expression ‘trees’ (vanaspatayaΔ), to signify multiplicity, whereas, if not interested in signifying each single tree as something different from the collection of trees, one will use the singular ‘thicket’ (vånaspatyam), as a collective noun. Thus, by saying yathå laukikavaidikeßu, Kåtyåyana, in Bhartr¢hari’s opinion, intended, at least, to highlight the fact that the world and the Veda are playing here the role of ‘substratum’ (ådhåra).8 In the alternative view, continues Bhartr¢hari, the relation is no more that of parts and totality: the secondary formations laukika and vaidika, having an adjectival meaning, presuppose another entity (in this case, the kr¢tånta), somehow linked with the object signified by the primary bases, i.e. the world or the Veda. 7 It was currently accepted, in grammatical tradition, that words signifying a totality may also signify parts of this totality (see M v. I, p. 12, l. 18-21, ad vt. 14). The simple primary base veda-, therefore, may as well signify the whole veda as well as a part of it, without the need of another suffix. 8 Ådhåra is a påñinian term, used while defining the grammatical locus of the action in A 1.4.45: ‘the substratum (ådhåra) [of the action] is called adhikaraña’ (ådhåro ’dhikarañam). In fact, the action is considered by grammarians to be principally located in the subject (in intransitive constructions) or in the object (in transitive constructions), the adhikaraña nevertheless is considered to be the substratum of the action indirectly, through the subject or the object.

394

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

In both views, the crucial point for Bhartr¢hari seems to be not to identify the world and the Veda with vaidika and laukika-things, but rather to consider the firsts as a substratum of the others. This substratum is linked to ‘things’ (be it words, texts, beliefs or events) either by a relation between parts and totality or (in the second view) by another relation which needs yet to be identified. Because of this relation these ‘things’ are referred to as vaidika or laukika. In fact, until now we have deliberately set aside one point which should nevertheless be important from a grammarian’s point of view. What exactly is the formation of the words laukika and vaidika implicit in Pata∞jali’s and Bhartr¢hari’s discussions? Which is the suffix involved in their formation? Neither here nor in other passages Pata∞jali gives us no clue. Kaiya™a, on the contrary, discusses the formation of these two words right the first time they appear at the beginning of Mahåbhåßya when, answering the question which words grammar is supposed to account for, Pata∞jali answers laukikånå∫ vaidikånå∫ ca (†abdånåm). Kaiya™a proposes two derivations for the word laukika; the first derivation is obtained through the suffix ™ha˚ (= -ika) in the sense of ‘(words) known in the word’ (loke viditåΔ [†abdåΔ]), as taught by A 5.1.44 (lokasarvalokå™ ™ha˚). The second derivation is taught in the sense of ‘(words) located in the world’ (loke bhavåΔ [†abdåΔ]), through A 4.3.60 (antaΔpürvapadå™ ™ha∞). This second derivation is also the only one proposed to account for the word vaidika, again in the sense of ‘(words) located in the Veda’. Still, as Kaiya™a itself points out, it is impossible to obtain vaidika and laukika directly from A 4.3.60; it is necessary to have recourse to an integration establishing a list of additional nominal bases to which the above mentioned suffix ™ha˚ may be added.9 This is surely not a very straightforward procedure of word-formation; it seems quite safe to assume that Kaiya™a proposed to derive laukika from A 4.3.60 (despite the existence of a more satisfactory derivation) only to keep a strict parallelism with the word vaidika, a parallelism which perhaps was not meant by Pata∞jali. This implies also that the difference between laukika and vaidika words seemed less sharp by Kaiya™a’s time.10 More than that, in the commentary ad vt. 1, commenting again on laukika and vaidika, Kaiya™a does not even mention the possibil9 Let us notice that this list does not mention either the base loka- or the base veda(see Birwé 1961: 441); to avoid this problem K ad A 4. 2.60 says the list is an åkr¢tigaña, i.e. a list which shows some examples of the elements concerned by the rule but which is not supposed to be exhaustive. 10 The unique Dœpikå’s manuscript available to us begins right in the middle of the discussion concerning vaidika and laukika words. Still, some interesting elements may be obtained from that passage. Particularly interesting is Bhartr¢hari’s statement that, among Vedic words taught by grammar, there are, of course, words found in Vedic texts, but also others, not handed down as such by tradition, which still have the power to bring about apürva, the sacrifice’s reward. Bhartr¢hari gives as an example the Vedic injunction ‘Counting the potsherds he disjoins them’ (sa∫khyåya kapålåny udvåsayati),

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 395

ity of deriving laukika in the sense of ‘known in the word’. Why? Perhaps because here the qualified term is no more †abda but, according to Pata∞jali, kr¢tånta, which Kaiya™a glosses as siddhånta, ‘authoritative opinion, doctrine’: these authoritative opinions seem thus to be found only in texts which became therefore their ‘context’ in a strict verbal construction of the domain (tatra bhåva). Even more explicitly, Kaiya™a further specifies that the laukika siddhånta is an authoritative opinion expressed in a text belonging to the smr¢ti, the vaidika siddhånta in a text belonging to the †ruti (see P v. I, p. 34, ad vt. 1). 2. Tripartition of the domain of linguistic usages: loka, veda and †åstra Let us now have a look at the examples Pata∞jali gives each time he comments Kåtyåyana’s vårttikas containing the parallelism yathå laukikavaidikeßu, and then check reactions of later commentators (principally Bhartr¢hari and Kaiya™a) to these examples. The first occurrence is the already mentioned commentary to vt. 1 (M v. I, p. 8, l. 10-22, ad vt. 1) where Kåtyåyana states that, linguistic usage being independent from grammar, grammar’s aim is to establish some limits in view of merit, just as it is done in well-established practices in the world and in the Veda: Loka

Veda

‘A tame cock should not be eaten, a tame pig should not be eaten’ (abhakßyo gråmyakukku™o ’bhakßyo gråmya†ükara ity ucyate).11 The act of eating is carried out to appease hunger. Still, hunger may be appeased also by eating dog’s meat. Therefore a restriction is made in the world, teaching that this may be eaten and this must not. ‘That woman is fit for marriage, that other is not’ (iya∫ gamyeyam agamyeti). Desire may be satisfied with woman fit for marriage as well as with one which is not. Therefore a restriction is made concerning which woman one must choose to satisfy his desire. ‘A Brahmin takes the vow of living on milk, a kßatriya of living on gruel, a vai†ya of living on curds’ (payovrato bråhmaño yavågüvrato råjanya åmikßåvrato vaicÛya ity ucyate).12

where the priest is supposed to disjoin the potsherds counting them eko dvau trayaΔ, ‘one, two three’, D1, p. 1 I, 9-14. The implicit reference seems here to be the grammatical process of üha, ‘modification, adaptation’ which, using (Deshpande 1985: 136) words, represents the “ability to effectively link [...] diachronic and synchronic facts of Sanskrit in the course of ritual”. 11 The statement is traced by Filliozat to Baudhåyanadharmasütra, 1.12.34 (Filliozat 1975: 101) and by Joshi and Roodbergen to Baudhåyanadharmasütra, 1.5.1-3 (Joshi, Roodbergen 1986: 121). 12 The statement is traced by Filliozat to Taittirœyårañyaka, 2.8, v. I, p. 141 (Filliozat 1975: 101) and by Joshi and Roodbergen to Taittirœyasa∫hitå, 6.2.5 in the context of the Soma sacrifice (Joshi, Roodbergen 1986: 124).

396

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

One could choose to observe the vow of living on rice or on meat. Therefore a restriction is made in the Veda. ‘The sacrificial post must be made of bilva or khådira wood’ (bailvaΔ khådiro vå yüpaΔ syåt ity ucyate).13 The sacrificial post is erected for the sake of tying the animals to it and one could tie the animals to any piece of wood. Therefore a restriction is made. ‘Once he has put the potsherds near the fire he recites the mantra: be you heated by the heat of the swat of the Bhr¢gu and of the Aõgiras’ (agnau kapålåny adhi†rityåbhimantrayate Ù bhr¢güñåm aõgiraså∫ gharmasya tapaså tapyadhvam iti).14 The fire would heat the potsherds even if the sacrificer does not recite the mantra; therefore a restriction is made to perform the sacrificial action together with the recitation of the mantra. ‡åstra Grammar too, considers that the intended meaning may be conveyed by correct words as well as by incorrect words. Therefore a restriction is made by grammar to express meaning only by using correct words. Bhartr¢hari, at the end of his commentary on the laukika-examples, introduces the Vedic ones by these words: ‘till now [Pata∞jali] has recalled the smr¢ti-science considered as human, now he shows that this is expounded also in the Veda’.15 Again, Kaiya™a inserts completely these examples in the well-known couple †ruti/smr¢ti and identifies laukika-examples with smr¢ti, and vaidika ones with †ruti.16 Let us notice that this sets a purely textual partition of domain, where it is not that easy to define the specificity of the †åstric domain as opposed to smr¢ti and †ruti.17 The second occurrence of yathå laukikavaidikeßu is found in a passage where Pata∞jali is discussing the concept of imitation (M v. I, p. 20, l. 16-p. 21, l. 2 ad ‡ivasütra 2 vt. 3). At first, Pata∞jali interprets imitation as mere repetition: having heard someone else uttering a word, 13 The statement is traced by Filliozat to Aitareyabråhmaña, 6.1, v. I, p. 140 (Filliozat 1975: 102) and by Joshi and Roodbergen to Aitareyabråhmaña, 2.1 (Joshi, Roodbergen 1986: 124). 14 The statement is traced by Filliozat to Våjasanehisa∫hitå, 1.118 (Filliozat 1975: 102) and by Joshi and Roodbergen to Taittirœyasa∫hitå, 1.1.17 (Joshi, Roodbergen 1986: 124). 15 D1, p. 26, l. 21-2, ad vt. 1: (eva∫ tåvat paurußeyåbhimata∫ smr¢ti†åstra∫ sm®tvopanibaddha∫ vede tad api pradar†ayati). 16 P v. I, p. 34, ad vt. 1: (laukikaΔ smr¢tyupanibaddhaΔ Ù vaidikaΔ †rutyupanibaddhaΔ). 17 Similar to this passage there is another one (M v. III, p. 57, l. 14-21 ad A 6.1.84 vt. 5) where again the discussion concerns the nature of grammatical rules. Precisely, Pata∞jali affirms that, grammatical rules being a dharmic teaching, they must apply in the totality of cases and are not fulfilled by applying only to some. The laukika-example is that of the rules ‘a Brahmin must not be killed’ (bråhmaño na hantavyaΔ), ‘one must not drink liquors’ (surå na peyå) and ‘a Brahmin must be saluted by a younger Brahmin getting to his feet’ (pürvava`yå bråhmañaΔ pratyuttheyaΔ), too generic to be reliably traced down.

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 397

a man may decide to utter it again; if the original word is correct, its repetition is correct too, if the original word is not correct, its repetition isn’t either; the same way it happens in well-established practices in the world and in the Veda: Loka

‘A man who, [seeing another man] offering a gift, a sacrifice or reciting something, imitates him and offers a gift, a sacrifice or recites something, that man also will obtain prosperity’ (ya evam asau dadåti ya evam asau yajate ya evam asåv adhœta iti tasyånukurvan dadyåc ca yajeta cådhœyœta ca so ’py abhyudayena yujyate). ‘A man who, [seeing another man] hiccoughing, laughing or scratching himself, imitates him and hiccoughs, laughs or scratches himself, for this man there will be no harm nor prosperity’ (ya evam asau hikketi ya evam asau hasati ya evam asau kañ∂üyatœti tasyånukurvan hikkec ca hasec ca kañ∂üyec ca naiva taddoßåya syån nåbhyudåya). ‘But if a man, [seeing another man] killing a Brahmin or drinking liquor, imitates him and kills a Brahmin or drinks liquor, that man too, in my opinion, would be a sinner’ (yas tu khalv evam asau bråhmaña∫ hanty evam asau surå∫ pibatœti tasyånukurvan bråhmaña∫ hanyåt surå∫ vå pibet so ’pi manye patitaΔ syåt). Veda ‘A man who, [seeing] that the Vi†vas®j set up the great sacrificial sessions, imitates them and set up some great sacrificial sessions, this man also will obtain prosperity’ (ya eva∫ vi†vas®jaΔ satråñy adhyåsata iti teßå∫ anukurva∫s tadvat satråñy adhyåsœta so ’py abhyudayena yujyate). ‡åstra ‘Here too a man who, [seeing another man] using a corrupt word, imitates him and uses a corrupt word, that man also will make use of corrupt speech’ (evam ihåpi ya evam asåv apa†abda∫ prayuõkta iti tasyånukurvann apa†abda∫ prayu∞jœta so ’py apa†abdabhåk syåt). It is quite evident from these examples that the loka-domain can’t refer here to a corpus of mundane texts or rules. The reference is simply to the practice of imitating/repeating someone else’s actions and to the implicit assumption in the word that both actions, the original and the imitation, get the same reward. Still it is somewhat bewildering that Pata∞jali puts among the laukika-examples not only some dharmic actions like that of offering gifts, but even the action of accomplishing a sacrifice! Shouldn’t that be a Vedic example? In fact, what seems to be crucial for Pata∞jali’s classification here, is the source of the action to be imitated: if a man sees someone else performing a sacrifice and imitates what he has seen, this is a laukika-example. On the other hand if, while hearing or reading a Vedic text, one happens to come upon a description of sacrificial sessions not performed any-

398

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

more,18 and decides to imitate them, this is a Vedic example. In this sense one must be cautious in interpreting Deshpande’s claim that there was an actual, present dimension of Vedic experience: this possibility of actualizing Vedic experience involved strictly the domain of recitation and transmission of the Vedic texts and not, in general, of all the ritual activity. There seems to be a part of the ritual experience that cannot be labeled as Vedic. Bhartr¢hari’s comment at this passage is unfortunately very corrupt, still one statement is reasonably certain and seems to give support to our interpretation. Commenting on the laukika-examples Bhartr¢hari observes ‘although the topic is Vedic, the words being laukika he (i.e. Pata∞jali) says that [the example is found] in the word. On the contrary the statement vi†vas®jaΔ satråñy åsata may be read also in the Veda’ (yady apy artho vaidikas tathåpi †abdånå∫ laukikatvåt loka ity ucyate Ù ida∫ tu vi†vas®jaΔ19 satråñy åsata iti vede ca [pa]™hyate [D2, p. 14, l. 4-5, ad ‡ivaütra 2 vt. 3]). This position may be fruitfully contrasted with that of Kaiya™a who, on the other hand, proceeds with his identification of the loka-domain with the domain of smr¢ti texts: ‘[Offering] a gift and so on, because of its being taught in the smr¢ti, is considered as a laukika [-example]’ (dånådœnå∫ tu smårtatvål laukikatvam ucyate [P v. I, p. 75, ad ‡ivasütra 2 vt. 3]). The third passage, which concerns linguistic practices of people, is particularly interesting (M v. I, p. 38, l. 15-21, ad A 1.1.1 vt. 4). Commenting the first rule of grammar A 1.1.1 (vr¢ddhir ådaic) –which teaches the name vr¢ddhi for the sounds å, ai, au–, Pata∞jali asserts that it is possible to identify vr¢ddhi as a technical name (sa∫j∞å) and å, ai, au as the objects named by it thanks to the practice of the teachers (åcåryåcåråt), the same way it happens in well-established practices in the world and in the Veda: Loka

Veda

‘In the world, parents, in a secret place, give a name to the son they have given birth to: Devadatta, Yaj∞adatta. From their subsequent practice, other people also know that this is his name’ (loke tåvan måtåpitarau putrasya jåtasya sa∫vr¢te ’vakå†e nåma kurvåte devadatto yaj∞adatta iti Ù tayor upacåråd anye ’pi jånantœyam asya sa∫j∞eti). ‘In the Veda the yåj∞ikas give names like sphya, yüpa and caßåla. From the subsequent practice of these revered [teachers] other people also know that this is its name’ (vede yåj∞ikåΔ sa∫j∞å∫ kurvanti sphyo yüpa† caßåla iti Ù tatra bhavatåm upacåråd anye ’pi jånantœyam asya sa∫j∞eti).

18 The thousand-years sacrificial sessions of the Vi†vasr¢j are accounted for principally in bråhmaña literature (see Mylius 1984). 19 Following here Abhyankar and Limaye’s edition (see Abhyankar, Limaye 1970: p. 62, l. 8). In Palsule’s edition (D2, p. 14, l. 4, ad ‡ivasütra 2 vt. 3) we find vi†vajitaΔ.

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 399

‡åstra ‘Here too it is the same. Right here (i.e. commenting A 1.1.1) some commentators say ‘the word vr¢ddhi is the sa∫j∞å, å, ai and au are the sa∫j∞in’. Other commentators, later on, [commenting] A 7.2.1 sici vr¢ddhi, quote the sounds å, ai and au’ (evam ihåpi Ù ihaiva tåvat kecid vyåcakßåñå åhuΔ Ù vr¢ddhi†abdaΔ sa∫j∞ådaicaΔ sa∫j∞ina iti Ù apare punaΔ sici vr¢ddhiΔ ity uktvåkåraikåraukårån udåharanti). The matter discussed in this passage is how it is possible to manipulate language and to transmit this modification, at least in some restricted domains. In the mundane context Pata∞jali provides the example of parents who give the name to their son in a secret place. It may be that nobody was present at the ceremony, nonetheless parents are authoritative as regards the name of their son. Hearing them, afterwards, call their son Yaj∞adatta or Devadatta, other people, even if they were not present at the name-giving ceremony, know the name of the boy. Of course there is no textual dimension in this creation of a tradition, nevertheless its establishment is not less effective. Unfortunately, the commentary of Bhartr¢hari on this passage is missing and Kaiya™a is silent. Let us now turn to the Vedic example. There, the yåj∞ikas ‘create names’ like sphya, yüpa and caßåla and, from their subsequent usage, other people know these names and their meanings. But who are the yåj∞ikas and in which sense may it be said that they create names (sa∫j∞å∫ kurvanti)? Någe†a affirms that by the term yåj∞ika here reference is made to the ancient seers (r¢ßi) who saw the sacrificial portion (yaj∞akåñ∂a) of the Vedic texts. The expression sa∫j∞å∫ kurvanti must therefore in his view be interpreted in the same way as some well-known statements on the r¢ßis being the authors of the Veda20: it is not strictly speaking creation of the text but rather direct supernatural vision of the text. Still, this seems to be quite a forced interpretation, prompted by the desire to give to the Veda a purely textual dimension. Let us then take up again the problem concerning the identity of these yåj∞ikas. The term is generally interpreted as conveying two different meanings, that of expert, theoretician of the sacrifice and that of sacrificer. This second meaning, nonetheless is a later one, and lexicons justify it only through a parallel yåj∞ika = yåjaka. The term is attested since ancient times. In ‡PaBr.(M) 14.6.7.1-4 (see also Br¢hadåranyakopanißad 3.6.7.1-4) we read the history of Uddålaka who, with other pupils, went to the sage Pata∞cala Kåpya to study the science of ritual (yaj∞am adhœyånåΔ). One day a gandharva passes by and stops to discuss with them. While discussing 20 U v. I, p. 126, ad A 1.1.1 vt. 4: (vede yåj∞ikå iti Ù yaj∞akåñ∂adraß™åra r¢ßaya ity arthaΔ Ù vedakartr¢tvavat sa∫j∞åkartr¢tva∫ teså∫ draß™avyam).

400

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

with them, he addresses them as yåj∞ikas, ‘scholars, experts of ritual’. The most interesting occurrences, nonetheless, are found in Yåska’s Nirukta; in this text yåj∞ikas are often opposed to nairuktas and vaiyåkarañas in the interpretation of Vedic texts. For example Nir. 5.11 comments on R¢V 8.77.4 (ekayå pratidhå pibat såka∫ sarå∫si tri∫†atam Ù indraΔ somasya kåñukå) and quotes different interpretations of the act of drinking thirty lakes of Soma: for the yåj∞ikas this should represent the thirty pots used during the pressuring of Soma, whereas the nairuktas consider that reference is made to the thirty elements of the couple of night and day in the first half of the lunar month.21 Yåska (Nir. 7.23) is even aware of a historic development inside this school and makes reference to the opinion of the pürve yåj∞ikåΔ. In the Nirukta it seems therefore quite safe to assume that the term yåj∞ika refers to disciples of a traditional school of interpretation of the Veda. But we have reasons to think that this was also the meaning intended by Pata∞jali himself. In the Aß™ådhyåyœ yåj∞ika is formed through A 4.2.60 ‘the suffix ™haK (= -ika) after [nominal bases] signifying sacrifice, after the group of nominal bases beginning with uktha, and the bases ending with sütra [in the sense of ‘he studies this, he knows this’]’ (kratükthådisütråntå™ ™hak).22 A yåj∞ika is, therefore, someone who studies or knows the sacrifice. Pata∞jali’s comment on this sütra adds an interesting element. He wonders if this same suffix ™haK may be applied also to the nominal bases yåj∞ikya and aukthikya (derived from the preceding ones in the sense of ‘it is one’s [tradition/doctrine]’23) in order to get the meaning of ‘one who knows the doctrines of the yåj∞ikas/of the aukthikas’. Still, in Pata∞jali’s opinion, this is not necessary: he proposes to give to the extant forms yåj∞ika and aukthika both the meanings ‘who studies the sacrifice’ and ‘who studies the doctrines of the yåj∞ikas’ or ‘who studies the ukthas’ and ‘who studies the doctrine of the aukthikas’. This is not easy to obtain, at least as far as strict påñinian derivation is at stake,24 but 21 Nir. 5.11: (tatraitad yåj∞ikå vedayante Ù tri∫†ad ukthapåtråñi mådhyandine savana ekadevatåni Ù tåny etasmin kåla ekena pratidhånena pibanti Ù tåny atra sarå∫sy ucyante Ù tri∫†ad aparapakßasyåhoråtråΔ Ù tri∫†at pürvapakßasyeti nairuktåh). Similar passages may be found in Nir. 7.4, where we find the yåj∞ikas opposed to the nairuktas in the debate concerning to whom one should ascribe hymns having no explicit mention of the deity praised, and Nir. 11.29 where they discuss the interpretation of the proper names Anumati et Råkå. Nir. 13.9 quotes the yåj∞ikas altogether with vaiyåkarañas, nairuktas and ‘others’ (eke) when expounding the different interpretations of the well-known hymn R¢V 1.164.45. 22 A 4.2.59 (tad adhœte tad veda). 23 The two secondary bases aukthikya- and yåj∞ikya- are formed from the nominal bases aukthika- and yåj∞ika- through A 4.3.129 (chandogaukthikayåj∞ikabahvr¢cana™å∞ ∞yaΔ). These formations are taught in a generic relational meaning by A 4.3.10 (tasyedam). Commentators generally agree that, in the present case, the meanings of dharma ‘law’ or of åmnåya ‘tradition, doctrine’ are involved. Cåndogya will therefore be the law or the doctrine of the cåndogas and yåj∞ikya the law or the doctrine of the yåj∞ikas. 24 It may be obtained through application of A 4.2.64 (proktål luk) that teaches substitution with phonic zero of a suffix taught in the sense tad adhœte tad veda, if it follows a

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 401

Pata∞jali seems to consider crucial that words like yåj∞ikas and aukthikas do not mean only those who study the sacrifice or the ukthas but also those who study the doctrine of the corresponding schools. If we come back now to the passage exemplifying yathå laukikavaidikeßu we started from, we can assume that the Vedic domain in that example includes not only the Vedic texts but also the whole tradition of their interpretation. It could thus seem easier to understand in which sense these yåj∞ikas, being members of an exegetical school, create names (sa∫j∞å∫ kr¢-): like other scholastic traditions the yåj∞ikas may have their own technical names for some objects or procedures valid within the restricted domain of their own school. The ‘technical names’ quoted are nevertheless a bit bewildering: sphya, yüpa, caßåla are all Vedic words found in the Sa∫hitås25 and it would be quite awkward to affirm that they have been ‘created’ by the yåj∞ikas. I have no definite answer for this difficulty. It may be that sphya, yüpa, caßåla have a non-technical value in the Sa∫hitås and that Pata∞jali therefore considers that yåj∞ikas have created their specialized meaning26. Or it may be that this difficulty is principally due to a kind of modern misunderstanding of the action of sa∫j∞å∫ kr¢- which, in the Bråhmañas, is principally linked to the act of establishing a convention, a link, between two entities, identifying one with the other. And in the Bråhmañas this identification is carried out for the above quoted ritual objects too: the sphya or sacrificial knife, for example, is always identified with Indra’s vajra.27 The sense of the Vedic example would therefore be that, the ritual trasuffix having the meaning ‘taught by him’ (see A 4.3.101: [tena proktam] . The word påñinœya, for example, denotes the Aß™ådhyåyœ (being the text taught by Påñini) and a person studying the Aß™ådhyåyœ. Still, applying rigorously the same rules to the case of yaj∞a/yåj∞ikya one should obtain *yåj∞ikya in the sense of ‘person studying the tradition of the yåj∞ikas’ and not, as it is desired, yåj∞ika. Patañjali is therefore obliged to propose either to apply zero substitution also of the suffix -ya (but this would involve an heavy modification of the påñinian text) or to presuppose the existence of a couple yaj∞a/yåj∞ikya without difference of meaning, following the example of avi/avika (A 5.4.28). 25 Concerning sphya see AV(‡) 11.3.9 and (Pai) 16.53.15; concerning yüpa see R¢V 1.51.14; 4.33.3; 5.2.7 and for caßåla see R¢V 1.162.6. 26The word vr¢ddhi too, for example, pre-exists Påñini’s usage, but its technical meaning is taught by Påñini. 27 The ‘sacrificial knife’ (sphya) is always related to Indra’s thunderbolt (vajro vai sphyaΔ); the ‘sacrificial post’ (yüpa), is at times related to that same thunderbolt in the split form it got when thrown by Indra against Vr¢tra (‡PaBr (M) 1.2.41 and ‡PaBr (K) 2.2.2.1). The ‘upper part of the sacrificial post’ (caßåla) is often related to the sacrificial post itself in traditional associations. See, for example, ‡PaBr (M) 5.4.4.15 which describes the sacrificer who, during the rite of coronation (råjasüya), addresses directly the sacrificial knife (sphya): ‘then the Brahmin gives him the sacrificial knife, the adhvaryu or the domestic priest of him [who offers the sacrifice says] ‘You are Indra’s thunderbolt, through this do came in my possession’, because the knife is the thunderbolt. Through this thunderbolt the Brahmin makes the king weaker than himself’ (athåsmai bråhmañaΔ sphya∫ prayachati Ù adhvaryur vå yo våsya purohito bhavatœndrasya vajro ’si tena me radhyeti vajro vai sphyaΔ Ù sa etena vajreña bråmaño råjånam åtmano ’balœyå∫sa∫ kurute). The identification seems less perspicuous as far as the sacrificial post is concerned : ‘Then he sprinkles it (i.e. the post) with the spoon [and says]: ‘Oh, you who are made for Vißñu’s sake’, because the post is

402

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

dition has established some links between Vedic ritual objects and other entities. During the performance of rites these objects are used to evoke those entities, and through this practice people understand what these ritual objects stand for. The †åstric example supports the previous interpretation of the Vedic one. The question is how one is supposed to know –when reading the rule A 1.1.1 (v®ddhir ådaic)– a) that it is a name-giving rule; b) which, between vr¢ddhi and ådaic (the set of sounds å, ai, au), is the technical name (sa∫j∞å) and which is the object named (sa∫j∞in). One of the tentative answers given by Pata∞jali is that this may be inferred by subsequent practice of commentators (vyåcakßåña). Some of them have established that in the rule ‘vr¢ddhir ådaic’ the word vr¢ddhi is the technical name and the word ådaic signifies the object named. Following this assertion, other commentators when dealing with påñinian rules like A 7.2.1 (sici vr¢ddhi) quote the sounds å, ai and au.28 From this kind of commentary –Pata∞jali says– we understand that words which make something else be known are the names (sa∫j∞å) and words which are made known are the named (sa∫j∞in). From this last remark we understand that the first step (identification of the sa∫j∞å and of the sa∫j∞in) is supposed not to be available anymore to Pata∞jali’s public. Pata∞jali has no authoritative statement to quote, to prove that, in A 1.1.1, v®ddhi is the sa∫j∞å and ådaic signifies the sa∫j∞in. Thus he proposes to infer it from the common practice of commentators who, when commenting other rules containing the word v®ddhi, exemplify these rules with occurrences of the sounds å, ai, and au. In all the three examples of Pata∞jali’s commentary on this last occurrence of yathå laukikavaidikeßu three different steps are involved: first the parents give a name to their child in a secret place, then they use it afterwards and from this subsequent usage (upacåra) other people understand which is the name of the child. The first step is hidden, it is inferred from further practice. Similarly, in the †åstric example, even if the definition itself of the name vr¢ddhi has been preserved (through A 1.1.1), its interpretation (which is the sa∫j∞å and which the sa∫j∞in) is not. Still, later commentators’ practice keeps memory of this assertion by other means. This way of reasoning reminds us strongly of Pollock’s interpretation of †ruti and smr¢ti in Sanskrit tradition (Pollock 1997). Pollock wonders what exactly is the content of the remembrance which characterises the smr¢ti and opposes it to the actual auditory perception of the †ruti. Shall not †ruti-texts what is made for Vißñu’s sake and the sacrifice is Vißñu. For the sacrifice he cuts down [the tree], this is the reason why he says ‘Oh, you who are made for Vißñu’s sake’’ (atha sruveñopaspr¢†ati Ù vißñave tveti vaißñavo hi yüpo yaj∞o vai vißñur yaj∞åya hy ena∫ vr¢†cati tasmåd åha vißñave tveti) (‡PaBr (M) 3.6.4.9). 28 Kaiya™a does not consider relevant the mention ‘afterwards’ (punar) and interprets the two steps (identification of the sa∫j∞å and the sa∫j∞in in the sa∫j∞åsütra itself and mention of the sa∫j∞ins in operative rules where the sa∫j∞å occurs) as alternative ways of getting the desired result, each being independently adequate.

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 403

themselves be called smr¢ti insofar as they are handed down by an unbroken tradition? Paraphrasing a passage from ‡abara, the author states that the difference lays in the source of the transmitted knowledge; as far as †ruti is concerned “our ‘memory’ of the Vedas themselves is not validated merely by its ‘unbroken tradition’ but by the fact that Vedas are actually perceptible to us. It is this actual perception of Vedic texts, –their existing during recitation– that constitutes the ‘prior cognitive experience’ necessary to substantiate the memory of them” (Pollock 1997: 404). The smr¢ti, on the other hand, lacks of this actual experience of the authoritative source upon which it is nevertheless based: Insofar as the same people who perform the acts of dharma required by the Veda also perform acts of dharma ‘not based on sacred word’, we must assume that the authority for these actions is conferred, not by directly perceptible Vedic texts, but by texts inferentially proven to exist. [...] It is not unreasonable to hold that the knowledge of these texts is remembered, while the texts themselves (i.e. their actual wording) have been lost (Pollock 1997: 404).29

The same cultural pattern seems to be at work in our passages, but here it is applied to all the three domains, loka, veda and †åstra: in all these domains the source of experience is crucial to define the domain itself (let us remember the imitation of a sacrifice classified as a laukika-example), and in all the three domains there seems to be room for a smr¢ti dimension, i.e. for an authoritative knowledge lacking an explicit textual source. Besides, the passage, while dealing with authoritative instauration of new meanings for words, adds an interesting dimension to grammarian’s conception of scholastic/restricted language too. It is generally attributed to the grammarians (and other schools dealing with the concept of language) a distinction between technical restricted language and natural language on the basis of the fact that the first has been modified by human intervention, and that memory of this modification is necessary to interpret it correctly (Scharfe 1971: 1). It is possible to interpret vr¢ddhi as meaning å, ai and au only through a recollection of the definition-rule A 1.1.1 which establishes it as a påñinian technical term. The passage we have just read, nevertheless, adds some new interesting dimensions to this picture. It shows, for example, that this memory may not take the form of an explicit statement handed down by tradition. It may also take the form of authoritative practices which point to such a statement without mentioning it. There is no rule in Påñini’s Aß™ådhyåyœ teaching which —in A 1.1.1 (vr¢ddhir ådaic)— is the sa∫j∞å and which is the sa∫j∞in; still this may be inferred from later commentators. By the way, commentators 29 The author further suggests that the couple †ruti/smr¢ti should be read alongside the opposition between pratyakßa and anumåna as sources of valid knowledge.

404

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

were also fully aware of the fact that Påñini, for example, used technical names (i.e. words whose meaning could not be derived from natural language) without defining them and without ‘quoting their creators’, because they were ‘technical names of the ancient teachers’ (pürvåcaryasa∫j∞å). The meaning of these terms, though not a natural meaning, is nevertheless not apprehended in a fully artificial way, i.e. by means of a definition, but it may be apprehended simply observing the practice of later grammarians. In this sense this process of apprehension is not radically different from the natural process of learning Sanskrit from the practice of the †iß™as. The difference between the apprehension of natural words and the apprehension of technical terms is thus less sharp than it may seem at a first sight.30 3. Textual and extra-textual dimension of tradition From what we have seen it seems quite evident that Pata∞jali identified three domains of human activities (two of them explicitly and the third implicitly, i.e. the domain of †åstra or grammar), and that these domains were considered by him as synchronically present in everyday experience of his public. These domains were differentiated following the source (ådhåra) of the experience involved. This is quite clear in the passage concerning imitation: imitating a sacrifice you may happen to attend to in the world is a laukika-example of imitation, while imitating the years-lasting sacrificial sessions described in the Veda is a Vedic example of imitation. The Veda may therefore be an independent source of experience in the same way Aß™ådhyåyœ may be another independent source of experience for grammarians. Inside these different domains of activities language may be (at least in part) modified and its modifications may be transmitted to following generations creating therefore languages valid within restricted domains. It would be impossible to interpret these domains on purely textual grounds, even if this is not to deny that the domain of the world, of the †åstra and of the Veda have also a textual dimension. Still, verbal testimony of texts is not the unique element at play: usage of experts in the different domains may be a hint to some authoritative statements not traceable any longer. The meaning of some technical names, for example, does not depend exclusively on the text in which they are 30 One last passage from the Bhåßya concerning the domain of linguistic behaviours is unfortunately less informative for us. It is M v. I, p. 137, l. 7 and ff, vt. 13 ad A 1.1.56 where Pata∞jali discusses the usage of the expression sthåne + genitive, in the sense of ‘in the place of’. The question is whether this expression carries the presupposition that the element ‘in the place of which’ another element is used must have been there previously. Pata∞jali’s answer is no, on the grounds that the expression may be used also in the case of an element that was not previously there as it is seen in mundane and Vedic examples. The laukika-example is that of the statement ‘the pupil instead of the teacher’ (upådhyåyasya sthåne †ißyaΔ) and the Vedic one is ‘instead of Soma one should press pütœka-grass’ (somasya sthåne pütœkatr¢ñåny abhißuñuyåt). Joshi and Roodbergen trace this quotation to Tåñ∂yabråhmaña, 9.5.3 (Joshi, Roodbergen 1990: 78).

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 405

uttered but also on the tradition to which these texts refer, and on the linguistic usages in force within that tradition. We have seen that even the meaning of the world vr¢ddhi cannot be completely ascertained through the sole Aß™ådhyåyœ. And it is not simply a matter of relationship between a text and its commentary; it covers all the linguistic production of a certain school: text, commentaries and usage of the †iß™as. Some passages from Bhart®hari are quite significant on this respect. One of these passages concerns Pata∞jali’s long discussion (M v. I, p. 6, l. 17-22, ad vt. 1) on the meaning of the word siddha in vt. 1 of Kåtyåyana (siddhe †abdårthasa∫bandhe): should siddha be interpreted as a synonym of nitya, thus meaning ‘eternally established’ or should it be interpreted in the meaning ‘ready, established, (already) prepared’ we find in utterances like ‘rice is ready’ (siddha odanaΔ )? As a first, tentative, answer to the question, Pata∞jali refers to the usage of this term in another text of the grammatical tradition, the Sa∫graha of Vyå∂i. In one passage of the Sa∫graha, Pata∞jali says, siddha is clearly opposed to kårya, ‘to be done, to be produced’: it shall therefore be interpreted as a synonym of nitya, both in that Sa∫graha passage and in the vårttika. of Kåtyåyana under discussion. Pata∞jali is satisfied with this answer, but Bhartr¢hari wonders on what grounds the meaning of the word siddha, obtained by analysis of the textual context31 in the Sa∫graha, may influence the interpretation of the word siddha in an altogether different textual context. Still, answers Bhartr¢hari, the Sa∫graha is also a part of the same †åstra and its author, Vyå∂i, is an authoritative person; the meaning he has attributed to the word siddha must therefore be considered valid in Kåtyåyana’s vårttika too.32 It is quite clear that there can be no question here of a purely textual conception of the domain: knowing the meaning of a certain word in a given utterance, let us say, to follow Bhartr¢hari’s examples, the meaning of the name Råma in råmårjunau33 does not allow any assumption whatsoever about the meaning of that same name in other points of the text and, even less, in other texts. But, in the case of the word siddha, something completely different is at stake; the text opposing siddha to kårya is nothing more than a hint pointing to something else, namely to the establishment of a specific meaning for a certain word in a restricted domain of activities (here in grammatical tradition) by some authoritative person. 31 See D1, p. 19, l. 23-28, ad vt. 1. The passage it is not easily interpreted in all its details, but from all the examples Bhartr¢hari quotes it is quite evident that he interprets the Pata∞jali’s context on pure textual grounds of association of words; e.g. an utterance like råmårjunau refers without doubt to Råma son of Da†aratha while råmake†avau refers to Baladeva. 32 D,1 p. 19, l. 28-p. 20, l. 2, ad vt. 1: ‘And here too (i.e. in the Bhåßya) it is the same. Why? The Sa∫graha too is a portion of this same science, thus, because of the condition of being one unique text and because of the authoritativeness of Vyå∂i, the world siddha is understood in the same way here too’ (ihåpi tad eva Ù kutaΔ sa∫graho ’py asyaiva †åstrasyaikade†aΔ tatraikatantratvåd vyå∂e† ca pråmåñyåd ihåpi tathaiva siddha†abda upåttaΔ). 33 See note 31.

406

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

The reasoning may also go the other way round. In another passage of the Dœpikå, Bhartr¢hari –following Pata∞jali on Kåtyåyana’s vt. 19 ad A 1.1.44– discusses the necessity for Påñini to teach explicitly the meaning of the term vibhåßå in the sense of ‘option’,34 as that meaning is already well-known in everyday usage and in some other technical traditions as well. The yåj∞ikas, when they employ the word vibhåßå, mean option, though they do not use it as a technical term.35 Pata∞jali’s comment stops here, but Bhartr¢hari’s argument goes further. Everyday usage, he argues, might be derived from grammatical usage (and this would justify the necessity of teaching the grammatical meaning of the word vibhåßå), but yåj∞ikas’ technical language cannot: as the yåj∞ikas belong to a different system (tantra), they would not use a technical term coming from another science (†åstra). Therefore, there exists a usage of vibhåßå to indicate option independent from grammatical tradition.36 Everyday usage may borrow some terms from restricted domains, but one restricted domain generally do not borrow from another because the terms each system uses depend on that same system. In both passages it is quite clear that Bhartr¢hari considers that the textual context is not sufficient to determine the meaning of some technical names.37 The meaning of these names is also dependent upon the tradition they belong to, a tradition which is of course built up with authoritative statements handed down by memory (e.g. Påñini’s definitions of technical terms) but also with practices which hint at likewise authoritative statements no longer available. Such a complex construction of the notion of tradition is not surprising in an author like Bhartr¢hari who has devoted much attention to the identification of sources of authority in knowledge. Aklujkar, while commenting on a term akin to those we have dealt with here, namely the term ågama (currently translated as ‘tradition’), points out the fact that this term is too often wrongly restricted to the mere †ruti (Aklujkar 1989). On the contrary, he maintains that the concept of ågama in Bhartr¢hari is a very wide one and “includes inherited knowledge of all kinds, from the biologically inherited knowledge or instincts of a species to the highly specific intentionally cultivated knowledge of a school. [...] Further Bhartr¢hari does not confine ågama to a written or spoken form; it can also exist and can be inferred from the practice of the †iß™as or the elite” (Aklujkar 1989: 18. See also Aklujkar 1991). 34

A 1.1.44 (na veti vibhåßå). M v. I, p. 105, l. 16-18, ad A 1.1.44 vt. 19 (ye ’pi hy etå∫ sa∫j∞åm anårabhamåñå vibhåßety ukte ’nityatvam avagacchanti Ù yåj∞ikåΔ khalv api sa∫j∞åm anårabhamåñå vibhåßety ukte ’nityatvam avagacchanti). 36 D6/2 p. 36, l. 20-1, ad A 1.1.44 vt. 19 (atha vå laukikåΔ †åstråd ågamya pratipadyeran Ù yåj∞ikås tu bhinnatantratvåt tasmin †åstre †åstråntarasa∫j∞ayå na vyavahareyur iti). 37 An opposition between a textual and an extra-textual context may be seen in the couple artha/prakåraña we find in traditional lists of factors determining the meaning of words. A detailed account for these lists would probably prove most fruitful but, as far as I know, is yet to be done. 35

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 407

The progressive tendency towards a much more text-centred interpretation of tradition by later commentators (let us remind here Kaiya™a equivalence laukika kr¢tånta = smr¢ti and Någe†a statement that the yåj∞ikas who create names are in fact the Vedic r¢ßis) shall not conceal the greater complexity of the picture in earlier times.

ABBREVIATIONS AND BIBLIOGRAPHY A

AV(Pai) AV(‡)

D1 D2 D6/2 K

M

Nir.

Aß™ådhyåyœ, edited by O. Böhtlingk, Påñini’s Grammatik. Herausgegeben, bersetzt, erläutert und mit verschiedenen Indices versehen von Otto Böhtlingk, Motilal Banarsidass, Delhi 1998 [1887]. Atharvaveda (Paippalåda recension), edited by Ragu Vira, Atharvaveda of the Paippaladas, International Academy of Indian Culture, Nagpur 1936. Atharvaveda (‡aunaka recension), edited by Vishva Bandhu, Atharvaveda: ‡aunaka. With the Pada-på™ha and Såyañåcårya’s Commentary; edited and annotated with text-comparative data from original manuscripts and other Vedic works by Vishva Bandhu et al., Vishveshvaran and Vedic Research Institute, Hoshiarpur 1960-1964. Mahåbhåßyadœpikå, edited by J. Bronkhorst, Mahåbhåßyadœpikå of Bhartr¢hari. Fascicule IV: Åhnika I, Bhandarkar Oriental Research Institute, Poona 1987. Mahåbhåßyadœpikå, edited by G. B. Palsule, Mahåbhåßyadœpikå of Bhartr¢hari. Fascicule V: Åhnika II, Bhandarkar Oriental Research Institute, Poona 1988. Mahåbhåßyadœpikå, edited by V.B. Bhagavat, S. Bhate, Mahåbhåßyadœpikå of Bhartr¢hari.Fascicule VII: Åhnika VI part I, Bhandarkar Oriental Research Institute, Poona 1990. Kå†ikå, edited by A. Sharma, K. Deshpande, D.G. Padhye, Kå†ikå, a Commentary on Påñini’s Grammar by Våmana and Jayåditya (†rœvåmanajayådityaviracitå påñinœyåß™ådhyåyœsütravr¢ttiΔ kå†ikå), Sanskrit Academy, Osmania University, Hyderabad 1969-1970, 2 voll. Mahåbhåßya, edited by F. Kielhorn, The Vyåkaraña-mahåbhåßya of Pata∞jali , Governement Central Press, Bombay 1880-1885, 3 voll. (quoted following K.V. Abhyankar (ed.), The Vyåkaraña-mahåbhåßya of Pata∞jali, edited by F. Kielhorn. Third edition, revised and furnished with additional readings, references and select critical notes by K.V. Abhyankar, BORI, Pune 1962-1972, 3 voll.). Nirukta, edited by L. Sarup, The Nighañ™u and the Nirukta, the Oldest Indian Treatise on Etymology, Philology, and Semantics, Motilal Banarsidass, Delhi 1967 [1920-1927].

408

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Pradœpa, edited by Vedavrata, ‡ribhagavatpata∞jalivicarita∫ vyåkaraña-mahåbhåßyam. ‡rœkaiya™a-kr¢tapradœpena nagojœbha™™akr¢tena bhåßyapradœpoddyotena ca vibhüßitam, Harayåñå Såhitya Sa∫sthåna, Gurukula Jhajjar (Rohatak) 19621963, 5 voll. R¢V R¢gveda, edited by T. Aufrecht, Die Hymnen des Rigveda herausgegeben von Theodor Aufrecht, O. Harrassowitz, Wiesbaden 1986 [1887]. ‡PaBr (K) ‡atapathabråhmaña (Kåñvœya recension), edited by W. Caland, Raghu Vira, The ‡atapatha Bråhmaña in the Kåñvœya Recension. Revised by Raghu Vira, Motilal Banarsidass, Delhi 1998 [1926]. ‡paBr (M) ‡atapathabråhmaña (Mådhyandina recension), edited by A. Weber, ‡atapatha-Bråhmaña in the Mådhyandina-†åkhå with the Extracts from the Commentaries of Såyaña, Harisvåmin and Dvivedagaõga, Chowkhamba, Varanasi 1964 [1849-1855]. U Uddyota, edited by Vedavrata, ‡ribhagavatpata∞jalivicarita∫ vyåkaraña-mahåbhåßyam. ‡rœkaiya™a-kr¢tapradœpena nagojœbha™™akr¢tena bhåßyapradœpoddyotena ca vibhüßitam, Harayåñå Såhitya Sa∫sthåna, Gurukula Jhajjar (Rohatak), 1962-1963, 5 voll.

P

Abhyankar, K.V., Limaye, V.P. (eds.) (1970), Mahåbhåßyadœpikå of Bhartr¢hari, Bhandarkar Oriental Research Institute, Poona. Aklujkar, A. (1989), Pråmåñya in the Philosophy of the Grammarians, in Kumar, A., Le Mée, J., Agrawal, M.M., Narang, S.P, Upadhyaya, V. (eds.), Studies in Indology, Prof. Rasik Vihari Joshi Felicitation Volume, Shree Publishing House, New Delhi, pp. 15-28. ______ (1991), Bhartr¢hari’s Concept of the Veda, in Deshpande, M.M. (ed.), Påñini and the Veda, E.J. Brill, Leiden-New YorkKöln, pp. 1-18. Cardona, G. (1990), On Attitudes towards Language in Ancient India, in «Sino-Platonic Papers», 15, pp. 1-19. Birwé, R. (1961), Der Gañapå™ha zu den Adhyåya IV und V der Grammatik Påñinis: Versuch einer Rekonstruktion, Harrassowitz, Wiesbaden. Deshpande, M.M. (1985), Historical Change and the Theology of Eternal Sanskrit, in «Zeitschrift fr Vergleichende Sprachforschung», 98, pp. 122-149. ______ (1993), The Changing Notion of †iß™a from Pata∞jali to Bhartr¢hari, in Bhate, S., Bronkhorst, J. (eds.), Proceedings of the First International Conference on Bhart®hari, University of Poona, January 6-8 1992, Peter Lang, Bern, pp. 95-115. Filliozat, P.S. (ed.) (1975), Le Mahåbhåßya de Pata∞jali avec le Pradœpa de Kaiya∂a et l’Uddyota de Någe†a. Adhyåya 1 Påda 1 Åhnika 14, Institut Français d’Indologie, Pondichéry.

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 409

Joshi, S.D., Roodbergen, J.A.F. (ed.) (1986), Pata∞jali’s Vyåkarañamahåbhåßya: Paspa†åhnika; Introduction, Text, Translation and Notes (= PCASS-C 15), University of Poona, Pune. Mylius, K. (1984), Vi†vasr¢j, Vi†vasr¢jaΔ and the Problem of Continuity in Indian Religious History, in Joshi, S.D. (ed.), Amr¢tadhårå. Professor R.N. Dandekar Felicitation Volume, Ajanta Publications, Delhi, pp. 285-300. Pollock, S. (1997), The ‘Revelation’ of ‘Tradition’: ‡ruti, Smr¢ti and the Sanskrit Discourse of Power, in Lienhard, S., Piovano, I. (eds.), Lex et Litterae. Studies in Honour of Professor Oscar Botto, Edizioni dell’Orso, Torino, pp. 395-417. Renou, L. (1941), Valid Forms in ‘Bhåßå’, in «The Indian Historical Quarterly», 15, pp. 245-250. Scharfe, H. (1971), Påñini’s Metalanguage, American Philosophical Society, Philadelphia.

VINCENZO VERGIANI Dealing with Conflicting Views within the Påñinian Tradition: On the Derivation of tyådr¢† etc.*

Most, if not all, intellectual traditions of ancient India were founded on a müla text the authority of which was –at least nominally– considered unassailable. In the case of vyåkaraña the cornerstone of the grammatical †åstra was the Aß™ådhyåyœ of Påñini, together with one of its earliest commentaries, the Mahåbhåßya of Pata∞jali that incorporates the vårttikas of Kåtyåyana.1 In time, the hierarchy of authoritativeness among the so-called three munis, the founding fathers, as it were, of Påñinian grammar, came to be reversed and greater authority was accorded to the latest author, Pata∞jali, followed by Kåtyåyana and Påñini.2 This reversal, whether justified or not, is itself the manifestation of a sense if not of history, of evolution, the implicit acknowledgment that Påñini’s grammar was perfectible, and consequently, that his first commentators had the task not simply of explaining it for the sake * This article has been to a large extent re-written in the spring of 2005 while I was a research fellow at the Pondicherry centre of the Ecole Française d’Extrême Orient (EFEO) thanks to a six-month’s fellowship granted by the EFEO. There I had the opportunity to discuss some of the issues I deal with here with Pandit Anjaneya Sarma, whom I wish to thank for his generosity in sharing his vast knowledge of vyåkaraña with me and his willingness to listen to a ‘non-traditional’ point of view. My thanks also go to Maria Piera Candotti for her relevant comments on an earlier draft of this paper. Of course, I alone am responsible for the views expressed here. 1 Påñini is generally assigned to the 5th century B.C.E., Kåtyåyana to the 3rd B.C.E. and Pata∞jali to the 2nd B.C.E. However, all these dates should be considered tentative because none is based on incontrovertible evidence. The same holds –with different degrees of approximation– for most of the dates given below. 2 On the emergence and evolution of this attitude, which is clearly expressed for the first time by Kaiya™a (11th C.E.) in the statement yathottara∫ hi munitrayasya pråmåñyam (Mahåbhåßyapradœpa, in Mahåbhåßya, Chaukhamba, vol. I, p. 339b), and its implications, see the important contributions of Madhav Deshpande (in particular, Deshpande 1993; 1998).

412

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

of men who were less quick-witted, but also of amending and supplementing it. Furthermore, this sense of evolution entails the recognition that the three munis may have and indeed had conflicting views on some issues, although all of them shared the basic principles laid out in the Aß™ådhyåyœ. As a matter of fact, the Påñinian tradition is anything but monolithic, although this is sometimes disregarded by contemporary scholarship, both ‘traditional’ and ‘Western’. But this impression of its unbroken continuity and general unanimity is, to a large extent, the result of the ways in which it came to represent and reproduce itself over the centuries and, especially, in medieval times. In fact, the respect for the earliest authorities gradually evolved into a sort of ‘canonisation’3 that tried to play down the disagreements by striving to reconcile conflicting views and, where that proved to be impossible, by silently passing over the opinions that were considered unacceptable. However, this was a gradual development. As I will try to show below, the first known post-Pata∞jali Påñinœyas, namely Bhartr¢hari (5th century C.E.) and, to some extent, the author(s) of the Kå†ikåvr¢tti (7th century) had a more nuanced attitude and still felt free, when they deemed it necessary, to take sides clearly with one or the other of the three munis and even openly criticise some of their views. And yet, from the very beginning any disagreement between the three munis seems to have been perceived as a very delicate issue that required the adoption of certain rhetorical strategies to be dealt with. Thus, retracing the history of the vyåkaraña tradition –indeed, of any intellectual tradition of ancient India– demands from modern scholarship that attention be paid not only to the notions and ideas handed down from one generation to the next, but also to the ways in which these are organised into a consistent discourse. For a tradition is constituted not only by the more or less deliberate choice of the texts to be transmitted4 and the topics to be treated, but also by its more or less conscious modes of dealing with internal conflict and with theoretical innovation. In this article I will focus on one such case of open disagreement among the three munis and the ways in which different authors in the later Påñinian tradition dealt with it. 3 According to Deshpande, already in Bhartr¢hari’s time, “Grammarians […] must look back to the golden age of the great ancient grammarians and seek authority in their statements. Thus, the ancient grammarians themselves become ‡iß™as with supernormal cognitive and mystical abilities […] The ancient grammarians are no longer to be treated as Åcåryas, but as R¢ßis” (Deshpande 1998: 20). I believe this is a fairly accurate general picture but, as I have tried to show here, in the late classical period it has to be taken with a pinch of salt and verified case by case by analysing the manner in which specific issues are discussed in relation to their treatment by the earliest authors. 4 It should not be forgotten that in pre-modern societies this amounted very concretely to deciding which texts should be committed to memory and/or writing.

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 413

1. The derivation of tyådr¢† etc. according to Aß™ådhyåyœ 3.2.60 The derivation of complex formations such as tyådr¢†/tyådr¢†a, sadr¢†/sadr¢†a etc., in which -dr¢† means ‘similar to, resembling’ while the first member is a pronominal base serving as the term of comparison,5 is provided for by Aß™ådhyåyœ 6 3.2.60. In itself, this looks like a minor issue of limited scope,7 for it concerns no more than a handful of forms.8 Nonetheless, Påñini’s formulation of the relevant sütra raises a number of issues that are first discussed in the Mahåbhåßya, where the apparently conflicting views of Kåtyåyana and Pata∞jali remain unreconciled. The same question is taken up again by Bhartr¢hari in two kårikås of the Sådhanasamudde†a,9 vv. 64-65, and the reason for his dealing there, uncharacteristically, with such a minute point of grammatical derivation is, I believe, that the brief and yet unresolved discussion on these forms found in the Mahåbhåßya gives him the opportunity to stress an important tenet of vyåkaraña methodology. But, before considering the views of Bhartr¢hari and other later grammarians in detail, let us first examine the sütra and the bhåßya on it. 2. A 3.2.60 and the remarks of Kåtyåyana and Pata∞jali thereon A 3.2.60, tyadådißu dr¢†o ’nålocane ka∞ ca, provides for the affixation of either of the two kr¢t suffixes Ka˚ or KviN10 to the root dr¢†- cooccurring11 with one of the items listed in the gaña beginning with tyad (tyadådißu)12 that serves as its object,13 provided that dr¢†- does not 5 With the exception of sadr¢† in which, according to Påñini, the first member is samåna: see below for details of derivation. 6 Henceforth, A, before sütra numbers. 7 This may be regarded as a weak point in the argument I develop in this article. However, I think that even if this were a unique case not only of disagreement among the early grammarians, but also of rejection of Pata∞jali’s views on the part of Bhartr¢hari and, less surprisingly, of the Kå†ikåvr¢tti, it would still be very significant. Moreover, I suspect that a careful examination of the commentators’ discussions on specific grammatical points would bring to light a number of similar instances. 8 Whitney (19245: 198) lists mådr¢†, tvådr¢†, yußmådr¢†, tådr¢†, etådr¢†, yådr¢†, œdr¢† and kœdr¢†, to which anyådr¢† and sadr¢† should be added. All these also have the variant form with -dr¢†a (mådr¢†a, tådr¢†a, etc.), and a few of them a third variant with -dr¢kßa. 9 The seventh chapter of the third book of the Våkyapadœya. 10 KviN is read here by anuvr¢tti from A 3.2.58; in the derivational process it is then replaced by ∅, namely deleted, according to A 6.1.67, ver apr¢ktasya. 11 A 3.2.60 belongs to the section governed by A 3.1.92, tatropapada∫ saptamœstham, according to which an upapada, i.e. an item co-occurring with one of the verbs mentioned in the following sütras, will be mentioned with a saptamœ, i.e. locative, ending. Such is the case of tyadådißu. 12 This gaña is a sub-group of the sarvådi gaña, the members of which receive the designation sarvanåman by A 1.1.27, sarvådœni sarvanåmåni, and therefore follow the ‘pronominal’ declension. In the established order of the list it includes the following items: tyad, tad, yad, etad, adas, idam and eka. Of these, adas does not occur with dr¢†-, while eka does but in a different meaning (ekadr¢† = one-eyed). 13 The condition karmañi is continued here through anuvr¢tti from A 3.2.1, karmañy añ.

414

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

mean ‘seeing’ (anålocane), to derive such terms as tyådr¢†/tyådr¢†a, tådr¢†/tådr¢†a, and so forth. Vt. 1 on A 3.2.60, dr¢†eΔ samånånyayoΔ ca upasaõkhyånam, adds samåna and anya to the group of words coming under the provision of A 3.2.60. This varttika is required because samåna is not a sarvanåman, while anya is not a member of the subgroup headed by tyad. Another group of rules in a different section of the Aß™ådhyåyœ helps to complete the derivation. A 6.3.89, dr¢gdr¢†avatußu, provides for the substitution of sa for samåna before -dr¢†/-dr¢†a (hence: samåna + dr¢† + KviN = sa + dr¢† + ∅ = sadr¢†). A 6.3.91, å sarvanåmnaΔ, then provides for the substitution of å for the final phoneme of a sarvanåman co-occurring with dr¢† (that is read here by anuvr¢tti from A 6.3.89) followed by Ka˚ or KviN (hence: tad + dr¢† + KviN = ta + å + dr¢† + ∅ = tådr¢†).14 Now, in vt. 2 ad A 3.2.60, kr¢darthånupapattis tu (Mahåbhåßya, vol. II, p. 107, l. 9), Kåtyåyana objects that in this case there is no justification (anupapatti) for resorting to primary suffixes (kr¢t), for these are generally agentive suffixes,15 but here no agent of seeing is intended.16 Consequently, vt. 3, ivårthe tu taddhitaΔ (Mahåbhåßya, vol. II, p. 107, l. 11), suggests that in tyådr¢† etc. one should rather recognise a formation with secondary (taddhita) suffixes17 having the meaning of iva ‘like, similar to’. Therefore, these suffixes ought to be prescribed among taddhitas, in the sub-section headed by A 5.3.96, ive 14 Note that A 6.3.89 seems to presuppose the existence of vt. 1 on A 3.2.60, since the latter rule does not provide for samåna. Similarly, in A 6.3.90, idaõkimor œ†kœ, kœ is given as the substitute for kim before -dr¢†/-dr¢†a, but again, kim is not a member of the tyadådi gaña either. Nor are tvad, yußmad and mad, that form similar derivatives (tvådr¢†, yußmådr¢† and mådr¢†). From the traditional point of view A 6.3.89 may be considered a j∞åpaka showing that Påñini intended A 3.2.60 to apply also to samåna (Pt. Anjaneya Sarma, personal communication, Pondicherry, March 2005). Similarly, I suppose that A 6.3.90 may be a j∞åpaka for kim and A 6.3.91, with its use of sarvanåmnaΔ instead of tyadådeΔ, a j∞åpaka for anya, yußmad, tvad and mad. However, this involves a good deal of pratipattigaurava, ‘prolixity in understanding’. Furthermore, as far as I can tell, none of the commentators explicity says that these are indeed j∞åpakas. At present I have no better answer to this riddle. 15 As is stated in A 3.4.67, kartari kr¢t. 16 Cf. Mahåbhåßya on A 3.2.60, vt. 2: “Here the meaning of a primary suffix is not justified. [The suffix] would wrongly apply to denote the agent of dr¢†-” (kr¢darthas tu na upapadyate. dr¢†eΔ kartari pråpnoti) (Mahåbhåßya, Kielhorn, vol. II, p. 107, l. 9). And Kaiya™a’s gloss thereon: “The meaning of a kr¢t cannot be expressed by words such as tådr¢† etcaetera. tådr¢† etc. convey a meaning characterised as ‘similar’, not the agent of the action of seeing” (yaΔ kr¢to ’rthaΔ sa tådr¢†ådi†abdavåcyo na bhavati. tådr¢†ådayo hi sadr¢†alakßañam artham åcakßate, na tu dar†anakriyåyåΔ [Rohatak: °å∫] kartåram) (Mahåbhåßyapradœpa, in Mahåbhåßya, Chaukhamba, vol. III, p. 171a; here the Chaukhamba edition is partially illegible due to bad printing, therefore I have integrated it with the Rohatak edition, vol. III, pp. 242-43). Incidentally, note that the final part of Kaiya™a’s comment is probably a quotation of a passage in Helåråja’s commentary on Sådhanasamudde†a 64: sadr¢†alakßaña∫ hy artha∫ tådr¢†ådayaΔ †abdåΔ pråhuΔ, na tu dar†anakriyåyåΔ kartåram iti vårttikårthaΔ (Prakœrñaprakå†a, in Våkyapadœya, Iyer, vol. III, part 1, p. 279, ll. 8-9). 17 For a possible reformulation of the rule, see the Uddyota on A 3.2.60, quoted and discussed below.

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 415

pratikr¢tau.18 Kåtyåyana does not elaborate further on the proposed modification of the sütrapå™ha.19 However, he may have possibly presented his own version of the rule in a vårttika that Pata∞jali chose not to include in the Mahåbhåßya,20 something along the lines of what is found more than a thousand years later in Någe†å’s commentary21 on the bhåßya on A 3.2.60: “if [this suffix were] a taddhita [listed] under the heading ive [in A 5.3.96], one should conjecture [the occurrence of] the suffixes dr¢†, dr¢†a™ and dr¢kßa after tyad etc”.22 It would certainly be in keeping with the Påñinian procedure to ignore the possible (and sometimes transparent) etymological origin of a certain element in a taddhita formation and treat it instead as an independent suffix for the sake of economy (låghava) in the grammatical description: consider, for instance, the taddhita affixes vidhaL and bhaktaL prescribed by A 4.2.54, rüpaP (A 5.3.66), kalpaP (A 5.3.67), and so on. It is evident that Kåtyåyana’s criticism of A 3.2.60, with its suggested reclassification of the element dr¢† in tyådr¢† etc. from verbal base (dhåtu) to secondary suffix (taddhita), entails quite a major change in the sütrapå™ha. It is not surprising, then, that Pata∞jali is unwilling to follow in his footsteps and tries to offer an exegetical solution to the problem raised by his predecessor that does not require any reformu18 Cf. Helåråja’s explanation: “This suffix should be called taddhita and prescribed in the section headed by [A 5.3.96] ive pratikr¢tau, ‘[kaN is introduced after a nominal base] to denote the sense of ‘like, similar to’ when a likeness or model is signified”. What is meant is that in this way the [desired] meaning, namely sa iva = tådr¢†aΔ etc., is realised’ (ive pratikr¢tåv ity atra prakarañe ’ya∫ taddhitasa∫j∞aΔ pratyayo vaktavyaΔ. tathå ca sa iva tådr¢†a ityådy arthasiddhir ity arthaΔ) (Prakœrñaprakå†a, in Våkyapadœya, Iyer, vol. III, part 1, p. 279, ll. 11-12). 19 Thus, in theory it is still possible to suppose that even for him this was just a pürvapakßa. This possibility is suggested by the fact that, as Subramania Iyer observes (Våkyapadœya, Iyer, vol. III, part 1, p. 279, critical apparatus, d), “The sentence ‘karmakartr¢tvåt siddham’ –reported by Helåråja as Pata∞jali’s confutation of vt. 3– is not found in the present text of the Mahåbhåßya … [and it] reads more like a Vårttika than a Bhåßya passage”. However, Helåråja, who is the only commentator, as far as I know, to express a clear opinion on this point, declares in unmistakable terms that vt. 3 indeed represents Kåtyåyana’s final view: “What is meant is that in this way the [desired] meaning, namely sa iva = tådr¢†aΔ etc., is realised. Thus, this is the final view established by the vårttikakåra” (tathå ca sa iva tådr¢†a ityådy arthasiddhir ity arthaΔ. eva∫ vårttikakåreña siddhåntitam) (Prakœrñaprakå†a, in Våkyapadœya, Iyer, vol. III, part 1, p. 279, l. 11-12). 20 We cannot be sure that the whole corpus of Kåtyåyana’s vårttikas has been preserved. In fact, it seems more than likely that Pata∞jali only reported and commented upon a selection of vårttikas. 21 By no means am I suggesting that Någe†a actually knew such a vårttika. As far as I can tell, he is the first Påñinœya to actually enunciate the essential elements of the new rule –tyadådibhyo dr¢gdr¢†a™dr¢kßåΔ– but this may well be nothing more than the result of the logical conclusions drawn from vt. 3 on A 3.2.60. Moreover, Någe†a’s remarks seem to be a kind of theoretical exercise in the context of a pürvapakßa, for in the second part of the Uddyota as well as in the Br¢hacchabdendu†ekhara he appears to follow Pata∞jali –as might be expected– in the latter’s rejection of any change to the sütra and his alternative semantic analysis of tyådr¢† etc (see below for a detailed account of Någe†a’s views). 22 ivådhikåre taddhite tyadådibhyo dr¢gdr¢†a™dr¢kßåΔ pratyayåΔ kalpyåΔ (Uddyota, in Mahåbhåßya, Chaukhamba, vol. III, p. 171b).

416

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

lation of the rule.23 The agent denoted by Ka˚/KviN (and their derivatives) is a karmakartr¢,24 he says, namely a former object that has turned into the agent of the same action: Otherwise, (let us say that) the meaning of a kr¢t is indeed appropriate here. This is a karmakartr¢. People see that one as similar to him. That one who is seen as similar to him sees himself as similar to him. (Hence) tådr¢k = similar25.

As is shown by the above quotation, Pata∞jali does not really explain what he means by karmakartr¢ in this context. He gives instead a sort of paraphrase of the meaning of a complex formation (vr¢tti) such as tådr¢†: the action of the agent (ayam) is first implicitly compared to that of other people (tam ivemam pa†yanti jånåΔ) and is ultimately rendered by an active verb form, pa†yati, governing the reflexive åtmånam as its object. In this way Pata∞jali rejects Kåtyåyana’s criticism and saves the letter, if not the spirit, of Påñini’s sütra. 3. Bhartr¢hari’s presentation of Pata∞jali’s view Although the reasoning behind Pata∞jali’s words is, I think, far from clear, two things should be noticed in his paraphrase. Firstly, in this context Pata∞jali seems to entertain a ‘notional’ view, as it were, of the category of karmakartr¢, rather than a strictly grammatical one, for in the analytical expression quoted above, by means of which he explains tådr¢†, there is no karmavadbhåva, i.e. passive verbal morphology, for the agent, which would lead to the derivation of an ungrammatical sentence such as *dr¢†yate sa svayam eva26. It may be argued that this is an unwarranted development of the Påñinian notion of karmakartr¢ as it appears from A 3.1.87, karmavat karmañå tulyakriyaΔ, that 23 That this is the preoccupation behind Pata∞jali’s rejection of Kåtyåyana’s suggestion is explicitly stated by Helåråja: “On the other hand, in order to confute what has been objected, namely that the meaning of the kr¢t has no logical justification, on the basis of the sütra as it is (yathånyåsam), the bhåßyakåra says: ‘it is achieved because it is a karmakartr¢’” (bhåßyakåras tu kr¢darthånupapattir iti codya∫ yathånyåsam eva parihartu∫ karmakartr¢tvåt siddham iti) (Prakœrñaprakå†a, in Våkyapadœya, Iyer, vol. III, part 1, p. 279, ll. 12-15). 24 According to A 3.1.87, karmavat karmañå tulyakriyaΔ, an agent (kartr¢ is read here by anuvr¢tti from A 3.1.68, kartari †ap) whose action is similar to the activity it performs when it is the object of the same action is treated as an object, that is, it requires passive verbal morphology. Such an agent is called karmakartr¢ (this term is used by Påñini himself in A 3.1.62, acaΔ karmakartari), while the operation prescribed by A 3.1.87 is commonly called karmavadbhåva ‘treatment [of the agent] as an object’ from the Mahåbhåßya onwards. 25 atha vå yuktaΔ eva atra kr¢darthaΔ. karmakartåyam. tam ivema∫ pa†yanti janåΔ so ’yam sa iva dr¢†yamånas tam ivåtmåna∫ pa†yati. tådr¢k. (Mahåbhåßya, Kielhorn, vol. II, p. 107, ll. 13-14). The bhåßya then goes on giving the same kind of paraphrase for anyadr¢†: anyam ivema∫ pa†yanti janåΔ. so ’yam anya iva dr¢†yamåno ’nyam ivåtmånam pa†yati. anyådr¢g iti (Mahåbhåßya, Kielhorn, vol. II, p. 107, ll. 14-15). 26 Structurally analogous to proper examples of karmakartari constructions such as ku†ülaΔ svayam eva bhidyate, lüyate kedåraΔ svayam eva, etc., found in the Mahåbhåßya (e.g., Mahåbhåßya, Kielhorn, vol. II, p. 67, ll. 17 and 19).

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 417

is precisely meant to provide for karmavadbhåva. Nonetheless, it is a development that, as will be shown below, is indeed explicitly argued for by some later Påñinœyas on the basis of Pata∞jali’s authority. Secondly, Pata∞jali’s interpretation is based on a meaning of dr¢†that appears to be quite different from its usual meaning, for it is hard to maintain that from the above sentence (as well as from tådr¢† etc.) one understands that the agent ‘sees himself’ in the same sense as he would ‘see himself’, for example, in a mirror. Rather, one may conclude that, according to Pata∞jali, here dr¢†- has more or less the sense of ‘manifesting oneself’ and, therefore, that which was the former object of seeing plays an agentive role in an action that maintains only a rather vague semantic relationship with seeing, namely, only to the extent that the usage of tådr¢† etc. presupposes a visual judgment on the speaker’s part. Even though it is barely articulated, Pata∞jali’s view has the advantage of preserving the integrity of the sütrapå™ha. At the same time, its very ambiguity paves the way for further semantic speculations. In Sådhanasamudde†a 64 Bhartr¢hari concisely gives the gist of Pata∞jali’s view in the first line and then proceeds to bring out the semantic implications of a karmakartr¢ interpretation in his own terms. However, judging from the conclusions he reaches in Sådhanasamudde†a 65, I believe he considers it to be a pürvapakßa. Sådhanasamudde†a 64 comes at the end of a long subsection in the karmådhikåra (section on karman) in which Bhartr¢hari shows how even the grammatical object, being a kåraka, a participant in the action, possesses a degree of agency (kartr¢tva)27 that can become fully manifest under certain conditions, precisely in the so-called karmakartari constructions, thus being reflected by the morphosyntactic structures of the language.28 It is in this light29 that he reads Pata∞jali’s interpretation of the semantic content of tyådr¢† etc.: sadr¢†ådißu yat karma kartr¢tva∫ pratipadyate Ù åpattyåpådane tatra vißayatva∫ prati kriye ÙÙ30 27 Cf., for instance, Sådhanasamudde†a 18ab: “As far as the mere realisation [of the action] is concerned, agency exists in every kåraka” (nißpattimåtre kartr¢tva∫ sarvatraivåsti kårake) (Våkyapadœya, Rau, p. 133). 28 Cf. Sådhanasamudde†a 55: “That [entity] which relies on its own activity, although there is a distinct notion of their activity [i.e. the activity of other agents], in certain contexts acquires the grammatical provisions assigned to the object as established in grammar” (tadvyåpåraviveke ’pi svavyåpåre vyavasthitam Ù karmåpadiß™å∫l labhate kva cic chåstrå†rayån vidhœn) (Våkyapadœya, Rau, p. 136). The ‘grammatical provisions’ are the operations prescribed by A 3.1.87 and related rules. See n. 24 above. 29 These notions are to a large extent the fruit of Bhartr¢hari’s original reflection, although it can be shown that they develop certain ideas and intuitions that can be traced back to the Mahåbhåßya, to which Bhartr¢hari himself and his commentator Helåråja constantly refer. 30 Våkyapadœya, Rau, p. 137. However, note that Rau has karmakartr¢tva∫ as one word, whereas Subramania Iyer separates the two terms (Våkyapadœya, Iyer, vol. III, part 1, p. 272). As is known, in the manuscripts the difference would not normally be noticeable.

418

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

With regard to that [object] which acquires the status of agent in [words like] sadr¢†a etc., both actions of ‘becoming’ and ‘causing to become’ [occur] in connection with its being involved [in dr¢†-, ‘seeing’].

The view presented by Bhartr¢hari in this kårikå appears to be an attempt to justify Pata∞jali’s attribution of a karmakartr¢ status to the agent denoted by the kr¢t suffix in words such as sadr¢† etc. This attempt is based on a sophisticated semantic analysis of the root dr¢†according to which it signifies two distinct activities, one pertaining to the object, the other pertaining to the agent. The former is described as åpatti ‘becoming’, the latter as åpådana ‘causing to become’ in 64c, and the implicit predicate is vißaya, an ‘object31 of cognition’, namely of visual perception, or more generally ‘that which is involved in the action’, as is indeed clarified in 64d by the clause vißayatva∫ prati. From this angle, in a sentence like gha™a∫ pa†yati ‘he sees the jar’, the jar is not just passively involved in the action, but it has an activity of its own, as is indeed expected of any kåraka.32 This activity is described by Bhartr¢hari in Sådhanasamudde†a 53 with the words “Becoming visible, being manifest and being amenable [to serving as the object of a certain action]: these, are the specific [properties] of an ‘object to be attained’ (pråpya) that are deployed in the accomplishment of the action”.33 On the one hand, any ordinary act of cognition requires an object for its accomplishment, and the very existence and availability of such an object of cognition is its contribution or assistance (upakåra) to the action. On the other hand, according to Bhartr¢hari no entity can be conceived and spoken about unless it is associated with an action, and the minimal action of any cognised object is its assumed existence.34 Thus, at first sight it could be reasonably argued that even an object of seeing has the prerequisites for turning into an agent of the karmakartr¢ type, as defined in Sådhanasamudde†a 56: “The object, for which any (external) instigation has ceased, relies on its own portion of the action. Since its status of object is suspended, it relies on its own agency”.35 Consequently, one may claim that once the instigation of the ‘real’ agent of seeing has stopped –the åpådana portion of seeing in the words of Sådhanasamudde†a 64– the object of seeing The latter reading seems to agree with Helåråja’s interpretation, for he glosses it as follows: yat karmañaΔ kartr¢tvam uktam (Prakœrñaprakå†a, in Våkyapadœya, Iyer, vol. III, part 1, p. 279, l. 24). I follow this reading in the translation. 31 Here, of course, in the non-grammatical sense of the word. 32 See n. 27 above. 33 åbhåsopagamo vyaktiΔ so∂hatvam iti karmañaΔ Ù vi†eßåΔ pråpyamåñasya kriyåsiddhau vyavasthitåΔ (Våkyapadœya, Rau, p. 136). 34 Clearly, this does not need to be actual ‘external’ existence. An object of cognition can very well have a merely mental existence. 35 nivr¢ttapreßaña∫ karma svakriyåvayave sthitam Ù nivartamåne karmatve sve kartr¢tve ’vatiß™hate (Våkyapadœya, Rau, p. 136).

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 419

can emerge as the agent of its own sub-action within the action of seeing, namely vißayabhåvåpatti,36 its becoming involved in the action. 4. Bhartr¢hari’s rejection of the karmakartr¢ interpretation However, this will not do. The karmakartr¢ interpretation of tyådr¢† etc. is forcefully rejected by Bhartr¢hari in Sådhanasamudde†a 65, although no direct allusion is made to the relevant bhåßya: kuta† cid åhr¢tya padam eva∫ ca parikalpane Ù karmasthabhåvakatva∫ syåd dar†anådyabhidhayinåm ÙÙ37 If one has this idea of supplying a word from somewhere in this manner, [even verbs] expressing seeing etc. would be karmasthabhåvaka ‘denoting states residing in the object’.

For a grammarian, the semantic analysis of an utterance, as well as of the words found therein, has to be confined to the ordinary speaker’s experience and practice. This principle is asserted in very clear terms in Sådhanasamudde†a 38cd: “The world accepts the authority of language, and this †åstra follows it”.38 Now, according to Bhartr¢hari an object of seeing –indeed, any object of a verb of cognition– is a karman of the pråpya kind,39 as defined in Sådhanasamudde†a 51: “When no accomplishment of specific traits produced by the action is cognised through either perception or inference, the [object] is called pråpya ‘to be attained’”.40 Such an object cannot turn into a karmakartr¢ because, according to the ordinary semantic value of verbs of cognition41 in Sanskrit, nothing happens to it when it serves as karman. In other words, it can36 See the following explanation given by Helåråja in the Prakœrñaprakå†a on v. 64: “There [in dr¢†-] the perceptual object’s activity, characterised as åpatti ‘reaching’, is subordinate to the agent’s activity, therefore, when the latter is present, the object’s independence with regard to its own activity is obscured. But when the perceptual object, located in a suitable place, happens to be in very bright light, then, being easily visible because its apparition is very distinct by itself, it excludes the reference to an agent. So, when the speaker wishes to convey its independence with regard to its own activity, that which was the grammatical object becomes the agent, [and] therefore it is in the domain of the karmakartr¢” (tatra kartr¢vyåpåre åpattilakßaño vißayavyåpåro nyagbhavatœti tatsannidhau svavyåpåre svåtantrya∫ vißayasya tirobhavati. yadå tu yogyade†åvasthitaΔ spaß™atarålokamadhyaparivartœ vißayaΔ tadå sudar†ano bhavan svayam abhivyaktadar†anaΔ kartr¢viniyogam apåkarotœti svavyåpåre svåtantryavivakßåyå∫ karma bhütvå kartå sa∫padyata iti bhavati karmakartr¢vißayaΔ) (in Våkyapadœya, Iyer, vol. III, part 1, p. 280, ll. 3-7). The term vißayabhåvåpatti itself is used by Helåråja a few lines below (ibidem, l. 12). 37 Våkyapadœya, Rau, p. 137. 38 †abdapramåñako lokaΔ sa †åstreñånugamyate (Våkyapadœya, Rau, p. 135). 39 One of the three kinds of œpsitatama karman as established by Bhartr¢hari in Sådhanasamudde†a 45, the other two being the nirvartya karman ‘object to be produced’ and vikårya karman ‘object to be modified’. 40 kriyåkr¢tå vi†eßåñå∫ siddhir yatra na gamyate Ù dar†anåd anumånåd vå tat pråpyam iti kathyate (Våkyapadœya, Rau, p. 136). 41 Or motion, the other category of Sanskrit verbs the objects of which are considered pråpya.

420

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

not appear in the role of an agent having an action similar to the one it had as an object (karmañå tulyakriyaΔ, in the words of A 3.1.87) because that action did not affect it in any way. The verbs denoting this type of action are in fact either kartr¢sthabhåvaka, namely referring to states residing in the agent, like dr¢†- ‘to see’, or kartr¢sthakriyaka, referring to actions residing in the agent, like gam- ‘to go’. It seems, then, that for Bhartr¢hari the karmakartr¢ interpretation of tyådr¢† etc. goes against the common sense of Sanskrit speakers. As Helåråja affirms in his commentary on Sådhanasamudde†a 65: “Here [in ordinary language] mere seeing, which resides in the agent, is denoted by dr¢†- in pa†yati ‘he sees’, for [the action of] becoming a (perceptual) object is not understood from it [i.e. from pa†yati]. On the contrary, the meaning of the verb is only the cognition of a (perceptual) object (vißayaj∞ånam)”.42 The kind of philosophical speculation on the nature of visual perception echoed by the semantic analysis of dr¢†- presented in Sådhanasamudde†a 64 violates a fundamental principle of vyåkaraña, which consists in following the laukika point of view in its understanding and explanation of grammatical forms and structures. By strictly adhering to the principles of Påñinian derivation, such risks are avoided, as is suggested by Bhartr¢hari in Sådhanasamudde†a 65ab. The derivation of any utterance must take into account only the words actually used therein and rely on their current meanings. The type of paraphrase alluded to in the Sådhanasamudde†a 64 does not respect this criterion and is therefore judged unacceptable. The importance attached by Bhartr¢hari to this tenet of grammatical methodology prompts the irony that seems to transpire from the first line of Sådhanasamudde†a 65 (kuta† cid åhr¢tya padam eva∫ ca parikalpane). In fact, it could hardly be maintained that the analytical expression of tådr¢† given by Pata∞jali –that I have quoted above– is a ‘natural’ semantic equivalent of tådr¢†. And, derivationally, the theoretical initial string of a sentence like ayam sa iva dr¢†yamånaΔ tam iva åtmånam pa†yati 43 is certainly not equivalent to that of aya∫ tam pa†yati, where dr¢†- is used in a sense other than ‘seeing’ (anålocane), according to Påñini’s wording of the sütra, for the transition from the former to the latter would require un-Påñinian operations such as the deletion of lexical items.44 Hence, I suspect, Bhartrhari’s ironical attitude towards this procedure that involves “supplying a word from somewhere”. Such an attitude is quite unusual in the author of the Våkyapadœya, who is known to be a champion of theoretical ‘perspectivism’.45 And it 42 iha dr¢†eΔ kartr¢stha∫ dar†anamåtra∫ våcyam pa†yatœti. na hi ato vißayabhåvåpattir avagamyate, ki∫ tu vißayavij∞ånam eva (Prakœrñaprakå†a, in Våkyapadœya, Iyer, vol. III, part 1, p. 280, ll. 11-12). 43 Or even of its reduced version, tam iva åtmånam pa†yati, which alone is considered by Någe†a (see below). 44 For a detailed treatment of this topic, see Deshpande (1985). 45 This expression, in relation to Bhartr¢hari’s general attitude towards theories and doctrines other than his own, has been proposed and vigourously defended by Jan

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 421

is all the more remarkable since the ultimate unnamed target appears to be none other than Pata∞jali. And yet, it must be noticed that not only is the bhåßyakåra not mentioned by name –which is the common practice in Sanskrit †åstra, no matter whether the mention of another’s opinion is appreciative or derogatory– but also the relevant bhåßya is in no way directly referred to in the words used by Bhartr¢hari in Sådhanasamudde†a 64. However, this shows that even though the respect for a venerable authority may prompt Bhartr¢hari to display maximum tact in criticising it, it could not prevent him from doing so if a principle of vital importance was at stake. It should also be noticed that while rejecting Pata∞jali’s interpretation, Bhartr¢hari does not explicitly accept Kåtyåyana’s suggestion either. Indeed, the Sådhanasamudde†a does not hint at vt. 2 and vt. 3 on A 3.2.60. And although it must be admitted that such a reference would be out of context in that chapter, I think it likely that Bhartr¢hari simply chooses to gloss over what he considers an unnecessary modification of the sütrapå™ha.46 A similar attitude –extending also to the view presented in the Mahåbhåßya– can be observed in the Kå†ikåvr¢tti and in its earliest commentary, the Nyåsa (8th century) of Jinendrabuddhi, to which I will now turn. 5. The Kå†ikåvr¢tti and the Nyåsa on A 3.2.60: the strategy of silence. The Kå†ikåvr¢tti on A 3.2.60 makes no explicit reference to the earlier debate on this sütra. Nonetheless, after explaining the purpose of the term anålocane by means of a counter-example47 according to its usual procedure of word-by-word analysis, its author feels compelled to add: “tådr¢† etc. belong to the class of conventional48 words (rü∂hi†abda), there is no action of seeing in them”.49 And the Nyåsa specifies: Houben in a series of articles (Houben, 1992-93, 1995b and 1997) and in his book on the Sambandhasamudde†a (Houben, 1995a). Although I believe that Houben’s analysis of Bhartr¢hari’s thought is essentially correct, the present verse seems to be one of those cases in which the author’s preferences are clearly and, indeed, unmistakably expressed, thus suggesting that the notion of Bhartr¢hari’s ‘perspectivism’ should somehow be mitigated, as pointed out by Cardona: “This does not mean, however, that Bhartr¢hari should be considered not to have held definite views of his own and to have argued [… ] against other positions” (Cardona, 1999: 93). 46 On the other hand, Helåråja reports and discusses Kåtyåyana’s vårttikas in his introduction to Sådhanasamudde†a 64, where he gives a complete picture of the underlying issues, but he does so in a neutral way without giving any clear indication that he approves of it. 47 anålocane iti kim. ta∫ pa†yati taddar†aΔ (Kå†ikåvr¢tti, vol. II, p. 585). When dr†- is used in its usual sense of ‘seeing’ and is combined with an upapada designating its object, it is followed by the kr¢t suffix a˜ according to A 3.2.1, karmañy añ, forming the nityasamåsa taddar†a ‘who looks at him’. See the Padama∞jarœ thereon: […] ta∫ pa†yati tu taddar¢†a iti cakßurvij∞åne ’ñ eva bhavati (in Kå†ikåvr¢tti, vol. II, p. 585). 48 That is, words with a meaning that cannot be arrived at from the meanings of their basic constituents, as is explained by Jinendrabuddhi in the passage I quote below. 49 tådr¢gådayo hi rü∂hi†abdaprakåråΔ naivåtra dar†anakriyå vidyate (Kå†ikåvr¢tti, vol. II, p. 585).

422

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

The meaning of the constituent parts will not be found in conventional words. What then? The meaning of the complex [formation]. […] Therefore, here [in tådr¢† etc.], although the action of seeing is resorted to for the derivation of the analytical expression, nonetheless it is not [found] in the complex formation. For instance, a suffix is introduced at the end of words such as tådr¢†, tådr¢†a, to refer to something similar [to something else], not to refer to the action of seeing. Thus, it is indeed correct (to say) that it is a complex formation of dr¢†- in a sense other than ‘seeing’.50

Clearly, the author of the Kå†ikåvr¢tti (and its earliest commentator) knew the views of Kåtyåyana and Pata∞jali and disagreed with both. Arguing that tådr¢† etc. are ru∂hi†abda amounts to defending the rule in its original Påñinian formulation. Pata∞jali’s interpretation preserves the rule but offers a kind of alternative semantic explanation that is apparently unacceptable to the Kå†ikåvr¢tti for it would make the meaning condition anålocane superfluous (in fact, false). In so far as the action carried out by a karmakartr¢ must be the same as the action it helps to accomplish when it is a karman,51 if one maintains that the suffixes Ka˚/KviN are introduced after the root dr¢†- to denote a karmakartr¢, the condition anålocane is virtually invalidated. If, on the other hand, no action of seeing is intended in tådr¢†- etc. –and the Kå†ikåvr¢tti bluntly declares that it is not– there is no reason to resort to the notion of karmakartr¢: the root dr¢†-, whatever its meaning may be, governs an object (tyad etc.) and its agent is denoted by the suffixes Ka˚/KviN. However, once again it must be emphasised the text does not say so explicitly. In fact, the word karmakartr¢ does not appear at all in the vr¢tti on A 3.2.60, but its insistence on the conventional nature of the meaning of tådr¢† etc. seems to be a reply to the attempt to establish a semantic connection between the usual meaning of dr¢†- and these formations, as is done in the context of the karmakartr¢ hypothesis. Nor is there any evidence of Bhartr¢hari’s influence with his sharp critique of the karmakartr¢ hypothesis. The whole conflict among previous authorities is passed over in silence in the Kå†ikåvr¢tti, but a clear stand –and for that matter, an orthodox Påñinian one– is taken on the controversial matter. 6.Haradatta: the rising supremacy of Pata∞jali’s authority A few centuries later, in another commentary on the Kå†ikåvr¢tti, the Padama∞jarœ of Haradatta (11th century), we come across the first 50 rü∂hi†abdeßu nåvayavårthena bhavitavyam. ki∫ tarhi? samudåyårthena. […] tasmåd iha yady api vyutpattyartha∫ våkyasya dar†anakriyopådœyate tathåpy asau vr¢ttau nåsty eva. tathå hi tådr¢k tådr¢†a ityådibhyaΔ †abdebhyaΔ sadr¢†avastuvißaye pratyaya upajåyate. na tu dar†anakriyåvißaya iti yuktaiva dr¢†er anålocane vr¢ttiΔ (in Kå†ikåvr¢tti, vol. II, p. 585). 51 For instance, softening (viklitti) of the rice is the final result of the action of cooking as denoted by pac- both in a sentence like sa odana∫ pacati, where odana is karman, and in odanaΔ pacyate, where it is karmakartr¢.

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 423

attempt to reconcile at least some of the views expressed by the earlier authors. No mention is made of Kåtyåyana’s vårttikas that had originally triggered the debate – again, I believe, a sign of rejection of his alternative suggestion. However, after commenting on the Kå†ikåvr¢tti’s view of tådr¢† etc. as rü∂hi†abdas, Haradatta first presents Pata∞jali’s view, ascribing it explicitly to the Mahåbhåßya from which he quotes one crucial sentence –the explicatory paraphrase of tådr¢†– and then tries to explain it with arguments that are clearly inspired by the Sådhanasamudde†a. It is worth quoting the relevant passage in full: But in the bhåßya the derivation (of tådr¢† etc.) is shown to be in the sense of karmakartr¢: ‘People see that one as similar to him. That one who is seen as similar to him sees himself as similar to him’. In this respect it should be understood that either dr¢†- is used to denote only the activity of the object, or the whole meaning of the verb is superimposed (on the activity of the object) to express the extreme easiness (with which the action is carried out).52

To begin with, it should be noted that Pata∞jali’s view is reported in a neutral way, seemingly on a par with the previous –and, as I have shown above, independent– rü∂hi†abda interpretation. But, according to the conventions of †åstric discourse, its very position at the end of the argument emphasises it and makes it appear almost as the siddhånta. Moreover, Haradatta reads the karmakartr¢ interpretation of tådr¢† etc. in the light of some considerations on the notion of karmakartr¢ that Bhartr¢hari develops in the Sådhanasamudde†a. As we have seen, the problem is how to justify that dr¢†- is used in a karmakartari construction to denote the original object’s action. Two alternative semantic analyses are possible here. One is the nivr¢ttapreßañakarma view, epitomised in Sådhanasamudde†a 56 (quoted above), according to which an object, for which all instigation (preßaña) on the real agent’s part has ceased (nivr¢tta),53 relies upon its being the agent of the activity that represents its own portion of the action (that is included in the verb meaning).54 Haradatta does not explicitly say what this activity could be in the case of dr¢†-, but we may suppose that it is the vißayabhåvåpatti ‘becoming a (perceptual) object’, as suggested by Bhartr¢hari in Sådhanasamudde†a 64cd. The other view, that came to be known later as adhyåropitapreßanapakßa, is also found in 52 bhåßye tu karmakartari vyutpattir dar†itå - tam ivemam pa†yanti janåΔ so ’yam sa iva dr¢†yamånas tam ivåtmånam pa†yatœti. tatra karmavyåpåramåtre vå dr¢†er vr¢ttiΔ, kr¢tsnadhåtvarthådhyåropo vå saukaryåti†ayapratipådanåyeti draß™avyam (in Kå†ikåvr¢tti, vol. II, p. 585). 53 It is worth recalling here that this refers to the speaker’s subjective conceptualisation and verbal expression, not to the ‘real-life’ situation, whatever that may be. 54 On the polisemy of roots, cf. Sådhanasamudde†a 58: “[Verbs like] pac- and the like, which are associated by their very nature with a similar form, apply to one aspect of the activities [they denote] or to the whole” (ekade†e samuhe ca vyåpåråñå∫ pacådayaΔ Ù svabhåvataΔ pravartante tulyarüpasamanvitåΔ) (Våkyapadœya, Rau, p. 136).

424

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Sådhanasamudde†a 6255: it assumes that the instigating role that was originally (and in reality) played by the agent is attributed to –superimposed upon (adhyåropita), in the vaiyåkarañas’ terminology– the object in order to convey the fact that the action is performed very easily, almost by itself, as it were. In the latter view, the whole verbal meaning (kr¢tsnadhåtvartha), not just part of it, is meant and expressed. As can be seen, Haradatta seems eager to accommodate Pata∞jali’s view, to the extent that he uses some ideas found in Bhartr¢hari’s Våkyapadœya to support it although Bhartr¢hari himself unmistakably rejects the karmakartr¢ interpretation of tyådr¢†. Indeed, Haradatta even disregards the fact that the Kå†ikåvr¢tti, the text he is commenting upon, is firmly opposed to the idea that the sense of ‘seeing’ can be understood from tyådr¢†, in accordance with the semantic condition anålocane stated by Påñini in A 3.2.60. It seems, therefore, that by Haradatta’s time the process of reversal of the chronological hierarchy among the three munis had made further progress. Faced with an overt conflict among the earlier authorities, Haradatta adopts three different strategies: he omits what is wholly unacceptable (Kåtyåyana’s view), juxtaposes the other views (Pata∞jali’s and that of the Kå†ikåvr¢tti), ignoring the fact that they start from different premises, and finally defends Pata∞jali’s view –his favourite?– by resorting to Bhartr¢hari’s arguments that were originally directed against that same view. This process is fully achieved some centuries later, as can be seen from the works of Någe†a, one the last great Påñinœyas from pre-colonial times, that I will examine in the next paragraph. 7. Någe†a: the ever-present past In the Uddyota, Någe†a generally follows the structure of the texts he is commenting upon, the Mahåbhåßya of Pata∞jali and the Mahåbhåßyapradœpå of Kaiya™a, as is customary. In the Uddyota on A 3.2.60, however, he elaborates upon vt. 3, ivårthe tu taddhitaΔ, on which Kaiya™a –like many other earlier commentators– is altogether silent. As I showed above, he even posits the taddhita suffixes (dr¢†, dr¢†a™, dr¢kßa) that should be provided by a rule reformulated in accordance with the vårttika and he goes as far as showing the advantages of the new rule 55 “According to some [thinkers], the activity which pertains to Devadatta etc. [= and other agents] when the verb is transitive is conveyed according to the speaker’s intention as pertaining to the mat etc. [= and other objects] in the absence of Devadatta etc.” (keßå∫ cid devadattåder vyåpåro yaΔ sakarmake Ù sa vinå devadattådeΔ ka™ådißu vivakßyate) (Våkyapadœya, Rau, p. 137). Helåråja introduces this kårikå saying that it serves “to justify (the agency of the object) […] through the superimposition (adhyåropa) of prompting” (praißådhyåropeñåpy upapådayitum) on the object itself and then explains: “When there is the intention to convey the extreme easiness of the action, this [the prompting] is conveyed according to the speaker’s intention as referring to the object [itself], in the absence of the agent” (asau yadå saukaryåti†ayavivakßå karmañaΔ tadå kartåram antareña tatra karmañi vivakßyate) (Våkyapadœya, Iyer, vol. III, part 1, p. 278, l. 5 and ll. 9-10).

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 425

in terms of derivation. For he explains that the taddhita suffix dr¢†a™ is ™it, i.e. it is provided with ™ as its marker (it), and that causes the introduction of the suffix ÕœP (= œ) to derive a feminine stem according to A 4.1.15, ™i∂∂håña∞dvayasajdaghnajmåtractayap™ha∞ka∞kvarapaΔ: tådr¢†a → tådr¢†œ. This, in turn, makes the suffix Ka˚, that is used by Påñini only in A 3.2.60, superfluous, so that this can be dispensed with and there is a gain in terms of låghava ‘economy’ of description.56 Then, without any explicit criticism of Kåtyåyana’s suggestion, Någe†a moves on and, following Kaiya™a, he approvingly reports and elaborates upon Pata∞jali’s statement that the agent in tådr¢† etc. “is a karmakartr¢” (karmakartåyam). The Pradœpa states that the suffix (Ka˚ or KviN) is indeed a kr¢t introduced after dr¢†- to denote an agent of the karmakartr¢ kind “when the speaker intends to express the agency of the object alone because of the easiness (of the action)”.57 Någe†a suggests then that the analytical equivalent of tådr¢† given in the bhåßya –tam ivåtmåna∫ pa†yati, literally ‘one sees himself as similar to him’– should be understood as tam ivåtmåna∫ janån dar†ayati, literally ‘one shows himself to people as similar to him’. For, even though there is an additional meaning, namely ‘inducing people to see’ oneself in a certain way, the causative suffix ˜iC is not used because its meaning is implicit in the verb,58 an interpretation which presupposes Bhartr¢hari’s semantic analysis of dr¢†- in terms of (vißayabhåva-)åpatti and (vißayabhåva-)åpådana, where the object of vision is somehow actively involved in and even triggers the process. Furthermore, between the two alternatives presented by Haradatta, the nivr¢ttapreßañapakßa and the adhyåropitapreßañapakßa, Någe†a explicitly defends the latter against the former. The analytical equivalent (vigraha) of tådr¢† according to the former view, he affirms in the Uddyota, would be sa iva pa†yati. Here any external instigation is suppressed. However, this vigraha is inappropriate (anucita) because the verb no longer has an object as its co-occurring word,59 as is required by the condition karmañy upapade that is to be read into A 3.2.60.60 In order to prevent this, as Nage†a explains in greater detail in the Br¢hacchabdendu†ekhara, the Mahåbhåßya presents a derivation of tådr¢† based on the adhyåropitapreßañapakßa: tam ivåtmåna∫ pa†yati = dar†ayati, literally ‘one sees = shows himself as similar to him’, where the karmakartr¢, the object-turned-agent, appears as the instigator of another object that is himself, signified by the reflexive 56 dr¢†a™aß ™ittvåc ca õœb api siddha iti ™i∂∂heti. sütre ka∞grahañam api na kåryam iti låghavam iti (Uddyota, in Mahåbhåßya, Chaukhamba, vol. III, p. 171b). 57 karmaña eva saukaryåt kartr¢tvavivakßåyå∫ […] (Mahåbhåßyapradœpa, in Mahåbhåßya, Chaukhamba, vol. III, p. 171b). 58 prerañåvr¢ddhåv api antarbhåvitañyarthatvån na ñic. tam ivåtmåna∫ janån dar†ayatœty arthåΔ (Uddyota, in Mahåbhåßya, Chaukhamba, vol. III, p. 171b). 59 karmañy upapade evåyam. tena nivr¢ttapreßañatayå sa iva pa†yatœti vigraho ’nucitaΔ (Uddyota, in Mahåbhåßya, Chaukhamba, vol. III, p. 171b). 60 See nn. 11 and 13 above.

426

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

åtmånam. Moreover, in this interpretation the condition anålocane is again justified: ålocana is explained as j∞ånamåtra,61 ‘cognition in general’, which is the activity pertaining to the real agent of dr¢†-; since this is suppressed, in that it is not understood from tådr¢† etc., the mention of anålocane is required. Thus, the integrity of the sütrapå™ha is preserved and Pata∞jali’s view is accepted and established, all possible objections having been answered.62 We should note one thing in particular in the way Någe†a expresses his view. He does not simply read back into the Mahåbhåßya ideas that were clearly formulated only at a later time –in fact, to a large extent derived from the Våkyapadœya– but he explicitly attributes them to the Mahåbhåßya itself. It is worth quoting the relevant passage in the Br¢hacchabdendu†ekhara to point out its crucial discursive articulations: In this view63 the heading karmañy upapade ‘when there is a co-occurring word serving as its grammatical object’ is prevented from applying. But in order to prevent this, an [alternative] derivation is presented in the bhåßya according to the adhyåropitapreßañapakßa: ‘tam ivåtmåna∫ pa†yati [=] dar†ayati’.64

Then, in the concluding lines of the Br¢hacchabdendu†ekhara, Någe†a briefly recalls the Kå†ikåvr¢tti view: “However, others interpret ‘aj∞ånårthåd’ saying that tådr¢† etc. can be derived (according to A 3.2.60), although the meaning of their (constituent) parts is not found (therein), because these are conventional words”.65 This unexpected mention of the rü∂hi†abda view is possibly motivated by the recognition that the Kå†ikåvr¢tti’s solution is in a way the most elegant, as it were. It preserves the sütra in its original form and, at the same time, it does not require any complex exegetical effort since it takes Påñini’s rule at its face value: the condition anålocane 61 Siddhåntakaumudi 429 on A 3.2.60, tyadådißu…, reads: tyadådißüpapadeßv aj∞ånårthåd dr¢†er dhåtoΔ ka∞ syåc cåt kvin (Siddhåntakaumudi, vol. 1, p. 435). Commenting on it, Någe†a says: ålocanam iha j∞ånamåtram. tad åha aj∞ånårthåd iti (Br¢hacchabdendu†ekhara, vol. 1, p. 664). And a few lines below he explains: aj∞ånårthåd ity asya ca vißayœkarañånukülavyåpårårthakabhinnåd ity arthaΔ, “(dr¢†-) when it does not have the meaning of ‘cognising’, namely when it differs from that which signifies an activity conducive to turning (something) into an object of cognition” (Br¢hacchabdendu†ekhara, vol. 1, p. 664), that is, when it does not denote the real agent’s apprehending activity. 62 Whether the answers are really convincing is another matter. For it seems to me that one crucial point is missed in Någe†a’s interpretation: the meaning of iva, that is nonetheless used in all the vigrahas of tådr¢† etc., is simply ignored at the final stage of derivation and cannot be accounted for through any of the morphological constituents of these words. 63 That is, the nivr¢ttapreßañapakßa. 64 atra pakße ‘karmañy upapade’ ity adhikårabådha åpadyate. bhåßye tu tadabådhåya ‘tam ivåtmåna∫ pa†yati dar†ayatœty adhyåropitapreßañapakße vyutpattir dar†itå (Br¢hacchabdendu†ekhara, vol. 1, p. 664). 65 anye tu tådr¢†ådœnå∫ rü∂hi†abdatvåd asatåpy avayavårthena vyutpådanam ity aj∞ånårthåd ity asya saõgatim åhuΔ (Br¢hacchabdendu†ekhara, vol. 1, p. 664).

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 427

unequivocally specifies that the root dr¢†-, though recognisable in these forms, is not used in its ordinary meaning. Nothing suggests that Någe†a disapproves of it. On the contrary, he presents this view as if it were an alternative or additional justification of Bha™™oji Dœkßita’s gloss of anålocane, and the fact that it concludes the argument seems to support this conclusion. As appears from the words underlined in the translation of the first passage, Någe†a anachronistically assumes that Pata∞jali had the two alternative views before him and, in formulating his analytical rendering of tadr¢†66, chose the second in order to avoid the difficulties raised by the first. Similarly, in the other passage an argument going back to the Kå†ikåvr¢tti is said to be given as an explanation of a gloss in the Siddhåntakaumudœ. 8. Conclusions. Saying that this is just one more example of the notorious Indian lack of a sense of history does not help us to understand the attitude of Någe†a and his contemporaries and the motives behind it, and even less to confront and contrast it with those of Bhartr¢hari and the authors of the Kå†ikåvr¢tti and, at an even earlier period, of Kåtyåyana and Pata∞jali with regard to Påñini. Någe†a’s attitude stemmed from the deeply-rooted conviction that the seed of all possible theoretical development was already contained in the ancient texts on which the tradition was built, and in particular in the Mahåbhåßya67 But this was the result of a long historical development the different stages of which should be ideally read, as far as possible, within the wider context of the changing cultural conditions in which the grammatical tradition existed and flourished.68 This is an ambitious and difficult goal, but even if we confine ourselves to the history of vyåkaraña we can identify significant shifts that can help us to understand the underlying assumptions of this intellectual tradition. The grammatical issue that is the focus of this article is admittedly a minor one, but I think that the thorough study of the treatment of specific grammatical issues by different authors can throw light on 66 It should be noticed that, unlike Haradatta, he quotes only the final part of Pata∞jali’s vigraha, extracting it from a longer and rather baffling sentence (that I have quoted in its entirety above). 67 For a hypothesis on the historical reasons of the supremacy attributed to the Mahåbhåßya, see Deshpande 1998, in particular pp. 23-24. 68 Clearly, it would be highly desirable to be able to relate the history of the Sanskit grammatical tradition –or, rather, traditions, for it should not be forgotten that alongside the Påñinian school, others schools of grammar flourished at different times and places– to the varying socio-political conditions in which it developed. However, as is well-known and often-lamented in Indological scholarship, the very lack of historical references in the works of the Sanskrit intellectuals makes this task extremely difficult, if not impossible (on this point see, e.g., recently, the remarks of Sheldon Pollock in Pollock 2001).

428

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

the possible inconsistencies between their ‘official’ discourse on the tradition to which they belonged and the authorities on which this was founded, on the one hand, and their actual practice when it came to finding the best technical and theoretical arguments to deal with certain problems.69 Moreover, the issue of the derivation of tyådr¢† etc. gave rise to an interesting debate that serves to illustrate very well the different sensibilities of various authors at different times and the way in which they conceived their role as grammarians and, more generally, as intellectuals. In the earliest phase, that of the three muni –or the lakßyaikacakßußka grammarians, those ‘whose sole attention is fixed on the usage of Sanskrit to be described’, as they were called by the late Påñinœyas70– we can clearly witness a tradition that is still in the making. Kåtyåyana and Pata∞jali show a reverential respect for the teacher Påñini and, interestingly, in their works one can already detect a reluctance to modify the sütrapå™ha that is modelled on the reverence for the Vedic texts, which demanded and ensured that they be transmitted integrally. Nonetheless, this did not prevent either of them from proposing to modify, supplement or suppress rules if those found in the Aß™ådhyåyœ were considered inadequate, as is evident in the case of the derivation of tyådr¢† etc., for which Kåtyåyana suggests a radical reformulation of the relevant rule. Even Pata∞jali’s exegetical solution, while preserving A 3.2.60 with its original wording, appears quite innovative, as it propounds a semantic interpretation of these formations that brings in a Påñinian notion like that of karmakartr¢ which, however, is not evidently indicated in Påñini’s sütra. In Bharthari’s time, more than five hundred years after Pata∞jali, the respect for the earlier authorities had already turned into a quasi veneration, as shown by Deshpande (1998). However, the two verses from the Sådhanasamudde†a discussed here show that, if necessary, he would not refrain from criticising Pata∞jali’s views, although in a tactfully oblique way. What does this tell us about his attitude towards his own tradition? I think this is a positive indication that Bhartr¢hari is aware of being contributing to a tradition that he still considers creative and not just exegetical. As far as linguistic theory is concerned, he voices radically new ideas –first and foremost, that of the sentence as the primary meaning-bearing unit of speech– while understating the novelty of his views, as pointed out by Houben (2002). Moreover, starting from the principles and intuitions found in the works of his predecessors, 69 Therefore, I consider many of the conclusions drawn here working hypotheses that need to be verified through other similar studies of particular issues and in a wider range of texts. 70 In opposition to the lakßyañaikacakßußka, ‘those whose sole attention is fixed on the inherited rules of Sanskrit grammar’. On this phrases and their origins and implications, see Deshpande 1998, pp. 24-25, from which I have taken the English renderings. For a somewhat different interpretation of the same, see Aklujkar 2002-2003.

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 429

he brings grammar into the arena of the philosophical debate of his time as of equal standing with other schools of thought. The most conspicuous evidence for his awareness of his own originality is, of course, the very fact that the Våkyapadœya is not a commentary, but an independent treatise.71 This consideration helps us to view the boldness that Bhartr¢hari shows in Sådhanasamudde†a 65 in the correct perspective. There, I suspect, he is not only defending a fundamental principle of grammatical methodology, namely, the adherence to the ‘wordly’ point of view in the explanation of linguistic forms, but also stressing a major point in his philosophical programme as a grammarian-philosopher: the pronouncements of †åstra –of any †åstra– are not substantially different from the utterances of ordinary language with which they share the inevitable limitations of discursive thought (vikalpa); hence, the task of vyåkaraña is in a way to show through the analysis of language that while such pronouncements may be a further step in the quest for truth, they also contribute to the proliferation of vikalpa, away from that immediacy of experience in which, according to Bhartr¢hari, men fleetingly grasp the unity of being. Far from being a break with the past, his views represent the ideal continuation of the empiricism that characterises the early grammarians’ approach to language and cognition. And, I would suggest, it is in the name of the continuity of this philosophical view that he feels entitled to correct the occasional défaillance of the great teacher Pata∞jali, as in the case presented here. To put it differently, in the work of Bhartr¢hari72 we see the author shedding a completely new light on old issues by problematising them in original ways that echo the complexity of the contemporary philosophical debate with its many voices. There is hardly any trace, here, of the “traditionalism of the problematics” that can be observed in later grammatical texts.73 A couple of centuries after Bhartr¢hari, the Kå†ikåvr¢tti offers a new exegetical solution (as we have seen, implicitly critical of Pata∞jali’s) to the problem raised by Kåtyåyana, while refraining –as is generally its style– from directly engaging in the debate and dealing with broader theoretical issues. However, this interpretative freedom is found alongside the well-known modifications to the sütrapå™ha introduced elsewhere in the Kå†ikå and shows that its authors still felt entitled to contribute to the advancement of the grammatical tradition, not merely to its preservation. 71 Conversely, it is tempting to speculate that his commentary on the Mahåbhåßya, the Mahåbhåßya™œkå or °dœpikå, was left unfinished for the same reason. 72 And of his commentator Helåråja, as far as we can judge his views from his only extant work, the Prakœrñaprakå†a, considering that this is primarily a commentary on the Våkyapadœya. 73 This phrase is used by Pollock in connection with the various schools that called themselves navya –including the navyavyåkåraña– of late medieval times (Pollock, 2001: 12).

430

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

It is only with Haradatta, the second commentator of the Kå†ikåvr¢tti, and more fully, a few centuries later, with Någe†a, that we perceive a distinct change in mentality and attitude. As Pollock has remarked, in the eyes of these authors, “all intellectual generations, disembedded from any spatio-temporal framework, were thought of as coexistent: the past was a very present conversational partner […] viewed as a superior partner, the master who made the primary statement in a discussion upon which later participants could only comment”.74 Indeed, with regard to the problem of derivation discussed here,75 Någe†a goes a step further, virtually blurring all the distinctions between texts of different authors and periods. Most strikingly, he completely ignores the important theoretical point made by Bhartr¢ari in Sådhanasamudde†a 65ab and the conclusions he draws in 65cd, namely, that verbs of cognition cannot be considered karmasthabhåvaka, ‘denoting states residing in the object’.76 It seems likely that these opinions, which Någe†a may have possibly subscribed to in another context, were not discarded, but simply dropped here because they could not be reconciled with Pata∞jali’s conclusions. On the whole, from the passages examined here one gets the impression that Någe†a sees himself as the custodian –as in a museum– of the ever-present, rather than timeless, truths contained in the ancient texts of the grammatical tradition that are thus safeguarded and handed down to posterity, intact but inert. It may seem paradoxical that for scholars such as Någe†a and his contemporaries, who proudly defined themselves navyavaiyåkarañas, the ‘new grammarians’, the ultimate source of legitimation of any notion, theory or exegetical device was to be found in the works of those venerable authorities. But perhaps this is not so surprising, after all. Broadly speaking,77 it may be argued that the brahmanical monopoly over intellectual discourse and its self-perpetuation through the alliance with and legitimation of political power demanded that the role of the intellectual brahmanical elites across South Asia should be conceived and presented as unchangeable in an unchanging world. As a consequence of this fictional self-representa74

Pollock, 2001: 7. The following considerations can certainly not be generalised to the whole of Någe†a’s complex contribution to the field of vyåkaraña. As I hinted above, I think it is methodologically unsound to expect absolute ideological consistency in the works of any author. 76 As far as I can tell, Någe†a is not the only one to overlook these kårikås. In fact, none of the texts I have examined –except, of course, the Prakœrñaprakå†a of Helåråja– refer to the views expressed there. And, according to the list of quotations from the Våkyapadœya prepared by Abhyankar and Limaye (1965, Appendix III), the two kårikås are not found in any later work. 77 I present these final considerations as a very tentative attempt to delineate some of the relations between the late vyåkaraña tradition and the broader context of the South Asian civilisation, but this topic is too complex to be treated in a few lines and I am not particularly qualified to develop it. 75

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 431

tion, the established †åstras largely condemned themselves to walk along the old paths of tradition –and these too, were carefully selected– over and over again. In particular, a discipline such as vyåkaraña proved unable to transcend itself and, in spite of its highly sophisticated conceptual tools, hardly ever tried to expand the field of linguistic speculation beyond Sanskrit to the many languages that served as the means of expression for the South Asian civilisation not only in daily life but also in several –indeed, in Någe†a’s time, most– contexts of literary production.78

BIBLIOGRAPHY Sources Bhartr¢hari, Våkyapadœya of Bhartr¢hari with the Commentary of Helåråja, Kåñ∂a III, Part I, critically edited by K.A. Subramania Iyer, Deccan College Monograph Series 21, Poona, 1963. Bhartr¢hari, Bhartr¢haris Våkyapadœya. Die Mülakårikås nach den Handschriften herausgegeben und mit einem Påda-Index versehen, critically edited by Wilhelm Rau, Franz Steiner, Wiesbaden, 1977. Bha™™oji Dœkßita, Siddhåntakaumudœ, with the Bålamanoramå of Vasudeva Dœkßita and J∞ånendra Sarasvatœ’s Tattvabodhinœ, edited by Giridhara ‡armå Caturveda and Parame†varånanda ‡armå Bhåskara, Motilal Banarsidass, Delhi, 199597, 4 voll. Haradatta Mi†ra, Padama∞jarœ: see Kå†ikåvr¢tti. Helåråja, Prakœrñaprakå†a: see Våkyapadœya of Bhartr¢hari with the Commentary of Helåråja, Iyer edition. Jinendrabuddhi, Nyåsa: see Kå†ikåvr¢tti. Kaiya™a, Mahåbhåßyapradœpa: see Pata∞jali’s Vyåkaraña Mahåbhåßya, Chaukhamba edition. Kå†ikåvr¢tti of Våmana-Jayåditya, with the commentaries Nyåsa or Pa∞cikå of Jinendrabuddhi and the Padama∞jarœ of Haradatta Mi†ra, edited by Dwårikå Dås ‡åstri and Kålika Prasåd Shukla, Sudhi Prakashan, Varanasi, 1983-85, 6 voll. Någe†a, Br¢hacchabdendu†ekhara, edited by Sitarama Shastri, Sarasvati Bhavana Granthamålå, no. 87, Vårañaseya Sanskrit Vi†vavidyålaya, Banaras, 1960, 3 voll. Någe†a, Uddyota: see Pata∞jali’s Vyåkaraña Mahåbhåßya, Chaukhamba edition. 78 On the competition between Sanskrit and the ‘vernacular’ languages –both IndoAryan, Dravidian and of foreign origin, like Persian– in South Asia, see, most recently, Pollock (2003) and the contributions therein.

432

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Påñini, Aß™ådhyåyœ of Påñini, Roman Transliteration and English Translation, by Sumitra M. Katre, University of Texas Press, Austin (First Indian edition: Motilal Banarsidass, Delhi 1989). Pata∞jali, Pata∞jali’s Vyåkaraña Mahåbhåßya with Kaiya™a’s Pradœpa and Någe†a’s Uddyota, edited by B.B. Joshi et al., Chaukhamba Sanskrit Pratishthan, Delhi, 1987-88, 6 voll. Pata∞jali, Vyåkaraña-mahåbhåßya of Pata∞jali, edited by F. Kielhorn, 3 voll., Government Central Press, Bombay, 1880-85. [3rd edition: K.V. Abhyankar, BORI, Poona, 1962-72.] Studies Abhyankar, K.V., Limaye, V. P. (1965), Våkyapådœya of ‡rœ Bhartr¢hari, Poona, University of Poona. Aklujkar, A. (2002-2003), Where Do Lakßañaika-Cakßußka and LakßyaikaCakßußka Apply?, in «Bulletin of the Deccan College», 62-63 (= Ashok R. Kelkar Felicitation Volume), pp. 179-85. Cardona, G. (1999), Approaching the Våkyapadœya, in «Journal of the American Oriental Society», 119, pp. 88-125. Deshpande, M.M. (1985), Ellipsis and Syntactical Overlapping: Current Issues in Påñinian Syntactic Theory, Pandit Shripad Shastri Deodhar Memorial Lectures, (= BORIS 24), Bhandarkar Oriental Research Institute, Pune. _______ (1993), The Changing Notion of †iß™a from Pata∞jali to Bhartr¢hari, in «Asiatische Studien», 47, 1, pp. 95-115. _______ (1998), Evolution of the Notion of Authority in the Påñinian Tradition, in Histoire Epistémologie Langage, Tome XX, Fasc. 1, Les Grammaires Indiennes, pp. 5-28. Houben, J.E.M. (1992-93), Bhartr¢hari’s perspectivism (3): On the Structure of the Third Kåñ∂a of the Våkyapadœya, in «Sambodhi», 18, pp. 1-32. _______ (1995a), The Sa∫bandha-samudde†a (Chapter on Relation) and Bhartr¢hari’s philosophy of language: A Study of Bhartr¢hari’s Sa∫bandha-samudde†a in the Context of the Våkyapadœya with a Translation of Helåråja’s Commentary Prakœrña-Prakå†a, Egbert Forsten, Groningen. _______ (1995b), Bhartr¢hari’s perspectivism (2): Bhartr¢hari on the Primary Unit of Language, in Dutz, K.D., Forsgren K.-A. (eds.), History and Rationality: The Skövde Papers in the Historiography of Linguistics (= AUSSL 1), Nodus Publikationen, Mnster, pp. 29-62. _______ (1997), Bhartr¢hari’s perspectivism (1): The Vr¢tti and Bhartr¢hari’s perspectivism in the First kåñ∂a of the Våkyapadœya, in Preisendanz, K., Franco, E. (eds.), Beyond Orientalism: the Impact of the Work of W. Halbfass on Indian and Cross-Cultural Studies, (Poznan Studies in the Philosophy of the Sciences and the Humanities, 59), Rodopi, Amsterdam, pp. 317-358.

IV. Experiencing Boundaries within Tradition: The Case of the Sanskrit Grammarians 433

_______ (2002), The Brahmin Intellectual: History, Ritual and “Time out of Time”, in «Journal of Indian Philosophy», 30, pp. 463-79. Pollock, S. (2001) New Intellectuals in Seventeenth-Century India, in «The Indian Economic and Social History Review», 38, 1. _______ (ed.) (2003) Literary Cultures in History. Reconstructions from South Asia, University of California Press, Berkeley-Los Angeles. Whitney, W.D. (19245), Sanskrit Grammar, reprint: Motilal Banarsidass, Delhi 1983.

V. Violating Tradition and its Boundaries

FEDERICO SQUARCINI Traditions against Tradition. Criticism, Dissent and the Struggle for the Semiotic Primacy of Veridiction * 1. The voices of dissent as an essential element in understanding the textual production of an ‘orthodox’ tradition In a short essay written in 1990, Friedhelm Hardy drew attention to a singular gap in the field of Indological studies. Taking as his example the minor and marginal position to which certain South Asian religious traditions have been consigned, Hardy raised various thorny methodological issues. On the one hand, he indicated how many of the supposed monolithic ‘givens’ (as the ‘true Vedic life’) to which we are accustomed are such only by virtue of the history of the discipline. On the other hand, he argued that the part played by ‘unorthodox distortions’ and ‘sectarian developments’ has not been attributed suffi* With the technical term ‘veridiction’ (borrowed from the French véridiction, in itself a cognate of the medieval Latin term verdictum [from verumdictum or veredictum]) I wish to emphasise the pretension (or the veridic intention) which sustains statements claiming that a certain conviction, or persuasion, is the ‘real view’ or the ‘proper understanding’. This veridic act, a peculiar property of each nomos, is the prerogative of those juridical, religious and philosophical discourse, in which speakers wants to be perceived and acknowledged as ‘pronouncing truth’. I have chosen to use the word veridiction also because other English terms –such as truthfulness (‘the quality of being truthful’), truthful proposition, truth telling, truth claim (or truthful claim), true statement, or veridic discourse– do not properly express the political, relational and competitive dimension of this specific form of speech act. Aspect, this latter, not at all extraneous to the ancient South Asian context. See Thompson 1998. See, on various usages of the notion of ‘veridiction’, Fortier 1997; Landowski 1988; Coquet 1983; Greimas 1980. In this paper, all the quotations, translations and references to the Månavadharma†åstra are taken from the critical edition of Patrick Olivelle (Olivelle 2005), whom I would like to thank for having given me the opportunity to consult and use it before its publication. Then, too many things should be said to address the debts that I have accumulated during the compilation of this essay. I want to mention at least the names of those who directly helped me: Piero Capelli, Alessandro Graheli, Marina Rustow, Francesco Sferra, Lara Tavarnesi.

438

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

cient weight, and that their historical role may have been anything but secondary or marginal. With the penetrating originality characteristic of his work, Hardy outlined a short but intensive research program: Naively it has been assumed that what the dharmashastras lay down as rules corresponds to actual life. In fact, it is no more than an ideal, a blueprint for a perfect society. […] From this follows that a considerable amount of religious life has been going on that is not as such described in the books on the Vedic dharma. No doubt these books, along with the belief in the Vedas etc., played a far wider role as prestigious norms and ideals than such a hypothetical calculation reveals. But they prescribe, not describe. Once it is realized that things need not actually be what they are supposed to be, according to some normative interpretation, it becomes possible to look at them in their own right and not write them off as ‘unorthodox distortions’ or ‘sectarian developments’ when acknowledging their existence at all. Thus a vast realm is opened up for the study of a ‘Hinduism’ after and outside the Veda in senso stricto [sic].1

Hardy’s reasoning is attractive. Nevertheless, we should try to resist its appeal and seriously consider the problematics of his proposal. How can we argue that we should shift the study of certain social environments (Lebenswelten) from the centre to the margins without appearing to make arbitrary choices? What would be the reason for giving importance to such materials –or what remains of them– and removing them from their classic status of irrelevance? What could these contribute to our understanding of a certain historical event or a particular work belonging to a majority tradition? Is there not a risk of supporting instances of mere collection and compilation, or even factious intentions, concealed beneath the pretext of raising methodological problems? In working on the ‘unsaid’ is there not a danger that we may hear voices that have never actually been raised? Effectively, the perplexities related to the feasibility of such research programs are numerous. Nevertheless, there are valid arguments for considering that the withdrawal of materials produced by dissident and non-majoritarian traditions from their ‘statutory’ patch of shade need not be an act of mere antiquarianism. It can significantly increase both the quantity of new data, which it is the historian’s task to address, and the quality of the reading of all the other elements already present. Introducing unaccustomed information into an already existing framework requires space to be made and the rearrangement of the elements in play. The network of the relations and dynamics of the entire historical context has to be redesigned. 1 Hardy 1990: 147. Of course, Hardy is not the only one aware of these problems. Consider also what Witzel said: “It is often overlooked that the Hindu texts, whether in Sanskrit, Middle or New Indo-Åryan, or Dravidian, usually represent the voice of the Bråhmañ (the Buddhist, the Jaina, and so forth) establishment. Even originally countermovements (Tantra, for example) made their way into the received texts only after a period of ‘Bråhmañization’ –entailing purification and adjustment to the norms of Hinduism” (Witzel 1997b: 501).

V. Violating Tradition and its Boundaries

439

Thus, re-qualifying and shifting the epistemic status of neglected materials and traditions would lead to at least two results. While losing the marginal status of being a mere accessory fact, such neglected materials would also acquire the new status –corresponding more closely to reality– of an essential part of the formative dialectic of South Asian intellectual production. Unfortunately, because of a scholarly tendency to separate the parts from the whole –justified by the objective complexity of similar scenarios–, the tackling of such aspects has frequently been considered secondary, with the result of impoverishing, and therefore damaging, both the interpretation and the method. Here the history and sociology of intellectual approaches can come to our aid.2 Effectively, these disciplines show us how every thought or symbolic system produced by an agent, while being the expression of the individual’s agency, is also subject to and influenced by the various forms of constraint exercised upon him by the horizon of theoretical possibilities of which he is part. The elements of these constraints enter into dialectical connection with agent’s cogitatio, which is always socially situated and historically positioned, in the same way as the other conditions which enable its exercise. The effects of such links then penetrate the heart of the concepts, notions and categories which he uses. Hence the imaginative capacity of an agent must always be set in relation to the objective interests deriving from his being situated in a specific position within the social sphere (a place with contours that are always definite and never abstract), with all that this brings with it. The act and the outcome of intellectual production are therefore closely connected with the social position of the producer. Transferring these principles to our case means that the intellectual production of a majority tradition is not only connected with its social position, but is also greatly influenced by the historical specificity of its epistemic condition. This intellectual production takes place within a context of relative autonomy, which is the objective outcome of the competitive climate established between the various occupants of the social field, including its dissident traditions and marginalised subjects. A competitive climate, in which abilities and personal interests, the demands of the professional order or the parent corporation, the need –common to all corporations– to preserve the legitimacy of the existing objective and symbolic heritages, the processes of alliance or rupture which emerge from the interrelations of the various social entities, contribute to, and are responsible for, the morphological outcome of the various forms of intellectual production. Consequently, it is clear how important, if not actually essential, it is to seek out and listen to the dissenting voices. And not, as I said above, for reasons of mere par condicio or fascination with minorities. Listening 2 Like those at the basis of Collins’ work (Collins 2001), which is partially devoted to South Asia. Furthermore, Collins 2002; Bourdieu 1983.

440

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

to such voices, on the contrary, is essential if we are truly seeking to understand the reasons behind the theoretical structures that support the text of a dominant writer or the works of a majority tradition. 2. Tradition, criticism, subalternity: strategies for establishing the marginality of dissent Before starting to think this way, we must remove some of the structural faults of this method. We have to reconsider the epistemic implications underlying the marginal and liminal status to which the materials produced by the exponents of the various forms of ‘dissident’ or ‘anti-traditional’ intellectual activity have been consigned. I am referring, in particular, to our understanding of the reasons why many South Asian traditions that do not adhere to a dominant doxa have been omitted or given a permanent status of subordination. Here the paucity of textual material produced by the ‘dissidents’, which now survives in a few pages of fragments,3 is the most striking proof that over the centuries these elements have not been considered for what they really are. The first phase of this policy of omission and marginalization is linked to ‘traditional’ intellectuals (smårta), who have excluded these awkward arguments from their treatises, except when driven by the need to contradict dangerously persuasive concepts. Subsequently, the oversight was reinforced by those who shared the historiographic idleness of certain ‘traditions’ of Indological studies. The force of inertia imposed by the received view has led them to a wariness in the quest for strategic and political reasons behind the silence of the sources and in the related inquiry into the literary corpora transmitted by ‘hegemonic’ traditions. Thus the words of dissent, which the ‘orthodox’ (åstika) authors had acutely consigned to the margins of their cultural universe, acquired a permanent, almost ontological, status of oversight. The traditional desire to shift these intellectual traditions to the sidelines has led, on the one hand, to the conviction that they were indeed marginal and, on the other, to establishing the unified, intact, concordant and undifferentiated image of the majority tradition. This produced the dominant view –that is, exempli gratia, the epistemic foundation for the legitimation of the myth of sanatånadharma–,4 according to 3

See Bhattacharya 2002a.

4 Note, in fact, that the famous theme of sanåtanadharma enters Sanskrit juridical dis-

course from the Månavadharma†åstra. See Månavadharma†åstra, 4.138 (dharmaΔ sanåtanaΔ); 7.98 (dharmaΔ sanåtanaΔ); 9.64 (dharmaΔ hanyuΔ sanåtanam). The fact that it is not present in the previous dharmasütras indicates in part that new historical and political conditions nourish the author’s intentions. See, for other references to the condition of eternity of the norm and to other related themes, Månavadharma†åstra, 9.325; 10.7; 12.99. Another coeval work in which the theme of the sanåtanadharma appears several times is the Mahåbhårata. Further, for some considerations about modern revisiting of the notion of sanåtanadharma, Zavos 2001; Kaviraj 1999: 38-39; Halbfass 1990: 322; 341-346. Moreover, Goldman 1997; Heesterman 1985: 2.

V. Violating Tradition and its Boundaries

441

which the transmission of traditional knowledge has been carried out without interruption or disturbance, since not exposed to any real dialectical opposition. To undermine the plausibility of this unrealistic argument, we have only to consider the position adopted by certain ‘traditional’ authors in relation to the specialist tradition that preceded them. This clearly reveals that their degree of concurrence with and deference to the authorities was significantly more limited than the nonconflictual image of tradition would lead us to believe. Consider, for example, the strength of the opposition and the ‘anti-traditional’ reasoning of an author such as Kau™ilya.5 The same can be said of other fields and genres in which dialectic, innovation, and the assertion of specificity or subjectivity have played a decisive role in the establishment of a tradition.6 And if this was the case within an individual tradition, we can imagine what took place outside it, where the coercive influence of corporate interests and bonds of belonging was less relevant. Removing permanently such structural flaws from our method, however, demands that we distance ourselves from a whole series of stereotyped forms, foremost among them that which Olivelle neatly defined as “[...] the marvellous illusion of sanåtanadharma and changeless India”.7 Fortunately, today we are in the position to review this state of affairs, thanks to the efforts of various researchers and the recent introduction of new approaches to the understanding of the process of 5 The formula ‘Kau™ilya does not [say] so’ (na iti kau™ilyaΔ) –expressing an evident distance and opposition from the opinions of other people (opinions that are mostly attributed by the author to prominent personalities [åcårya] or, less frequently, to specific authorities [Manu, U†ana, Br¢haspati etc.], when not to previous teachings [iti upadi†anti])–, it is very frequently used in the work of Kau™ilya and it marks the argumentative and dialectical nature of its discourse. This formula is recurrently used within portions of the Artha†åstra (exempli gratia, Artha†åstra, 7.9.1-51; 7.12.1-25; 8.2.1-26; 8.3.1-61; 8.4.1-47; 9.1.1-34) dedicated to themes which are, from the political point of view, extremely important. See, for the occurrence of this formula in detail, Artha†åstra, 1.4.7; 1.15.32; 3.11.47; 3.14.7; 3.19.20; 3.20.5; 7.4.9; 7.5.4; 7.5.13; 7.9.10; 7.9.15; 7.9.20; 7.9.24; 7.9.28; 7.9.33; 7.9.51; 7.10.13; 7.11.14; 7.12.10; 7.12.15; 7.12.19; 7.12.23; 7.13.32; 7.15.16; 7.17.4; 8.1.12; 8.1.22; 8.1.28; 8.1.37; 8.1.46; 8.1.55; 8.2.6; 8.2.10; 8.2.14; 8.2.22; 8.3.13; 8.3.27; 8.3.34; 8.3.42; 8.3.52; 8.3.58; 8.4.3; 8.4.6; 8.4.10; 8.4.14; 8.4.17; 8.4.22; 8.4.25; 8.4.28; 8.4.32; 8.4.35; 8.4.39; 8.4.42; 9.1.6; 9.1.13; 9.1.32; 9.2.22; 12.1.6. Kau™ilya employs also a more neutral way –but not less assertive– of reinforcing his statements and judgements, using the formula ‘Kau™ilya [says] so’ (iti kau™ilyaΔ). See Artha†åstra, 1.2.8; 1.7.6; 1.8.27; 1.10.17; 1.15.50; 1.17.22; 1.17.30; 2.7.15; 2.9.12; 3.4.12; 3.4.36; 3.5.24; 3.7.3; 3.17.5; 3.17.14; 3.19.18; 5.6.23; 5.6.32; 7.1.5; 7.6.31; 7.11.38; 7.15.11; 9.1.43; 13.4.5; 15.1.22. Here we are clearly dealing with an intellectual practice which implies the critique of previous authorities and that, for obvious reasons, gained popularity, up to the point that Kau™ilya’s statements themselves have become the object of further criticisms. See, exempli gratia, Zimmermann 2000: 186-199. 6 This is clear if we consider the case of the authors of juridical texts: “They gave their texts a particular structure; they argued for particular positions in law and morality; they disagreed with other experts, both their contemporaries and their predecessors; and they had particular social, economic and political axes to grind. In all this they are not much different from modern authors” (Olivelle 2002: 536-537). See, for a theoretical proposal concerning the authorial condition in classical South Asian textual materials, Squarcini 2004. 7 Olivelle 1993: 246.

442

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

intellectual history and the relationships between the different fields of cultural production, even in the sphere of South Asian antiquities. Once the reasons for the oblivion to which the dissident traditions have been consigned are understood, we can then attempt to redesign the structure, the systemic typologies and the dynamics of intellectual production and transmission in South Asian antiquity, thus redeeming from an artificial status of marginality an agonistic aspect of cultural production which is intrinsically in no way marginal at all. 3. The struggle between ‘traditions’ for the semiotic primacy of veridiction My basic assumption is that the monopoly on veridiction, which all traditions claim to possess, is always achieved through the imposition of certain principles of epistemic vision and social division. This monopoly is a privileged status and can be achieved by a tradition only after passing over, assimilating, defeating, or even concealing the logical and historical legitimacy of its opponents and competitors. However, even if we accept the rationale of the model of semiotic conflict, talking about ‘traditions against the tradition’ is only effectively possible when we can proof the existence, at an historical level, of a set of agents that actually hold, or claim to do so, the monopoly on veridiction. To substantiate this necessary condition, we must analyse the tradition’s representations of itself, its claims to truth, and its pretensions to primacy. In other words, we have to gather evidence of the operation of concrete corporative forms that have benefited from the outcome achieved by an initial policy of knowledge governed by a desire for cultural supremacy. Fortunately, as regards the status of the Gangetic area in the centuries prior to our period, the existence and the scope of such a necessary condition has recently been corroborated by several studies, to which I would refer.8 Consequently, it is now historically plausible to speak of the existence of a wide-ranging cultural project that attempted to establish, canonise and regulate specific practices, cults, beliefs, visions, customs, institutions and values. If this is true, then we have satisfied the basic conditions and arguments that allow us to assume the existence of dialectic and dispute between the various bearers of different truths. Among these, those who prevailed then sought to subordinate, minimise, obstruct and disregard,9 the semiotic and cultural heritage of the others. From these premises, we can begin to collect, and then review, what remains of such conflict, seeking to understand how and to what extent the different parties were dialectically involved in the 8 See, exempli gratia, Wezler 2004: 643-646 (where the processes of ‘vedification’ and ‘brahmanization’ are discussed); Witzel 1997a; Witzel 1997c: 266-268; 293-296; 301-303; 307-313; Witzel 1995; Deshpande 1999; Pollock 2005a; Pollock 2005b; Pollock 1990. 9 See Piantelli 1996: 50; 53.

V. Violating Tradition and its Boundaries

443

internal struggles within the South Asian intellectual domain of their time. So we come to the ‘traditions against the tradition’ issue, which should thus be understood as a model in which different knowledgeproducing and knowledge-spreading agencies, immersed within a pluralistic climate of semiotic and political strife, competed against each other for the monopoly on veridiction, aiming at an objective establishment of their own forms of knowledge about the world. This model becomes even more plausible if, as Ramkrishna Bhattacharya has argued, in classical times words such as lokåyata referred to a sort of disputatio.10 Giving central importance to the practice of dialectics and dispute between different points of view appears now anything but out of place. On the contrary, this perspective allows us to re-examine a whole series of significant South Asian cultural phenomena, such as the strongly agonistic spirit of some Vedic poets11 or the structure of certain dialectical practices (such as vikalpa, apoha, vivåda, etc.). These practices were widely used and therefore achieved high levels of practical, theoretical and lexical formalisation.12 Thus, as Kaviraj has recently pointed out, we must attribute considerable importance to the fact that [t]he intellectual universe of Sanskrit possessed a culture of energetic, intense and vivid critical discussion, a culture that revelled in controversy; and precisely because it considered disputation so important in the search for truth, it devoted great attention to devising rules for rational prosecution of intellectual debates.13

If, however, we want to discover the reasons that allowed for the development of these intellectual practices, we should not confine ourselves to the speculative and rational aspects of cultural production, risking a banal accounting for everything by recourse to the ‘Indians’ love for specialised technicism’, but should rather look at the semiotic-political need to define reality. Effectively, if we look deeply into the social and economic conditions of ancient South Asia, we can find cogent reasons to conclude that the development, justification, institution and diffusion of such intellectual products could 10 See Bhattacharya 1998a; Bhattacharya 2000b. The word lokåyata is attested since Påta∞jali (Mahåbhåßya, 7.3.45) and Artha†åstra (Bhattacharya 1998b), but from the sixth century CE was largely used by epitomists, polemists and doxographers to indicate ‘materialist’ philosophy. See Bhattacharya 2002b. Further, on the denigratory depictions of lokåyata and cårvåka by the nine century’s author Jayantabha™™a, see Bhattacharya 2002c. 11 Emblematic is the poetic dispute (vÌjina) within the college of Vedic poets, where the revelation of the ‘connections’ (bandhu) was comprised within a context of serious objective interests. See, exempli gratia, R¢gveda, 10.35.1-14; 10.71.1-11 (on poetics agonism); 10.81.1-7; 10.166.1-5 (on the victory on colleagues). Further, Oguibénine 1998: 158-159; Gonda 1984: 1-67; Gonda 1975: 79-82; Durante 1976: 194-201. 12 See, on the practices connected to the dialectical debate and on its various implications, Matilal 1998: 31-87; Solomon 1976-78; Stcherbatsky 2000: 457-505. 13 Kaviraj 2005: 125.

444

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

not come about merely for reasons of enjoyment. These practices, extremely demanding and complex, were necessarily stimulated and sustained by both logical and political reasons. A very good example, among others,14 is found in one of the seven treatises of the Buddhist Abhidhammapi™aka. The work, eloquently entitled Kathåvatthu15 –probably written during the reign of A†oka,16 and hence in a situation characterised by marked political and ideological tensions– is entirely devoted to a guided exposition of dialectical practice. That subsequent Buddhist authors felt the need to ascribe the structure of the Kathåvatthu to Buddha himself only confirms the importance they attributed to the proper conduction of a dispute.17 Buddha thus plays the role of guarantor and custodian of the principles of argumentative dialectics which have to be observed during the search for the semiotic primacy on veridiction. As recently stated by Jonardon Ganeri, [t]he Kathåvatthu or Points of Controversy is a book about method. It describes, for the benefit of adherents to various Buddhist schisms, the proper method to be followed in conducting a critical discussion into an issue of doctrinal conflict.18

Nothing shows the strategies behind the intellectual vitality of South Asia in the centuries previous to our era better than this interweave of veridical discourse, method of dispute, doctrinal politics and ideological interests.19 It is the inherent vitality of a context in which the agonistic procedures of dispute and opposition between different conflicting forms of veridiction were granted the power to define the arrangement of the social space occupied by the exponents of different points of view. It is here that the theme of the recovery of the agonistic context in which, since the Vedic era,20 the practices of intellectual production have taken place merges with the need to imagine, and then identify, the actual profiles of the opponents, which are not recorded 14 Within Buddhism a very important antecedent, indeed, is constituted by the four mahopadesas. See Yang Gyu An 2003; Davids 1977. 15 See Kathåvatthu (Taylor 1894; 1897). 16 See Barua 1937: 367-370. 17 See Buddhaghosa, Atthasålinœ (in PTS, p. 8). 18 Ganeri 2001: 485. 19 There is no doubt that these conditions are not exclusive of this period. In primis, we can refer also to the important work of Vasubandhu (c. 360 CE), named Vådavidhi. See, for the edition and translation of its remaining fragments, Frauwallner 1957; Tucci 1928. Further, Rangaswami Iyengar 1953. Then, the work that Dharmakœrti dedicated to the logic of the debate is another significant indicator of the prosecution of such a spirit. See Dharmakœrti, Vådanyåya (Gokhale 1993). 20 I’m refering to the existence of ‘deniers of gods’ in the R¢gveda (exempli gratia, R¢gveda, 2.25.5; 8.100.3-4 [where the god Indra is denied]) and to those who deny the ‘divine’ status of the Veda. See Ronzitti 2001. Another example of this attitude are Kautsa’s words cited in Yaska, Nirukta, 1.15-16. Further, but related to other period, Slaje 1998.

V. Violating Tradition and its Boundaries

445

in detail in the accounts and productions of the traditions that attained primacy. Revealing, in this regard, is the exploration of the way in which the sources connect the relation between description of the argumentative dialectic and prescription of the moral judgement. The more strictly theoretical treatises technically codified the dialectic method in line with precise forms and roles –later codified in various forms of debate (våda, jalpa, vitañ∂å, etc.).21 Elsewhere, instead, the desire to usefully employ the semiotic power derived from the success achieved in the dialectic conflict pursued the path of deprecation of the adversary’s moral status, seeking to support the theoretical dimension with the strength of the practical reasons. It is no coincidence that the moralisation of differences and opposing positions has always been the preferred approach of those who promoted specific symbolic and ethical construction of social reality, strongly marked by hegemonic claims.22 This is also why Brahmanic sources describe their adversary interlocutors with disparaging terms, promoting a negative picture of their values and moral qualities. They are divided into ‘opponents’ (vipratipakßin, prativådin, pürvapakßin), ‘skeptics’ and ‘dialecticians’ (tårkika, hetumat, hetuvådin), and ‘disputants’ (vivådåtmaka), but they are also presented as ‘bearers of false visions’ (mithyådr¢ß™in), ‘polemicists’ (vaitañ∂ika), ‘deniers’ (nåstika) and ‘materialists’ (bhütavådin). Some of these epithets are morally pregnant marks of real disputes, whose real protagonists are opposing social agents –although their morphology is often mystified in the texts. They act dialectically within a specific domain and are driven by opposed interests, in conflict with each other and seriously involved in the contest for the primacy on veridiction. It is clear that they do not debate simply out of vis polemica. Some dissidents, for example, are perfectly aware that demolishing a given belief, or an axiomatic system, provokes an epistemic rupture that, in its turn, generates other epistemic breaks and destabilises the whole system of representation. In other words, when a ‘denier’ (nåstika) claims that one must attribute primacy and ‘priority’ (jyeß™hatå) only to ‘sense perception’ (pratyakßa) he asserts that the scriptures (ågama) are not a valid means of knowledge (pramåña).23 By saying so he induces others to think that what the auctoritates declare about 21 These notions are among the constitutive elements of the epistemological structure of the initial nyåya knowledge system. See Gautama, Nyåyasütra, 1.2.1-20; 1.1.32-41. 22 See Bourdieu 1991: 5-13. 23 This is briefly what is said in one of the oldest fragments of the ‘materialist’ philosophy preserved –included within the conversation on the liberation of Janaka, king of Mithila (Mahåbhårata, 12.211.1-48)–, where the lack of real evidence of the existence of an entity separate from the body is discussed as a sound reason not to believe in the authority of the ‘scripture’ (ågama). See Mahåbhårata, 12.211.21-45 (in part. 12.211.26-27: pratyakßa∫ hy etayor müla∫ kr¢tånta iti hy ayor api Ù pratyakßo hy ågamo ’bhinnaΔ kr¢tånto vå na ki∫cana ÙÙ yatra tatrånumåne ’sti kr¢tam bhåvayate’pi vå Ù anyo jœvaΔ †arœrasya nåstikånå∫ mate smr¢taΔ ÙÙ [Critical ed.]). The †loka following this one is ascribed to the nåstika in the commentary of Nœlakañ™ha.

446

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

other aspects of real life is also unfounded. Such polemic attitude, if not disproved, risks assuming the features of a more general ‘doctrine of the negation of the Veda as a valid means’ (vedapråmåñyanißedhavåda).24 This is the type of threat most feared by ‘orthodox’ (åstika) traditions,25 such as that championed by the author of the Månavadharma†åstra. In this regard, the arguments used to stigmatise the critical approach mentioned above are emblematic: 10. ‘Scripture’ should be recognized as ‘Veda’, and ‘tradition’ as ‘Law Treatise’ [dharma†åstra]. These two should never be called into question in any matter, for it is from them that the Law has shined forth. 11. If a twice-born disparages these two by relying on the science of logic, he ought to be ostracized by good people as an infidel [nåstika] and a denigrator of the Veda [vedanindaka].26

In the wake of these threats, other exponents of the ‘orthodox’ tradition later declared that the primary intention of logical reasoning (nyåya) was to ‘protect the validity of the Veda’ (vedapråmåñyarakßå).27 It is evident that this is not just a struggle for knowledge but also –or perhaps mainly– a struggle for symbolic power and primacy of veridiction. The object of such a dispute is the power of having the right to speak, which consists in the possibility of formulating veridical propositions legitimately. These propositions, owing to the link between the performative force of the speech-acts and the political interests of establishing certain sets of meanings, have the power to give knowledge reality. The irenic image of the struggle for knowledge could still make some sense in relation to intellectual practices such as ‘philosophical scepticism’,28 even if it would be reductive to see them as devoid of non-theoretical interest, since –extending Randall Collins’ explana24 We have evidence of this attitude cited by its detractors. The following aphorisms ascribed to the cårvåka or lokåyata are indicative of such a position: ‘let the rites of dharma not to be practiced’ (na dharmå∫† caret) in Våtsyåyana, Kåmasütra, 1.2.21-24 (21. na dharmå∫† caret. eßyat phalatvåt. sa∫†ayikatvåc ca Ù 22. ko hy abåli†o hastagata∫ paragata∫ kuryåt Ù 23. varam adya kapotaΔ †vo mayüråt Ù 24. vara∫ så∫†ayikån nißkåd aså∫†ayikaΔ kårßåpañaΔ Ù iti laukåyatikåΔ Ù); ‘let dharma not to be done’ (dharmo na kåryaΔ) and ‘let its instructions not be considered trustworthy’ (tad upade†eßu na pratyetavyam) in Jayantabha™™a, Nyåyama∞jarœ, åhnika 4 (ed. Varadacharya 1969: 647-648). 25 One of the traditional etymologies of the term –based on Påñini, Aß™ådhyåyœ, 4.4.60 (astinåstidiß™am matiΔ)– says ‘he whose opinion is that œ†vara exists’ (asti œ†vara iti matir yasya). Other definitions include ‘opposite of nåstika’ (nåstika bhinna); ‘he whose idea is that œ†vara exists’ (œ†vara asti iti vådœ); ‘he who considers the Vedas as authorities’ (vedapråmåñyavådœ). Further, according to Hemacandra, åstika is a synonym for ‘he who believes’ (†raddhålu) and for †raddhå. See Råjårådhåkånta 1961: s.v. 26 Månavadharma†åstra, 2.10-11. Further, on these ‘offenders of the Veda’ (vedanindaka), Månavadharma†åstra, 2.11; 3.161; 4.163; 11.57. 27 See Jayantabha™™a, Nyåyama∞jarœ, åhnika 1 ([p. 7] nyåyavistaras tu mülastambhabhütaΔ sarvavidyånå∫ vedapråmåñyarakßåhetutvåt Ù); åhnika 1 ([p. 11] yasya hi vedapråmåñye sa∫†ayånå viparyastå vå matiΔ, ta∫prati †åstrårambhaΔ Ù) (ed. Varadacharya 1969). 28 The analysis of the role played by skepticism and by systematic doubt within South

V. Violating Tradition and its Boundaries

447

tory model to this sphere–29 these too can no longer be represented as mere philosophical approaches. But this argument loses strength in the face of more ‘committed’ forms of dissent. The ‘materialistic’ discourse connected with the figure of Br¢haspati, for example, was not just a sort of intellectual and doctrinal distancing from a given point of view, but rather a clear ethical and moral rejection, through which certain individuals reacted to a given social condition. It’s fairly evident that certain dissenters voiced their criticisms against cultural orders that they considered as oppressive and harbouring hegemonic pretensions. Their main objective was to reveal the unsoundness of the theoretical basis of certain ethical-regulatory structures. This is what can be gathered from the intentions of certain ancient theoreticians at the forefront of ‘materialist’ traditions, such as Br¢haspati.30 Indeed, “B®haspati’s assertion that inference is not a means of valid cognition had a pragmatic purpose: to cut off any divine or supernatural factor from the foundation of social and ethical theories”.31 A similar critical approach can be found in the specialised domain of Buddhist pramåña tradition, with the same sceptical and polemical stance toward the Brahmanic social vision.32 If the dispute among different traditions over the monopoly on veridiction had not had non-theoretical foundations, South Asia would never have experienced the extraordinary intellectual ferment of its writers, who took part in lengthy and heated debates with clear monopolistic intentions. A good example of this tension is the secular dispute on the criteria to be adopted to discern the validity of the ‘means of knowledge’ (pramåña). What has been developed as the ‘discourse on reliable warrants’ (pramåñavåda), indeed, was at the centre of a heated debate that lasted for centuries. Through the production of several specialised works, originating from the most diverse ideological and religious convictions, the criteria for establishing the authority of the sources of knowledge were subjected to meticulous scrutiny. Consider, for example, the long-standing debate –which generated an immense literary production– that can be traced to the idea of pramåña presented in the work of Bhartr¢hari (450-510 CE) and extends through the Pramåñasamuccaya of Dignåga (480-540 CE), to the Pramåñavini†caya and the Pramåñavårttika of Dharmakœrti (600660 CE) –an important text which has given rise to numerous commentaries–,33 to the Pramåñalakßaña of Sarvaj∞åtman (c. 1027 CE),

Asian intellectual traditions has recently obtained new contributions. See Bhattacharya 2000a; Butzenberger 1996; Franco 1994; Bhattacharyya 1987; Matilal 1985; Franco 1984; Franco 1983; Gupta 1981. 29 See Collins 2002; Collins 2001: 177-271. 30 See Namai 1976. 31 Franco 1983: 148. Further, Franco 1994: 8; 46-47; Bhattacharya 1997. 32 See Eltschinger 2000: 33-60; 83-86. 33 Such as the Pramåñavårttika™œkå of ‡åkyamati (c. 710 CE), the Pramåñavårttika™œkå of

448

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

the Pramåñåntarbhavaprakaraña of Ratnakœrti (c. 1070 CE), the Pramåñanayatattvålokåla∫kara of Vådi Devasüri (1087-1150 CE), the Pramåñamœmå∫så of Hemacandra (1089-1172 CE), the Pramåñama∞jarœ of Sarvadeva (c. 1200 CE), the Pramåñalakßaña by Madhvåcårya (1238-1317 CE), the Pramåñapaddhati of the famous dvaitavedånta™œkåkåra Jayatœrtha (1335-1385 CE), the Pramåñasundara of Padmasundara (c. 1550 CE), the Pramåñapramoda and the Pråmåñyavåda of Hariråma Tarkavågœ†a (c. 1640 CE), up to the later Prameyaratnåvalœ of Baladevavidyåbhüßaña (c. 1780 CE). This is to mention only a fragment of the vast number of treatises and texts –the list of which would take many pages– produced in the attempt to ground and legitimise distinct theoretical constructs through the favour granted to certain sets of pramåña.34 But behind the technicism of the ‘discourse on reliable warrants’ (pramåñavåda) nestles the much more thorny question of authority, which inevitably brings into play issues of the sociology of knowledge. This has been clear since the origin of the notion of pramåña. While investigating one of the first occurrences of the term, Wezler recently defined pramåña as “authority as the relevant instance” (Wezler 2004: 638). And it is precisely this placing of the notion of pramåña in close relation to the problem of authority which brings to the fore the agonistic climate underlying the debate on the definition of pramåñas. It is therefore, to all effects, a dispute on veridiction, which clearly exemplifies the difficulties faced by anyone who wants to raise his own criteria to super partes principles. This became evident with the turn which the discussion on the pramåñas assumed under the pressure of Buddhist authors such as Dharmakœrti or Dignåga, who introduced the notion of ‘one who is (like) a pramåña’ (pramåñabhütapurußa), so as to invest particular persons with full veridical authority, leaving them the last word about the correctness of modes of knowing.35 Such interested agonistic usage of the notion clearly shows how contemporary research into the types and history of pramåña must now avail of interpretative models borrowed from the sociology of knowledge, unless we wish to keep repeating that “[i]t is not absolutely clear what theories of pramåña were meant for” (Oetke 2003: 199). All this confirms the fact that every forma mentis exists, and can be experienced, starting from a tangible, objective and incarnate modus vivendi. This is the concrete dimension that gives strength to the ‡a∫karananda (c. 800 CE), the Pramåñavårttika™œkå of Jina (c. 920 CE), the vr¢tti of Manorathanandin (c. 950 CE), or the Pramåñavårttikåla∫kåra of Praj∞åkaragupta (c. 750-810 CE). 34 See, in order to explore the main contexts in which the pramåñavåda developed, Balcerowicz 2005; Phuntsho 2005; Kataoka 2003; Kellner 2003; Oetke 2003; ColasChauhan 2002; Arnold 2001; Krasser 2001; Houben 2001; Houben 1999; Gokhale 1993; Taber 1992; Potter 1992; Dreyfus 1991; Sarma 1989-1990; Aklujkar 1989a; Aklujkar 1989b; Hegde 1982; Bandyopadhyay 1979; Bhattacharya 1977. 35 See Silk 2002; Krasser 2001; Franco 1997: 15-43; Seyfort Ruegg 1995; Seyfort Ruegg 1994; Tillemans 1993.

V. Violating Tradition and its Boundaries

449

abstract. And it is precisely because of such osmosis that intellectual postures and positions occupied within the social sphere need to be seen as closely connected. 4. Threats to tradition: getting to the heart of the intellectual history of traditions against the tradition If we accept the approach presented so far, then the presence of agonistic elements, otherwise considered marginal or incidental, become essential for understanding any phenomenon belonging to a given intellectual context. This makes the study of traditions against ‘the tradition’ in South Asia a question of method. Consequently, we have to tackle the theme of the dialectics between opposing parties through a more conscious and analytically justified scrutiny of the binding relation between the social context (made up of different specific spheres) to which each individual –or group of individuals– belongs and the imaginative and intellectual possibilities that they consequently have at their disposal.36 It is in this multifaceted and pluralistic environment that we have to set the struggles for the semiotic primacy of veridiction, where specific historical ‘traditions’ are counterpoised with other ‘traditions’ threatening their monopoly on practical and theoretical truths. Recent studies on the numerous ancient urban settlements of the Gangetic area –which, during the firsts centuries BCE, were economically active and engaged in a flourishing trade–37 offer a new context,38 within which, alongside the representatives of established ‘orthodox’ traditions (åstika), we can acknowledge the agonistic presence of ‘heretics’ (påßañ∂in), ‘deniers’ (nåstika), ‘thinkers’, ‘logicians’ (hetuvådin, tarkin), and ‘offenders of the Veda’ (vedanindaka), who repudiated certain forms of dharma (dharmavidveßin), denied the authority of the Veda (apråmåñya∫vedånåm) and sought defects in the scriptures (†åstradoßånudar†in).39 This dissent regarding dharma represented a real threat to ‘orthodoxy’ and, consequently, led to the inception of new forms of epistemic regulation. 36 Various theoretical scenarios can strongly facilitate this new critical interpretation of the two levels of the doxological discussion. An example is Bourdieu’s disquisition on the relation between doxa, orthodoxy and heterodoxy. See Bourdieu 2000; Bourdieu 1988b. Further, Berlinerblau 2001. Moreover, on the same topic but related to South Asian materials, Eisenstadt, Kahane, Shulman 1984. 37 See Chakravarti 2005; Lahiri 1999: 268-399; Champakalakshmi 1996. Further, on the situation south to the Gangetic area, Morrison 1995. 38 The levels of social diversification and the structure of administrative systems reached in the northern area of South Asia in the first centuries BCE were remarkable. The same is true for the development of forms of local ‘state’ (råß™ra). See Witzel 2003: 72111; Thapar 2002: 137-164; Sharma 2001: 15-20; 82-96; Allchin 1995: 73-151; Kane 1993: vol. 3, 132-177; Scharfe 1989: 127-227; Sharma 1989; Erdosy 1988. 39 See, for some analyses of these individuals and of the various forms of dissent, conflict, and ‘heresy’ in ancient South Asian regions, Ronzitti 2001; Thapar 2000b; Thapar

450

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

The pluralism generated by the development of urban life was undoubtedly a source of complication and proliferation of the possibilities of understanding the world. This is a pluralism which has aroused the attention of doxographers and authors of treatises since its mention in an important edict of A†oka (Pillar Edict, VII). It is the existence of these different schools and traditions that led to the production of a variety of definitions of dharma –a variety that cannot be truly accepted if the normative value of dharma is to be maintained.40 For example, after discussing ‘heretical’ and non ‘heretical’ doctrines for four chapters,41 a jaina text –which together with the Åcåråõga is the oldest part of the †vetåmbara canon (dating to the third or second century BCE)– refers to the existence of as many as 363 schools of thought: 180 of them are listed as kriyåvådin (believing in the force of the act), 84 as akriyåvådin (rejecting the force of the act), 67 as aj∞anika (asserting that they do not know, sceptics) and 32 as vainayika (believing in salvation through good deeds).42 Something similar is expressed in the section of the Såma∞∞aphalasutta of the Dœghanikåya,43 where six different teachers and heads of schools are mentioned (Püraña Kassapa, Makkhali Gosåla, Ajita Kesakambalœ, Pakudha Kaccåyana, Sa∞jaya Belatthiputta, Nigañ™ha Nåtaputta [in the Brahmanic sources these are all considered as initiators of ‘heretical’ movements]), who acted as temporary teachers of Ajåtasattu, the famous monarch in the capital of Magadha.44 We can therefore agree with what Wilhelm Halbfass asserts when he claims: [t]he increase of reasoning and argumentation appears as a symptom of the Kaliyuga, the worst world period, and of the religious degeneration by which it is accompanied. Out of ‘fear of the science of reasoning’ (tarkavidyåbhaya), people seek refuge with the sources of tradition, with the Upanißadic ‘revelation’.45

If this is even partially true, it becomes easier to understand the reasons for certain representations given by ‘orthodox’ traditions, 2000a; Hudson 1998; Ballanfat 1997; Eisenstadt, Kahane, Shulman 1984; Franci 1975: 93-141; Morton Smith 1974; Morton Smith 1977; Doniger 1971; Tucci 1971: 49-155; Nakamura 1961. 40 The existence of pluralistic and agonistic dimensions that conditioned the definitions of dharma is very well documented in a collection of essays recently edited. See Olivelle 2004. About the anxiety for the presence of too many dharmas, see, exempli gratia, Mahåbhårata, 14.48.14-17; 12.135.1-171. 41 See Sütrakr¢taõga, 1.1.1-4. 42 See Sütrakr¢taõga, 2.2.79. 43 See Dœghanikåya, 2 (såma∞∞aphalasutta).1-33 (in PTS vol. 1, p. 47-59). 44 See Basham 1951: 11-18. 45 Halbfass 1990: 279. Actually, the fear that the spread of similar way of reasoning can damage the ‘Vedic’ system of belief remained for centuries, up to the time of Medhåtithi. See Medhåtithi, Manubhåßya ad Månavadharma†åstra, 12.106 (ed. Jha, vol. 2, p. 485) (tarkeñeti tarkapradhånå granthå laukikapramåñanirüpañaparå nyåyavai†eßikalokåyatikå ucyante Ù tatra vedaviruddhåni bauddhalokåyatikanairgranthådœni paryudasyante Ù).

V. Violating Tradition and its Boundaries

451

faithful to what is ‘recalled by tradition’ (smårta). In such sources the representatives of ‘other’ traditions are perceived and presented as a threat to a given order, while their discourses are seen as an open delegitimisation of the ideal social scenario (varñå†ramadharma). The dissenters were not just seeking out or introducing ‘flaws’ (doßa) into the ‘revealed’ scriptures (†ruti)46 and inciting sovereigns to deviate from the dictates of the authoritative texts (lubdhasyocchåstravartinaΔ).47 They also denied the existence of both the other world (paraloka)48 and the gods (devatå, œ†vara). They even contested the existence of an entity (åtman) that survives the death of the body, which is instead the only reality,49 thus weakening the foundations of the system of ethical government which bound the given doctrine of the retribution of actions (karmaphala)50 to that of rebirth (sa∫såra). Moreover, even in the oldest dharmasütras, we find references to individuals who would ‘attack or profane the Veda’ (vedaviplåvaka).51 46 See, exempli gratia, Mahåbhårata, 13.24.70 (vedånå∫ caiva düßakåΔ); 13.25.8 (yaΔ pravr¢ttå∫ †ruti∫ samyak †åstra∫ vå munibhiΔ kr¢tam Ù düßayaty anabhij∞åya ta∫ vidyå brahmaghåtinam ÙÙ). 47 See Månavadharma†åstra, 4.87. 48 See Medhåtithi, Manubhåßya ad Månavadharma†åstra, 4.30 (ed. Jha, vol. 1, p. 342) (haitukåΔ nåstikåΔ Ù nåsti paralokaΔ, nåsti datta∫, nåsti hutam ity eva∫ sthitapraj∞åΔ Ù). 49 An example in this regard is reported in Dœghanikåya, 23 (påyåsisuttanta). 1-34 (in PTS vol. 2, pp. 316-357), of which we have also a sort of jaina ‘recension’. See Bollée 2002. 50 In fact, the consolidation of this important causal nexus between ‘action’ and ‘result’ is one of the objectives to which the Brahmanical world at the time devoted much attention. See, in reference to the crucial ethical role played by this pair, Månavadharma†åstra, 12.82 (“82. I have declared to you above all the fruits arising from actions. Listen now to these rules of action for a Brahmin, rules that secure the supreme good”). See also Månavadharma†åstra, 4.70; 11.230-231; 12.1; 12.3; 12.81-82. An exemplary proposal for the redemptiveness of this constraining ‘law’ can be found in Bhagavadgœtå, 2.47 (karmañy evå’dhikåras te må phaleßu kadåcana Ù må karmaphalahetur bhür må te saõgo ’stv akarmañi ÙÙ). It should also be noted that nearly the entire twelfth adhyåya of Månavadharma†åstra (12.1106) is dedicated to the relationship between ‘action’ and ‘result’. According to this ethical model, the act exerts a univocal causal force (called karmahetu in Månavadharma†åstra, 1.49) whose results (phala) are sufficient to recognize the cause that produced them. On the notion of ‘results’ (phala), see Månavadharma†åstra, 1.109; 2.157-158 (where it is used as a criterion of proof in analogy); 3.95 (mentioned as sign of the obtainment of merit); 3.128; 3.142-143 (again, mentioned as a criterion of proof in analogy); 3.176-178; 4.172-173; 6.97; 7.86; 7.144; 7.208 (case of unwanted results); 9.51-54 (an exemplar use of the metaphor ‘seed-fruit’ to confirm the analogy ‘action-result’); 11.5. For a panoramic view of the complexity of defining the relations between ‘actions’ and ‘ends’, see Bronkhorst 2000b. 51 See Gautamadharmasütra, 20.1; Vasiß™hadharmasütra, 15.11; Månavadharma†åstra, 11.198 (here viplåvya indicates the improper use of the Veda). Is interesting to notice here, while talking about vedaviplåvaka, that one of the very few skeptical and ‘heretical’ treatises that we know, the Tattvopaplavasi∫ha –ascribed to Jayarå†i and dated at about the eighth century CE–, includes in its title the term upaplava (which, too, derives from the root plu-, that, besides ‘confusion’, means ‘doubt’, ‘disaster’, ‘calamity’). Individuals who ‘damage dharma’ (dharmaviplava) are also mentioned in an interesting satirical work of Jayantabha™™a (850-910 CE). See Ågama∂ambara, 3.36 (ye tu prastutadharmaviplavakr¢taΔ påpås tapopåyinas te ced å†u na yånti ghåtayati tån dasyün iva kßmåpatiΔ). Further, on the improper use of reasoning (tarka) to the detriment (viplåva) of Veda’s authority, Jayantabha™™a, Nyåyama∞jarœ, åhnika 1 ([p. 7] vedeßu hi dustårkika racitakutarkaviplåvitapråmåñyeßu †ithilitåsthåΔ katham iva bahuvittavyayåyåsådisådhya∫ vedårthånuß™hånam ådriyeran sådhavaΔ Ù) (ed. Varadacharya 1969).

452

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

In this perspective, we can see why these historical figures were seen as bearers of intellectual attitudes and practical conduct that had to be openly opposed, up to the point of comparing them to other human qualities that were more clearly damaging.52 They were invariably denigrated, stigmatised and strongly condemned,53 as can be inferred from what Bhœßma said to Yudhiß™hira: “sellers of the Vedas, corrupters of the Vedas, and those who write the Vedas, these surely go to hell”.54 There can be no doubt, therefore, that such individuals, belonging to intellectual traditions considered menacing by the dominant view, defined dharma in ‘their way’, propagated teachings, gathered together in groups, established practices and produced texts designed to spread their vision of the world. It is precisely to this production that the ‘orthodox’ sources refer, putting their readers seriously on their guard against such ‘non traditional’ materials: 95. The scriptures that are outside the Veda [yå vedabåhyåΔ †rutayo], as well as every kind of fallacious doctrine [kudr¢ß™ayaΔ] –all these bear no fruit after death, for tradition takes them to be founded on Darkness. 96. All those different from the Veda that spring up and then flounder –they are false and bear no fruit, because they belong to recent times. (Månavadharma†åstra, 12.95-96)

Their ‘untrue’ (asat) treatises (†åstra) are effectively listed among the things that the prudent bråhmaña must avoid: 66. Remaining without establishing the sacred fires; acting like a woman; non-payment of debts; studying fallacious treatises [asacchåstrådhigamana∫]; living a corrupt life; engaging in vices. (Månavadharma†åstra, 11.66)

Certain Brahmanic sources, charged with vindicatory spirit and intent on undermining the reputation of the dissident schools, associated the study of such ‘fallacious’ literature with the sterile use of ‘logical-argumentative reasoning’ (yukti, buddhir ånvœkßikœ, ånvœkßikœ tarkavidyå, hetu†åstra), here significantly counterpoised to the canonical and more certain reference to the scriptural tradition.55 52 See, exempli gratia, Månavadharma†åstra, 4.163 (nåstikya∫ vedanindå∫ ca devatånå∫ ca kutsanam Ù dveßa∫ stambha∫ ca måna∫ ca krodha∫ taikßñya∫ ca varjayet ÙÙ). 53 On the tendency of such intellectual stances to denigrate, see Menon 1987-88; Sharma 1979-80; Hopkins 1925. 54 Mahåbhårata, 13.24.70 (vedavikrayiña† caiva vedånå∫ caiva düßakåΔ Ù vedånå∫ lekhakå† caiva te vai nirayagåminaΔ ÙÙ). For an entire section of the text devoted to those who, because of their transgression of the †ruti, deserve to go to the hellish realm (niraya), see Mahåbhårata, 13.24.59-8o. Forewarning those who hold unorthodox ideas and engage in unorthodox practices about the possibility of going to hell (naraka), or being destroyed (vinå†a), is a standard custom of classical Brahmanical texts. See what it is said to lokåyata –and about their parentage– in Bårhaspatyasütra, 2.29-30 (laukåyatiko mr¢to bhavatyarthakåmadharmamokßavihœno nårakœ ca ÙÙ kule ca tatkula∫ tatputrapautråntare vina†yati ÙÙ). 55 See Månavadharma†åstra, 2.10-11; 12.95-96; 12.105-106.

V. Violating Tradition and its Boundaries

453

Starting from various references to a generically defined Br¢haspatisütra (a text which has not come down to us, generally attributed to the mythic Br¢haspati founder of the ‘heretical’ doctrines), within the ‘orthodox’ literature we can find various mentions of works by dissidents and materialists (lokåyata†åstra, nåstika†åstra).56 Unfortunately, no complete versions of these ‘heretical’ works reached us, but only fragments, mostly surviving within the ‘orthodox’ texts57 –with the sole exception of the late, but extraordinary Tattvopaplavasi∫ha, probably written in the eighth century CE.58 However, we do have some ‘historical’ names of the representatives of these intellectual traditions, which, frequently intermingled with the names of the mythical founders of sceptical-materialistic thought (such as Br¢haspati, Suraguru, Cårvåka), have been known to the academic world since Henry Thomas Colebrooke gave his memorable lecture ‘On the materialist tradition in India’ in 1827 at the Royal Asiatic Society of Calcutta.59 We also know of the existence of ‘personalities’ such as Aviddhakarña, Kambalå†vatara, Purandara (who were discovered thanks to the Tattvasa∫graha treatise by ‡åntarakßita and the Tattvasa∫grahapa∞jikå by Kamala†œla), Bha™™a Udbha™a, Bhåvivikta (known through the Granthibhaõga by Cakradhara), through to the decidedly more ‘concrete’ figure of Jayarå†ibha™™a (c. 770-830 CE). The struggle for the primacy on veridiction, however, did not only involve marginal sceptics and materialists. There was also great rivalry between more mainstream traditions, such as jaina and bauddha, and the ‘orthodox’ world based on the Veda. Let us take as an example the case of ancient Buddhism. 5. The bauddhadharma as a challenge to the ‘tradition’ Early Buddhism can definitely be included among those traditions of thought and practice that challenged the monopoly of the Brahmanic tradition, with its claims of holding the primacy on symbolic practices and on regulatory social veridiction. To different extents and with varying intensity, Gautama Buddha and some of the related traditions which immediately followed him, devoted themselves to a meticulous criticism –that sometimes appear as a quest for renewal of 56 See, exempli gratia, Månavadharma†åstra, 12.95 (vedabåhyåΔ †rutayo); Br¢haspatidharma†åstra, 1.1.64 (nåstika†åstra); Medhåtithi, Manubhåßya ad Månavadharma†åstra, 2.11 (ed. Jha, vol. 1, p. 72) (hetu†åstra∫ nåstikatarka†åstra∫ bauddhacårvåkådi†åstra∫ yatra vedo ’dharmåyeti punaΔ punar udghußyate Ù); Medhåtithi, Manubhåßya ad Månavadharma†åstra, 12.106 (ed. Jha, vol. 2, p. 485) (tarkeñeti tarkapradhånå granthå laukikapramåñanirüpañaparå nyåyavai†eßikalokåyatikå ucyante Ù tatra vedaviruddhåni bauddhalokåyatikanairgranthådœni paryudasyante Ù); Kr¢ßñami†ra, Prabodhacandrodaya, 5.5-6; 5.10 (påßañ∂ågama, påßañ∂åtarka†åstra). Further, Tucci 1971: 133-135; Dasgupta 1988: vol. 3, 515-516. 57 See Bhattacharya 2002a; Bhattacharya 1999. 58 See Franco 1994: 9-15. 59 See Colebrooke 1837: 402-405.

454

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

the ‘true’ Brahmanic tradition (Lindtner 1998; Bhattacharya 1973)– of the ‘orthodox’ (åstika) world, addressing and requalifying a) the very status of the bråhmaña; b) the plausibility of the symbolic practices of the bråhmaña –primarily represented by the ritualistic activity of sacrifice (yaj∞a)–; c) the legitimacy of the social representations that they provided to society through their works. Let us rapidly run over some examples of these three aspects to confirm the fact that, here too, we are dealing with a semiotic and political dispute on veridiction: a) adversus bråhmaña. The context in which Buddha ‡åkyamuni acted featured a considerable consensus on the ‘revealed’ nature of the Vedic sa∫hitå, on the correctness of the identity and the practices of the årya –whatever status we wish to attribute to this cultural indicator– and on the centrality of the sacrificial act (yaj∞a). From an analysis of the literature of the time, these elements appear to have been the subject of a more complex discourse, aimed at the construction of a cultural primacy and at the development of a precise identity: [a]s Aryans moved into the interior of India [...] there was a gradual process of naturalization or indigenization of Aryans and Aryanization of non-Aryans. This process was by no means smooth and uniform. It involved many violent social, religious, political and other conflicts, impositions and adjustments, and resulted in various groups holding different opinions on questions of identity, Aryanhood, Barbarianness etc. By the time the Buddhist and the Jaina canonical texts and the Hindu Dharma†åstras were formulated, there were several competing conceptions of Aryanhood and Aryan lands.60

Such a process inevitably had to develop in different directions that were at once rhetorical, poetical and political. Among these, that of Brahmanic superiority was of particular importance, as proved by the fact that it became the focus of the criticism of the traditional Brahmanic Weltanschauung raised by early Buddhism. To understand such criticism, first we have to consider the Brahmanic discourse on the superiority of bråhmañas. Starting from the text of ‡atapathabråhmaña –clearly concerned with the need to structure and rationalise the far from linear relations between political and secular power (råjanya), between the interests of the sacerdotal class (bråhmaña) and the whole community (vi†)– we have a systematic representation of the arguments that were later to characterise the classic forms of rhetoric on this Brahmanic primacy. For example, when the sources describe a bråhmaña descended from a r¢ßi as the earthly representation of the celestial order and all the gods,61 they show the axiological path of the foundation and the justification of the privileges and the preroga60 61

Deshpande 1993: 84. Further, Thapar 2000c. See ‡atapathabråhmaña, 12.4.4.6-7.

V. Violating Tradition and its Boundaries

455

tives of the bråhmaña himself (of which, moreover, the ‡atapathabråhmaña itself provides an initial list).62 Later, when the verses of Månavadharma†åstra were written, the concern for the cardinal aspects of the årya cultural discourse evolved into a new and more cogent phase. The topic of Brahmanic primacy was developed as never before, becoming the primary interest of a new intellectual circle, albeit within itself pluralistic and agonistically differentiated.63 The fact that the sacrificial practices cannot be separated from the Vedic collections (yaj∞asiddhyartham r¢gyajuΔsåmalakßañam),64 was then designed to establish not only that the Vedas –that is the very reason for Brahmanic existence– are both intrinsic and necessary to creation, but that the same must hold for the bråhmaña themselves. According to the author of the Månavadharma†åstra, the naturalisation of the primacy of knowledge and its possessors is indeed the central path to the foundation of their mutual relations and their ‘consubstantial’ dependency.65 It’s no coincidence that the discourse on Brahmanic excellence –despite already existing in nuce in other Brahmanic sources– finds systematic form in the verses of this important traditional piece of work.66 Here it is promoted and motivated through a skilful analogical logic that derives great semiotic power from the references to the very status of the nature of things. Far from being an arbitrary privilege, being a bråhmaña is presented as a condition innately connected to the ‘sense of the world’: it is said that the bråhmaña were generated through the evolution (vivr¢ddhyartha∫) of the world itself.67 The reassessment of a Vedic metaphor, that dictates the gen62 See ‡atapathabråhmaña, 11.5.7.1-2. It’s a list of exemptions then expanded in Gautamadharmasütra, 8.12-13, besides large reference in the Månavadharma†åstra. See infra. 63 See, on the intra-Brahmanic pluralism, Kosambi 2002: 87-97; Benveniste 2001: 215220; Kane 1997: vol. 2.1, 123-154; Inden 1992; Minkowski 1991: 20-22; 43-44; 111-128; Upadhyay 1979: 131-140. 64 See, exempli gratia, Månavadharma†åstra, 1.23. 65 See, exempli gratia, Månavadharma†åstra, 1.21-23. 66 Note that entire sections of the text devoted to celebration of the ‘condition of bråhmaña’ (bråhmañatva) occur only in one of the latest dharmasütra. See, exempli gratia, Vasiß™hadharmasütra, 3.1-12 (with some problematic adjustments); 30.1-11 (a veritable hymn to Brahmanic excellence). Despite this, the centrality of the bråhmaña was not first introduced in the Månavadharma†åstra, since we also find it in the oldest dharmasütra, such as the Åpastambadharmasütra, where out of the total 1364 sütra, 1206 revolve around the bråhmaña. However, there is no doubt that the Månavadharma†åstra marks the dawn of a singular phase of Brahmanic self-reference, since from the †loka 26 of the second book up to the †loka 97 of the sixth, the text deals with Brahmanic duties (dharmo bråhmañasya caturvidhaΔ). In this regard, it is important to note that the celebration of the bråhmaña at the end of the first adhyåya (Månavadharma†åstra, 1.92-101) continues at the end of the last adhyåya of the treatise (Månavadharma†åstra, 12.94-103). For further eulogistic verses, see Månavadharma†åstra, 9.313-322. The epic further extended this eulogistic tone towards the bråhmaña. See, exempli gratia, Mahåbhårata, 13.33.1-25; 13.34.1-29; 13.35.1-23. On the origin of this pretension to excellence see Smith 1994a: 26-36; 79-82. Further, Kane 1997: vol. 2.1, 134-140; Gonda 1981: 48-55. 67 See, exempli gratia, Månavadharma†åstra, 1.31.

456

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

esis of the social body through a sociogonic narration,68 now serves to define even more clearly the value and classification stances of the binary hierarchy (uttama Ù carama; utkarßa Ù apakarßa; utkr¢ß™aja Ù apakr¢ß™aja; †reß™ha Ù antyaja)69 inscribed within the ‘natural’ divisions of the social world which, precisely because considered natural, could not become the object of criticism. All this leads to a number of ‘inevitable’ considerations: the bråhmaña ‘that masters the Veda’ is ‘the lord of the whole world’ (œ†aΔ sarvasya jagato bråhmaño vedapåragaΔ)70 and –almost regardless of the level of his learning and of the correctness of his behavior– is, and remains ‘by nature’, ‘lord of all the classes’ (varñånå∫ bråhmañaΔ prabhuΔ),71 ‘great divinity’ (bråhmaño daivata∫ mahat),72 or rather ‘the greatest divinity’ (para∫ daivata∫)73 if not even ‘the divinity of the gods themselves’ (devånåm api daivatam).74 Therefore he is attributed a unique position and role, after which there is little else to say about the condition of exclusive primacy to be attributed to him. At this point all that remained for the author of the Månavadharma†åstra to do was to declare –thus generating a sensational petitio principii– that the bråhmaña enjoys not only the elevated status of ‘authoritative source of knowledge for the whole world’ (pramåña∫ caiva lokasya),75 but must also be considered the ‘owner of everything that there is on earth’ (sarva∫ sva∫ bråhmañasyeda∫ yatki∫jjagatœgatam) since, due to his pre-eminence and high birth (†raiß™yenåbhijanena), the ‘bråhmaña is entitled to the whole world’ (ida∫ sarva∫ vai bråhmaño ’rhati).76 The criterion devised to legitimise the dictates of the †loka is that according to which Manu, son of the self-existent, supposedly wrote a treatise (†åstra) to define the characteristics and activities of the bråhmañas,77 so that these can now be distinguished and recognised. In a context that has the cloying flavour of the self-referencing panegyric,78 Manu, the †åstra and the bråhmaña thus become parts of an indivisible trio.79 It was on the basis of such arguments –specifically conceived and structured in the Månavadharma†åstra– that the ‘discourse on 68 See, for the original Vedic narration of the social quadripartite model, R¢gveda, 10.90.11-12. 69 See, for examples of such contrapositions regarding the spatiality of social status, Månavadharma†åstra, 1.92-93; 10.42 (further, Bhåruci, Manu†åstravivaraña ad Månavadharma†åstra, 10.42); 8.281; 8.279. 70 Månavadharma†åstra, 9.245. 71 Månavadharma†åstra, 10.3. 72 Månavadharma†åstra, 9.317. 73 Månavadharma†åstra, 9.319. 74 Månavadharma†åstra, 11.85. 75 Månavadharma†åstra, 11.85. 76 Månavadharma†åstra, 1.100. 77 See Månavadharma†åstra, 1.102. Further, Månavadharma†åstra, 1.26-30. 78 See Månavadharma†åstra, 1.92-101. 79 See Månavadharma†åstra, 1.102-106.

V. Violating Tradition and its Boundaries

457

Brahmanic excellence’ and on the superiority of his activities began to assume its definitive shape. But, in order to understand, in all its depth and repercussions, the diffusion and the structuring of the claim to superiority of the bråhmaña, we need to consider this in the light of polemical tension with early Buddhism. It is otherwise difficult to justify the intensity and the reasons for this defence of the corporation by the Brahmanic corporation itself. This instead appears in all its cogency when we associate with it the Buddhist criticism comprised in early Påli literature, where the bråhmaña are ridiculed for their presumption, inconsistency, ostentation and contradictions. Buddha’s words, as recorded in the bråhmañavagga of the Dhammapada, are fairly exemplary of this polemical climate where a given fact traditionally seen as valid is exposed to delegitimisation: 393. One is not a brahmin by virtue of matted hair, lineage or caste. When a man possesses both Truth and truthfulness, then he is pure, then he is a brahmin. 394. What use is your matted hair, you fool? What use is your antelope skin? You are tangled inside, and you are just making the outside pretty. 395. The man who wears robes made from rags off the dust heap, who is gaunt, with his sinews standing out all over his body, alone meditating in the forest –that is what I call a brahmin. 396. I do not call him a brahmin who is so by natural birth from his mother. He is just a supercilious person if he still has possessions of his own. He who owns nothing of his own, and is without attachment –that is what I call a brahmin.80

Beyond these distinctly polemical considerations, there are many more present in the Påli sources.81 As a consequence of the acute cogency of such criticism, a widespread apologetics for the status of the bråhmaña emerges within Dhammapada, 26.393-396 (trans. by J. Richards 1993). See, exempli gratia, Dœghanikåya, 1 (brahmajålasutta).1.11 - 1.3.74 (in PTS vol. 1, p. 546); Dœghanikåya, 4 (soñadañ∂asutta).3-23 (in PTS vol. 1, p. 112-124); Dœghanikåya, 13 (tevijjasutta).1-36 (in PTS vol. 1, p. 235-248); Dœghanikåya, 27 (agga∞∞asuttanta).3-7 (in PTS vol. 3, p. 80-83); Aõguttaranikåya, 191 (bråhmañavagga).1-6 (in PTS vol. 3, p. 221-222); Suttanipåta, 1 (uragavagga).7.25-27 (in PTS p. 24) (here the fact that certain bråhmañaborn [jåti] sin [påpakamma] is discussed, together with the fact that just being born bråhmaña it is not enough to be qualify); Suttanipåta, 2 (cülavagga).7.1-32 (in PTS p. 50-55) (in this decisive bråhmañadhammikasutta Gotama explains and criticizes different aspects of the status of the bråhmañas of his time –sometimes also calling them back to the respect of their disappeared integrity, thus appearing ‘orthodox’–, starting immediately with a negative answer to the question if there were still bråhmañas faithful to the ancient dharma [‘sandissanti nu kho bho Gotama etarahi bråhmañå poråñåna∫ bråhmañåna∫ bråhmañadhamme’ ti. ‘na kho bråhmañå sandissanti etarahi bråhmañå poråñåna∫ bråhmañåna∫ bråhmañadhamme’ ti.]. Finally he concluded it quoting a term destined to become famous: bråhmabandhü [2.7.32]); Suttanipåta, 3 (mahåvagga).9.1-63 (in PTS p. 115-123) (here the discussion between Gotama, Våse™™ha and Bhåradvåjå on what makes someone be a bråhmaña). Further, Bailey, Mabbett 2003: 113-129; Collins 1993: 349-351. Moreover, for further examples of criticism to the status of the bråhmañas later made by jaina and bauddha, Osier 2000: 139-153; Masset 2000: 155-173; Eltschinger 2000: 48-60. 80 81

458

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Brahmanic literature, which embraces both the above-mentioned eulogistic formulas and the ostentation of the primacy of the social position, almost irrespective of the subjective qualities of the individual. The desire to avoid the spread of mistrust of the corporation leads to the aprioristic defence of the condition of the bråhmaña, even when a bråhmaña is discovered to have acted reprehensibly. Apropos this, it is stated: “Similarly, even if they engage in every undesirable act, Brahmins should be honoured in every way; for they are the highest deity”.82 The attempt here is to convince of the fact that even after a grievous lapse, the bråhmaña can never be deprived of the respect and honour due to him, since he remains a symbol of divinity and a guarantor of the preservation of traditional views and praxis. This is a further example of how the existence of an articulated dissent dictated the syntax of the ‘orthodox’ structure of abstract norms and practical rules. b) adversus yaj∞a. Another occasion of criticism is that of the Buddhist debate on the centrality of the sacrificial rituals. A criticism, however, that already existed within the Brahmanic world of that time,83 a world that we consider as a single and unified cultural ‘universe’. Clear references to the anti-ritualistic and anti-sacrificial debate are present from the earliest Buddhist literature, comprising censure of the Brahmanic abuse of violent rituals and bloody practices connected with sacrifice.84 In a significant section of the Suttanipåta –polemically entitled bråhmañadhammikasutta–,85 Gotama Buddha reconstructs an itinerary of causes to motivate his disapproval of the sacrificial forms which he evidently witnessed in his time. He begins his engaging narrative by 82 Månavadharma†åstra, 9.319. In this regard it is significant to note that the section where the †loka 319 appears (9.315-320), is entirely devoted to re-establish the greatness of the bråhmaña –a privileged status evidently under threat at the time–, and that it have a parallel version in Mahåbhårata, 13.136.16-23. Also, it is important to consider the theological implications that in other parts of the epic are added to the principle of inviolability expressed in the †loka of Manu. See, exempli gratia, Bhagavadgœtå, 9.30 (api cet suduråcåro bhajate måm ananyabhåk Ù sådhur eva sa mantavyaΔ samyag vyavasito hi saΔ ÙÙ). 83 As is shown in the polemical references to specific celebrations or rituals (exempli gratia, Aitareyårañyaka, 2.1.2; 3.2.6 [ed. Keith: 101; 139]); to rites referred to as ‘insecure vassels’ (exempli gratia, Muñ∂akopanißad, 1.2.7-8); or to the ‘flowery language’ (pußpitå∫ våca∫) of the Veda and to the sacrificial Vedic practices (exempli gratia, Bhagavadgœtå, 2.4243; 2.45-47; 2.53; 9.20-21). 84 See, exempli gratia, Dœghanikåya, 5 (kü™adantasutta).1-30 (in PTS vol. 1, p. 127-149); Majjhimanikåya, 129 (balapañ∂itasutta).13-17 in PTS vol. 3, p. 167); Aõguttaranikåya, 89 (sœtivagga).1-2 (in PTS vol. 3, p. 438); Sa∫yuttanikåya, 1.50(gha™œkårasutta∫) (in PTS vol. 1, p. 75-76); Suttanipåta, 2 (cülavagga).7.12-32 (in PTS p. 52-54); Suttanipåta, 3 (mahåvagga).4.1-32 (in PTS p. 79-86) (where the discussion between Gotama and one bråhmaña on real oblation is cited); Suttanipåta, 5 (påråyanavagga).4.1-6 (in PTS p. 199-201) (as a response to the questions of Puññaka and contesting the Brahmanic ideology, Gotama says here that r¢ßis, bråhmañas and all the descendants of Manu performed sacrifices because of their longing for a good life in this world and out of eagerness for earthly pleasures). Further, Freiberger 1998; Bhattacharya 1973. 85 See Suttanipåta, 2 (cülavagga).7.12-32 (in PTS p. 52-54).

V. Violating Tradition and its Boundaries

459

invoking the correctness of the ancient way of performing sacrifices in which, as he recounts it, the killing of cattle was not allowed.86 He goes on to emphasise the significant historical occurrence of a change in conditions,87 for which the bråhmañas became thirsty for power and slaves to greed and lust.88 As a result of this, they were therefore driven towards the more remunerative sacrificial practices involving horses and human beings.89 It was this debasement which distanced the bråhmañas from the rectitude of the ancient dharma and exposed them to the disapproval of the people.90 These appear to be the reasons why, according to the Buddha, it was no longer justified for those in search of salvation to turn to the traditional institution of Brahmanic sacrifice. As I said above, this criticism was certainly not newly introduced by the Buddha, since even in the Upanißads we find statements that openly criticise Vedic ritual and its underlying violence. Certain utterances in these works play down the role of traditional sacrifice, introducing a symbolic interpretation of the elements of the sacrificial practice.91 The sacrificial controversy is in fact a long-standing and thorny issue,92 periodically invoked when certain social alliances fail.93 However, the teaching of the Buddha can be presented as revealing reasons for delegitimising a tradition of beliefs and practices and, consequently, for justifying his own decision to change direction or to invoke a radical renewal. iii) adversus caturvarña. The same polemic attitude towards the Brahmanic sacrificial vision is found –as far back as the documents of the earliest Buddhist traditions–, in relation to the traditional discourse on the natural condition of the social order based on the notion of varña.94 The question at stake is well defined in that section of the Dœghanikåya known as agga∞∞asuttanta. The beginning of the sutta 86 See Suttanipåta, 2 (cülavagga).7.12-13 (12. tañ∂ula∫ sayana∫ vattha∫ sappitela∞ ca yåciya dhammena samudånetvå tato ya∞∞am akappayu∫ Ù upa™™hitasmi∫ ya∞∞asmi∫ nâssu gåvo hani∫su te ÙÙ 13. ‘yathå måtå pitå bhåtå a∞∞e vå pi ca ∞åtakå gåvo no paramå mittå Ù yåsu jåyanti osadhå Ù annadå baladå c’etå vaññadå sukhadå tathå’ etam atthavasa∫ ∞atvå nâssu gåvo hani∫su te ÙÙ). 87 Suttanipåta, 2 (cülavagga).7.16 (tesa∫ åsi vipallåso). 88 Suttanipåta, 2 (cülavagga).7.16 (disvåna añuto añu∫ råjino ca viyåkåra∫ nariyo ca samala∫katå ÙÙ). 89 Suttanipåta, 2 (cülavagga).7.20 (assamedha∫ purisamedha∫). 90 Suttanipåta, 2 (cülavagga).7.30 (yåjaka∫ garahatœ jano ÙÙ). 91 See, for some examples of this interpretation of sacrifices, Br¢hadårañyakopanißad, 3.1 - 3.9; Kaußœtakyupanißad, 2.5. 92 The section titled Structure et expression du conflit entre le sacrificant et le non-sacrificant in Oguibénine (1985: 53-66) provides a precious overview of this issue. 93 See, exempli gratia, Mahåbhårata, 14.28.6-28 (on yati adhvaryu samvåda). Further, on the agonistic relation between ritual and renunciation, Heesterman 1985: 26-44. 94 See, exempli gratia, Dœghanikåya, 3 (amba™™hasutta).1.10-1.28 (in PTS vol. 1, p. 90-99); Dœghanikåya, 4 (soñadañ∂asutta).3-23 (in PTS vol. 1, p. 112-125); Dœghanikåya, 27 (agga∞∞asuttanta).3-7 (in PTS vol. 3, p. 80-83); Dœghanikåya, 13 (tevijjasutta).19 (in PTS vol. 1, p. 241-242); Suttanipåta, 1 (uragavagga).7.1-27 (in PTS p. 21-25). Further, Eltschinger 2000: 33-60; 83-86; Collins 1993: 301-393.

460

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

cites a series of questions put to the Buddha by Våse™™ha and Bhåradvåja (both names are in Påli and refer to important figures of Brahmanic society, here undoubtedly used with strategic intentions) regarding the status of those who violate the specific duties of the class (vañña in Påli) they belong to. These questions illustrate the problematic situation of those who attempted to alter their social position: Then the Exalted One said to Våse™™ha: You, Våse™™ha, being brahmins by birth and family, have gone forth from a brahmin family, your home, into homeless life. Do not the brahmins blame and revile you? Yeah, verily, lord, the brahmins do blame and revile us with characteristic abuse, copious, not at all stinted. But in what words, Våse™™ha, do they so blame you? The brahmins, lord, say thus: The brahmins class is the best. […] Only a brahmin is of the best social grade; other grades are low.95

This introductory section of the sutta ends with the significant words of the Buddha, meant to show his interlocutors that good or bad qualities do not belong to a specific social class, nor to elective principles connected with birth, but are to be found in any sphere of society: Now seeing, Våse™™ha, that both bad and good qualities, blamed and praised respectively by the wise, are thus distributed among each of the four classes, the wise do not admit those claims which the brahmins put forward.96

The legitimacy of the social representations proposed by the Brahmanic tradition is questioned in other portions of the Påli sources, as in that part of Suttanipåta –known as vasalasutta–,97 which contains an important discussion on the condition of the ‘out-caste’ (vasala). Recorded here is Gotama’s significant phrase: ‘it is not by birth that one is vasala and it’s not by birth that one is bråhmaña’,98 through which the Buddha launches a severe critique of the system of social reproduction and legitimisation at the basis of the Brahmanic discourse. Certainly, challenging a traditional social order does not signify invalidating the peculiar legitimising function of tradition, with all its tendentially monopolistic and exclusionist implications. Quite the contrary. Showing the faults of a tradition is a good way to establish one’s own version of that tradition. Effectively, Buddhist authors themselves criticized, in a way that we would be led to define as ‘traditionalist’, a vision of the world which for centuries had been present95 Dœghanikåya, 27 (agga∞∞asuttanta).3 (in PTS vol. 3, p. 81). Trans. from Rhys Davids 2000: 77-78. 96 Dœghanikåya, 27 (agga∞∞asuttanta).7 (in PTS vol. 3, p. 83). Trans. from Rhys Davids 2000: 79. 97

See Suttanipåta, 1 (uragavagga).7.1-27 (in PTS pp. 21-25). Suttanipåta, 1 (uragavagga).7.21 (na jåcca vasalo hoti na jaccå hoti bråhmaño Ù kammanå vasalo hoti kammanå hoti bråhmaño ÙÙ). These words are repeated in Suttanipåta, 1 (uragavagga).7.27, and then in Suttanipåta, 3 (mahåvagga).9.55-59 (in PTS p. 122). 98

V. Violating Tradition and its Boundaries

461

ed as ‘orthodox’ (åstika). No doubt they considered their system of knowledge as the only valid ‘tradition’. In fact, the ancient bauddhadharma was not just a challenge to the Brahmanic tradition, but expressed the will of Buddhists to gain their own space inside the religious milieu. As proof of this, there are many occasions in which, later on, Buddhist authors have spoken of the vedavådin –and hence the ‘orthodox’ (åstika) par excellence– as if they were ‘heretical’. Indeed, well known Buddhist figures of the early centuries CE, such as Någårjuna and Åryadeva,99 as well as others,100 often described Brahmanic ideas and concepts as ‘heresies’ or errant visions. Although this latter phenomenon brings with it a second set of problems –namely the polyvalence and the relational condition of ideological definitions, such as ‘heresy’ and ‘orthodoxy’–,101 it certainly does not undermine the stature and prominence of the issue, that is the conflict between traditions for the monopoly of veridiction, a conflict that moulded the intellectual profile of ancient South Asia. The case of baudddharma presented so far, indeed, still refers back to the profound structural role –affecting minimal facts of everyday life as well as the construction of broadest symbolic systems– of the dialectical and agonistic process that occurs when different traditions meet and co-exist. It also underlines the dependencies created between a given intellectual formulation and the surrounding and preexistent cultural terrain, which has substantially provided the conceptual tools through which new understandings have come into being.102 Richard Gombrich gives a clear example of this phenomenon, when he describes the tension between the Buddha’s practice and the thought of his contemporaries: “[…] the Buddha’s teaching evolved in dialogue with other religious teachers of his day, especially Brahmins”.103 The respective influences among diverse intellectual and social imageries can be multiple. Nothing can prevent one from adopting the basic concept of another, maybe even taking it more seriously or shaping it to a different mould. This can then give rise to the devising of new conceptual approaches, in a sort of reciprocal differentiation of the respective contents, implemented in dependency and caused by dependency.104 99

See Nakamura 1955: 93. See Nakamura 1955: 79, 84, 88-89, 94, 98, 102. 101 A meaningful example in this regard is provided by a Buddhist author such as ‡åntarakßita (750-810 CE), who, while involved in criticizing the work of a champion of Brahmanic ‘orthodoxy’ such as Kumårila Bha™™a (620-680 CE), bitterly hurls himself at one of the privileged targets of the traditional Brahmanical apologists, that is the ‘materialists’ (cårvåka). See, regarding the two sides of this controversy, Taber 2001; Bhattacharya 2000c. 102 See, for examples of mutual influence, Bronkhorst 2000; Bollee 1999. 103 Gombrich 1997: 27. Further, for a different –yet inverted– approach at the same issue, King 1997: 183-204. 104 See, on dialectical developments of intellectual productions in ancient time, Deshpande 1997; Silburn 1989: 147-164; Bronkhorst 1998: 79-88; Gombrich 1997: 27-64. 100

462

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

We need therefore to identify the modes of relation and the dynamics between the members of these social ‘parties’, whether perceived as marginal or not. Such ‘parties’, involved in an endless hermeneutic cycle, seek to comprehend –and hence draw to themselves– the difference of the other, transported by a sort of breathless pursuit that aims to ground, and then support, the pilasters of a given institution of thought or, more precisely, of a “legitimate social grouping”,105 which intends to transmit itself faithfully to its followers and become for them the benchmark tradition. 6. Tradition against Traditions. Defending tradition through the counterpoising of dharma and adharma All the above being said, I think it useful now to try and read the details of the ‘traditional’ production of binary oppositions (selecting among them the opposition between dharma and adharma).106 These are to be understood as the way in which a certain vision has sought to articulate the defence of its point of view –perceived and valued certainly not as one of many points of view, but as the mode of being of the nature of things–, while also appealing to criteria of perceptible efficacy and observable ethical motives. Effectively, the policy of binary division, which includes apologetic components, has developed through a discourse on the justice and error of certain convictions and beliefs. In many Brahmanic works this discourse unfolds through the delineation of both elective routes and exclusive models, through which the practices of the various cultural contexts are to be distinguished and judged. This has led, as in the case of the Månavadharma†åstra, to the production of moral ideal-types of ‘politeness’ and ‘goodness’ (åryatå, sådhutva),107 to be counterpoised with the antagonistic ideal-types of ‘impoliteness’ and ‘evilness’ (anåryatå, asådhutva).108 It is precisely here that the potential success of a particular point of view resides, intended no longer to be recognised for what it is –namely specific, restricted, arbitrary, partial and artificial– but as something that is above itself. This is where the true strength of tradition comes in. A strength that stems mainly from the establishment and universalisation of spe105

Douglas 1990: 82. Normative sources have extensively tried to draw a border between dharma and adharma. Remarkable is the frequency –although there are several semantic differences– on which the opposition between dharma/adharma is used in Manu’s work. See Månavadharma†åstra, 1.26; 1.29; 1.81-82; 2.111; 2.206; 3.11; 3.22-26; 4.60-61; 4.133; 4.170177; 6.64; 7.7-18. 7.28; 8.8-24; 8.41-44; 8.58-59; 8.122-123; 8.127; 8.174-175; 8.228-229; 8.304306; 8.310; 8.348-349; 8.353; 8.381; 9.233-234; 9.236; 9.248-249; 9.251; 9.270; 9.273; 10.106-108; 11.21; 11.29-30; 12.20-23; 12.117-118. 107 See, exempli gratia, Månavadharma†åstra, 7.211; 10.57. The sådhus are called to testify their condition of ‘goodness’. See, exempli gratia, Månavadharma†åstra, 2.11; 4.252. 108 See, exempli gratia, Månavadharma†åstra, 9.17; 10.57-58. Conversely asådhu is described in Månavadharma†åstra, 3.182; 11.19. 106

V. Violating Tradition and its Boundaries

463

cific practices, conduct and beliefs, from the historical existence of which a great power can be drawn.109 It is precisely the historical existence of a given doxological horizon (represented by recourse to the distinctive device separating ‘tradition’ from ‘novelty’, and between ‘right’ [dharma] and ‘wrong’ [adharma]) that constrains the development of reasoning within specific forms of internal coherence. It was the objective variants introduced by new historical conditions which then interrupted the reproducibility of such doxological schema, generating new articulations of the norm (dharma) itself,110 and stimulating the review and redefinition of the traditional heritage. Even the most conservative tradition knows that excessively rigid models for representing and explaining the world cannot always be used. More ductile methods are needed to conciliate dissimilar but indispensable elements, which up to then may have been considered as opposed and antithetical. Exemplary, in this regard, is the attempt to harmonise reason and tradition made, not without difficulty, in the Månavadharma†åstra: 106. The man who scrutinizes the record of the seers [the Vedic texts] and the teachings of the Law by means of logical reasoning not inconsistent with the Vedic treatise –he alone knows the Law, and no one else.111 109 Recourse to the strength of existing things by reference to the past is a powerful way of legitimising and transferring the status quo. Extensive examples of this practice are to be found in the Månavadharma†åstra, that inaugurates the assertive apodictic style of formulas such as ‘Manu has proclaimed’ (manur abravœt) –or ‘Manu has ordained/established’ (tan manor anu†åsanam)–, whereby the author calls upon the authority of Manu’s past edicts to enhance the weight of what he has just said. See Månavadharma†åstra, 3.150 (manur abravœt); 4.103 (manur abravœt); 5.131 (tan manur abravœt); 8.139 (tan manor anu†åsanam); 8.168 (manur abravœt); 8.242 (manur abravœt); 8.279 (tan manor anu†åsanam); 8.292 (manur abravœt); 8.339 (manur abravœt); 9.182 (manur abravœt); 9.239 (tan manor anu†åsanam). This is a practice already noted by various bhåßyakåra, since Medhåtithi’s comment underlines just this aspect of the emphatic use of the epithet ‘Manu’. See Medhåtithi, Manubhåßya ad Månavadharma†åstra, 4.103. In other cases, driven by the same desire for legitisation, in other cases the author of the treatise adds the charisma of ‘tradition’ (smr¢ti) to that of Manu, by invoking the weight of what is ‘recalled’ (smr¢ta) by it. See Månavadharma†åstra, 1.27; 1.47; 1.78 1.96 2.3 3.46 3.194-195; 7.156; 8.416; 10.10; 10.41; 10.45-46; 12.95. 110 An example of this practice is the creation of a ‘rule for adverse conditions’ (åpaddharma). Recourse to the ‘rule for crisis’, to the ‘rule for adverse times’ and to the ‘rule for state of emergency’ is however inevitable, as it is a common subject of juridical texts. See, on åpaddharma, the 38 adhyåya of †åntiparvan (12.129 - 12.167) of the Mahåbhårata and the 130 †loka (10.1-131) of the Månavadharma†åstra. Further, Upadhyay 1979: 122-124; 230-232. It is important to notice here the much smaller parts devoted to the notion of ‘adverse times’ (åpatkalpa) by the authors of ancient dharmasütras. See Åpastambadharmasütra, 1.20.11; 2.4.25; Gautamadharmasütra, 7.1-26; 9.67; Baudhåyanadharmasütra, 1.3.41-42; 2.5.7; 2.18.18; Vasiß™hadharmasütra, 2.22. Moreover, for general theoretical remarks on the notion of état de siège, Agamben 2003. 111 Månavadharma†åstra, 12.106. Just for its problematic character, this verse has been debated upon by several commentators. See, exempli gratia, Bhåruci, Manu†åstravivaraña ad Månavadharma†åstra, 12.106; 12.111; Medhåtithi, Manubhåßya ad Månavadharma†åstra, 12.106. Then, on the role to be given to logical deduction while understanding dharma, see Wezler 1999. Further, on the complex relationship between reason and tradition in South Asian classical texts, Mohanty 1992: 272-276; Halbfass 1991: 131-204.

464

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

While thus enjoying the benefits offered, in terms of legitimacy, by reference to the historical nature of a tradition and to ‘what is recorded by it’ (smr¢ti) –that is, what an established memory, oriented by a precise selective principle, considers opportune and functional to record–, specific agents seek to reproduce, albeit not servilely,112 the fundamental elements of the discourse underlying the process that has given structural strength to certain objective structures. If the idea that we ‘produce ourselves by producing’ holds here too, then ‘reproducing is essential to reproduction’.113 It is in this respect that the attempt to reproduce and redefine the original Vedic discourse on social division ratified by the order of the varña becomes clearly understandable.114 The same applies to the revival, in post-Vedic times, of the opposition between dharma and adharma. In both such cases, we are not dealing with a simple reproposal –in a revisionist and traditionalist key– of faded ideologies, but rather with a conscious procedure for justifying the introduction of renewed theological and political paradigms, designed to draw credibility and profit from recourse to the rhetorical and semiotic device of the ‘Vedic tradition’.115 Even for the Brahmanic traditions of the early centuries BCE –which can be defined as ‘reiterative institutions of Vedic legitimacy’–, looking at the past was a way of legitimising the governing of the present. Nevertheless, paying excessive attention to ideal oppositions, such as dharma and adharma, implicates the risk of thinking that general contentions are fuelled solely by questions of principle –therefore neglecting the historical and social role of the exponents of such an opposition. Effectively, no dispute on universal principles could survive independently, were it not to find new and more cogent articulations dictated by specific situations. 112 For example, the author of the Månavadharma†åstra is in line with the existing juridical-regulatory currents without being subject to them, in fact proving to be able to introduce several innovations. In this respect, attention is drawn to the fact that the Månavadharma†åstra is a ‘non-affiliated’ work (being deprived of a direct link with any kalpasütra and not restricted to a specific †åkhå, according to what has been said since Kumårila Bha™™a, Tantravårttika, 1.3.8 [ad ‡abarabhåßya, 1.3.15]), as for the Gautamadharmasütra –that too believed to be the result of the author’s attempt to “[...] produce an ideal sütra work” (Olivelle 2000: 8)–, which however shares several verses with the Månavadharma†åstra and along with it establishes a significant case for the merging of a political treatise with a ‘religious’ one. See Olivelle (forthcoming). 113 This is a special aspect of the social reproduction processes referred to by Pierre Bourdieu. See, exempli gratia, Bourdieu 2000: 58-71; Bourdieu 1998a: 131; Bourdieu 1995: 33-34; 58-62; Bourdieu, Passeron 2000: 3-68. 114 See, for a reading of social ideologies of the Vedic world which is very close to the quoted reproduction notion, Smith 1994b: 67-93. 115 Recourse to the Veda with the intent to establish and justify new beliefs and practices has been identified –since 1960 by Louis Renou– as a classical and widely used apologetic strategy. See Renou 1960: 2. Further, on the same subject, Smith 1998: 20; Gonda 1997: 7-8; Patton 1994: 1; Halbfass 1991: 1-3.

V. Violating Tradition and its Boundaries

465

The †loka of the Månavadharma†åstra are clearly a proof of this logic. Laden with hints and references to various forms of conflict –albeit veiled by the writer’s efforts to disregard their real status–, the verses of Manu’s treatise are explicit indicators of agonistic situations of a historical type. The arrangement of the syntax of this regulatory discourse, still formally linked to the ideal distinction between dharma and adharma, is dictated precisely by the conflictual encounter between diverging interests. A concrete context where rivals are viewed, from a hegemonic angle, either as menacing adversaries or as necessary –although inconvenient– allies. 7. Conflicting traditions: from the defence of the universals to the monopoly on the particulars Let’s now dwell on the ‘universal grammar’ that underlies the drafting of the Månavadharma†åstra, a text that is intimately related to the defence of specific interests. Effectively, as incisively observed, [...] Manu’s interest lay not in the lower classes of society, which he considered to be an ever-present threat to the dominance of the upper classes, but in the interaction between the political power and Brahmanical priestly interests, interests that were under constant threat ranging from the A†okan imperial polity to the foreign invasions toward the turn of the millennium.116

As a matter of fact, the author is objectively concerned that the educated classes, involved in administration and politics, will appeal to other authorities or will ‘deviate from the injunctions of the authoritative texts’ (lubdhasyocchåstravartinaΔ).117 He also fears an unfavourable outcome of the dispute with the various detractors of his tradition, who threaten it from several sides. The entire Månavadharma†åstra is imbued with this anxiety, which can be summarised in three forms: anxiety about the maintenance, within the Brahmanic field, of the orthodoxy or regularity of certain practices;118 anxiety about the incompatibility between the interests of priests and those of the king and politicians;119 anxiety about the conflict between the ‘orthodox’ tradition and subjects perceived as external to Brahmanic dharma. 116

Olivelle 2002: 547. See Månavadharma†åstra, 4.87. 118 Within the classical Brahmanic world if catching (and ‘having’ other people catch, that is ‘enabling them to recognize’) somebody’s mistake is with no doubt a symbolic form of violence, being disrespectful to the orthodox form of a given doctrine and vision is a serious act of delegitimization. Indeed, the ‘proper doxa’ is considered such and justified only through the strength given by observance, that signifies and confirms it. See, for a pertinent quote about the emphasis to be given on correct praxis, Månavadharma†åstra, 3.84-92. 119 See, for a study that allows to understand the level of complexity at that time and explains the implications behind the different terms used to indicate political-administrative powers, Kane 1993: vol. 3, 20-55; 104-131. 117

466

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Because of that, the Månavadharma†åstra is a proof of the fact that both the threat and the agonistic dialectics between defenders and aggressors of a given ‘tradition’, are key elements that forge the construction of a norm. As a consequence, the practical logic of this regulatory intention cannot be understood intimately if distanced from the elements that contributed to its construction. Below, I intend to focus only on the third type of this three-fold agonistic context, not having sufficient space here to present the parts of the Månavadharma†åstra concerning the first two.120 Following the sequence of the adhyåya, these are the sections of the text that include references and representations of points of view, beliefs, practices and individuals that the author of the Månavadharma†åstra perceives as a serious threat to his own vision and to the tradition that claims to be the owner, protector and perpetrator of the same. First of all, the author attempts to include the existence of critics outside his tradition in a wider framework, as he does with the two other types of conflict. Utilising theological logic, he launches an articulated process of naturalisation of such opponents, in a way that not only justifies their existence but also theorises their emergence. He seeks to convince his readers that the existence of rivals, opponents and heretics should not be seen as the result of an actual choice to look at the world in a different way nor as a response to the inefficacy of the existing modes of signifying reality. On the contrary, such existence should be traced to reasons external to the world –thus truly ‘other-worldly’–, namely a precise design of destiny inscribed within an equally accurate philosophy of history.121 In order to distinguish, classify and morally judge all antagonism, he appeals to the mythical and sociogonic reasons implicit in his cosmogonic model. Thus, being a great master of sociodicy, the author invites us to read agonism –and hence the very possibility of criticism– through the distorting lens of the ‘natural’ causes underlying the degeneration of customs and ‘irreligion’ (adharma). Certain of achieving results, he exploits the immediacy of the metaphor of ‘the four feet of dharma’:122 81. In the Kr¢ta Age, the Law is whole [dharmaΔ satya∫], possessing all four feet; and so is truth. People never acquire any property through 120

I have carried out this analysis elsewhere. See Squarcini (forthcoming): chap. 5.1. See, on the prescriptive character of the doctrine of the ‘epochs of the world’ (yuga), Gonzalez-Reimann 2002; Olivelle 1993: 234-237. 122 The four ‘feet of the law’ (carita, vyavahåra, dharma, råja†åsana) are described in Kau™ilya’s work too, within a context where the figure of the king as source of the norm (dharmapravartaka) is emphasized. See Kau™ilya, Artha†åstra, 3.1.39 (dharma† ca vyavahåra† ca caritram råja†åsanam Ù vivåda artha† catußpådaΔ pa†cimaΔ pürvabådhakaΔ ÙÙ). Further, Parpola 1975-76. Later on, the image of dharma and of its ‘four feet’ has been largely used in South Asia. In medieval texts, for example, we find the metaphor of dharma bull that sees his legs being hit by the degradation of passing epochs. See, exempli gratia, Bhågavatapuråña, 1.17.22-30. 121

V. Violating Tradition and its Boundaries

467

unlawful means. 82. By acquiring such property, however, the Law is stripped of one foot in each of the subsequent Ages; through theft, falsehood, and fraud, the Law disappears a foot at a time. (Månavadharma†åstra, 1.81-82)

This means that the suppression of those who are adhårmika and påpin represents a source of purification and redemption for the sovereign, just as the sacrifice leads to purification for the dvija.123 Following this logic, at the end of the first adhyåya (Månavadharma†åstra, 1.115) the author announces that part of his work will be devoted to what he defines as the practice of ‘eradication of the thorns’ (kañ™akoddharaña), or ‘removing of the thorns’ (kañ™akå†odhana).124 This part of the treatise (which comes immediately after, though not within, the discourse on the eighteen ‘spheres of legal dispute’ [vyavahårapada] and hence placed under the sovereign’s direct responsibility and authority) is extremely interesting for my argument. It illustrates the repressive policy to be applied against all those –euphemistically referred to as ‘thorns’ (kañ™akå)– who threaten the credibility and the legitimacy of the social order. As well as thieves and criminals of various kinds, these individuals also include prophets, heretics, astrologers, ascetics, etc., who, according to the author, contribute to a greater degeneration and confusion of customs. Now the ‘heretics’ (nåstika), the ‘offenders of the Veda’ (vedanindaka),125 the ‘logicians’ as well as the ‘reasoners’ (hetuvådin, hetu†åstrin) and those ‘not faithful to vows’ (vråtya),126 are at the centre of a manoeuvre to defend the tradition, the tone of which well illustrates the extent and the motivation of the concern which these adversaries arouse in the author and his corporation.127 123 See Månavadharma†åstra, 8.310-311. The historical significance of these verses can be better understood if we consider their moralistic nature, which becomes evident if, after checking the concordances (Olivelle 2005: 1009-1034), we take count of the fact that such stanzas do not exist in previous texts on dharma. Another significant example of this is in Månavadharma†åstra, 8.352-353. Here, after invoking general principles (Månavadharma†åstra, 8.348-351 [all these verses are taken from previous dharmasütras]) concerning the legitimate use of violence, the author infers by induction a particular use of that principle, according to which it is possible to make use of different forms of pre-emptive and corrective violence. Similar cases are cited in Månavadharma†åstra, 9.225-226; 9.252-293; 10.58-61. In my view, the very firm and apodictic tone of such examples is due to the interested vision of the treatise’s author, who responds to a precise socio-political context. 124 See Månavadharma†åstra, 9.252-293. Further, Månavadharma†åstra, 8.386-387. There is also a whole section of Kau™ilya’s treatise where the topic of ‘removing thorns’ (kañ™aka†odhana) is discussed, with references to three judges assigned to do that. See Kau™ilya, Artha†åstra, 1.1.6; 4.1.1 - 4.13.43. Again, Kåtyåyanadharma†åstra, 15 (prajånå∫ rakßaña∫ nitya∫ kañ™akånå∫ ca †odhanam Ù dvijånå∫ püjana∫ caiva etad artha∫ kr¢to nr¢paΔ ÙÙ). This image is also used in didactic Sanskrit literature, as in Pa∞catantra, 3.62. See, on this particular topic, Lingat 1999: 237-240; Kangle 1997: vol. 3, 232-243; Scharfe 1989: 41; 214. 125 See, on these ‘offenders of the Veda’, Månavadharma†åstra, 2.11; 3.161; 4.163; 11.57. 126 See Hiltebeitel 2001: 132-139; Sharma 2001: 54-63. Further, Macdonell, Keith 1982: 342-345. 127 See, for two descriptions of domains to be defended against this ‘dangers’, Månavadharma†åstra, 2.10-11; 2.39-40.

468

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

After claiming that dissidents and offenders threaten the transmission of tradition and do ‘things that should be avoided’ (vikarma) –such as resisting and repudiating the master or abandoning the sacred fire–,128 the treatise asserts that a ‘twice born’ (dvija) should not even address words of welcome to such individuals who ‘offend’ (påßañ∂in) and are dedicated to futile ‘sophisms’.129 ‘Twice born’ and bråhmaña are strictly prohibited from frequenting such people or the places that they live in or govern. Indeed, the text makes explicit reference to deviant sovereigns and to kingdoms where different ethical criteria and different systems of belief prevail.130 The intention is to inhibit dvijas and brahmañas from serving those that are ‘outside’ the tradition. Should such injunctions not be observed, the text –thus revealing the agonistic justification of what are presented as mere ‘descriptions’– condemns the transgressors to hell.131 The antagonists are then referred to as ‘unjust people’ (adhårmiko naro yo hi), while various references to the ill-omened effects produced by their beliefs and practices help the author to clearly define the value borders between different points of view, resorting even to serious eschatological threats.132 In order to prevent dangerous mixtures, therefore, it is said that neither rites nor sacrifices can be celebrated for rebels, ascetics, heretics and similar people, all bundled together in a single category (påßañ∂in).133 Even if, at first sight, the adharmin appear to prosper and enjoy good fortune (adharmeñaidhate tåvattato bhadråñi pa†yati),134 the author of the treatise is concerned to reiterate that their final end is destruction (vina†yati).135 Inviting his readers to reflect on such individuals, he shows that it is in nobody’s interests to follow such a path of ‘irreligion’ (adharma), since all it leads to is punishment.136 Clearly, he contemplates this contingency with astonishment, being well aware what damage such visions of the world can cause to the stability of his tradition, undermining its theoretical foundations. The author of the Månavadharma†åstra then sketches out baleful scenarios where –as the punitive power of the righteous king fails and under the pressure of such atheists and dissidents– traditional society is entirely overwhelmed, reverting to the injustices of the ‘law of the strongest’ (måtsyanyåya) and the misery caused by the violation 128

See Månavadharma†åstra, 3.153; 11.9-10. See Månavadharma†åstra, 4.30. 130 See Månavadharma†åstra, 4.60-61; 4.84-86; 4.163. 131 See Månavadharma†åstra, 4.87-91. 132 See Månavadharma†åstra, 4.170-173. 133 See Månavadharma†åstra, 5.89-90. 134 See Månavadharma†åstra, 4.174. 135 See Månavadharma†åstra, 4.174. 136 See Månavadharma†åstra, 6.61-64. 129

V. Violating Tradition and its Boundaries

469

of the regular performance of ritual practices.137 It is thus that the world order is overthrown and the earth is exposed to awesome and devastating phenomena: The entire realm, stricken with famine and pestilence, quickly perishes, when it is teeming with ‡üdras, overrun by infidels [nåstika], and devoid of twice-born people [åkråntamdvijam].138

Pursuing the argument, he seeks ‘natural’ reasons in order to drive further his attempt to preserve the cultural statutes of which he is spokesman. It’s easy to see that the need to predict such social destinies did not develop from nothing, but was triggered by threats that were real or seen as such. So much so that the fervour to pursue the clash with the ‘unrighteous’ (adhårmika) people drives the author to even suggest their suppression, articulated with varying degrees of intensity.139 A suppression of the ‘sinner’ (påpin) which assumes the features of a struggle in which religious salvation is at stake, through a comparison with ritual purifying activities (dvijåtaya ivejyåbhiΔ püyante satata∫).140 The fear that the ethical degeneration –seen in terms of sexual violation and modes of intercourse– caused by the heretics and by other visions of the world could rupture certain equilibriums and lead towards a ‘deviation from dharma’ (that, in its turn, ‘leads to the uprooting and destruction of everything’ [yena mülaharo’dharmaΔ sarvanå†åya kalpate]),141 is effectively at the basis of the argument of several †loka in the eighth section of the Månavadharma†åstra. Here, ethics, society, dharma and power are all part of a single enterprise. Not incidentally there is further reference to dharma, understood as the ‘given’ –and thus inviolable– order of all social class137

See Månavadharma†åstra, 7.20-21. See Månavadharma†åstra, 8.22. 139 See Månavadharma†åstra, 8.310. Note here the interesting –albeit decidedly later– references to the practice of eliminating those who do not confirm to a given dharma present in some works of Jayantabha™™a (850-910 CE). References to the suppression of the nœlåmbara sect from the kingdom of ‡aõkaravarman (883-902 CE) are present in his Nyåyama∞jarœ (“Some rakes, as we are told, invented this Black Blanket Observance, in which men and women wrapped together in a single black veil make various movements. King ‡aõkaravarman, who was conversant with the true nature of Dharma [dharmatattvaj∞aΔ], suppressed this practice, because he knew that it was unprecedented, but he did not [suppress] the religions of Jains and others in the same way” [asitaikapa™anivœtåviyutastrœpu∫savihitabahuceß™am Ù nœlåmbaravratam ida∫ kila kalpitam åsœt vi™aiΔ kai†cit ÙÙ tad apürvam iti viditvå nivårayåm åsa dharmatattvaj∞åΔ Ù råjå †aõkaravarmå na punar jainådimatam evam ÙÙ] [vol. 1, p. 649, r. 4-7. Citation and translation from Dezso 2004: ix]), and in his own satirical work (see Ågama∂ambara, 2.150-155; 3.10-19; 3.34-36: “36. Those criminal false ascetics, however, who devastate the established social and religious order –if they don’t leave immediately, the king will strike them like thieves” [ye tu prastutadharmaviplavakr¢taΔ påpås tapopåyinas te cedå†u na yånti ghåtayati tån dasyün iva kßmåpatiΔ]. Translation from Dezso 2004: xii). The same Jayantabha™™a referred to his ideological opponents as ‘wretched’ (varåka). See, on these aspects of Jayantabha™™a’s intellectual production, Bhattacharya 2002c; Wezler 1976. 140 See Månavadharma†åstra, 8.311. 141 See Månavadharma†åstra, 8.353. 138

470

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

es and occupations.142 If this is the way things are, it is logical that the author of the treatise demands the expulsion from urban centers of gamblers, bandits, performers, entertainers, impostors, liquor vendors, heretics and the unorthodox, in order to avoid that their ‘deleterious practices’ (vikarmakriyå) may pave the way to the corruption of the customs of all ‘pious people’ (bhadrikapraja).143 And if these are effectively the premises, then it is consistent that an advocate of tradition such as the author of the Månavadharma†åstra should continue his discourse by proposing an actual political doctrine for the ‘eradication of thorns’ (kañ™akoddharaña). Thorns that are, indeed, considered the first and last reason for all deviations from the ‘regulations’ (†åstra); for every transgression from the social order decreed by dharma; for every deviation from the ‘årya way of life that is defended [by the right sovereign]’ (rakßañåd åryavr¢ttånå∫); and hence for all distancing from tradition.144 In order to avert the possibility of a conflict with the exponents of different visions of the world (but not only, since punishment is envisaged even for those who make their living through the dharma but who then deviate from it),145 the author enjoins his ideal readers –ranging from bråhmañas in training to ministers and advisers to the sovereign– to act in such a manner that the †üdra, a social group that he sees as a serious threat to his cultural tradition, cannot amass goods. If this happens, he warns, they could use them for very dangerous ends: “Even a capable ‡üdra must not accumulate wealth; for when a ‡üdra becomes wealthy, he harasses Brahmins [bråhmañåneva bådhate]”.146 The degree of fear regarding the threat represented by the existence of ‘other’ visions can, moreover, be understood by looking at the list of violations to ‘right conduct’ (sadåcåra) –typical of the Brahmanic literature on dharma– which abound in references to forms of conflict with individuals that the author sees as external to his sphere of values. When, for example, he condemns ‘lying about one’s status’ (anr¢ta∫ ca samutkarße), ‘making false accusations against an old person’ (guro† cålœkanirbandhaΔ), ‘violating the Veda’ (brahmojjhatå), ‘disclosing the Veda’ (vedanindå), ‘eating food that is banned or unsuitable’ (garhitånådyayor jagdhiΔ),147 ‘remaining without the sacred fire’ (anåhitågnitå), ‘studying fallacious treatises’ (asacchåstrådhigamana∫) and ‘being unfaithful’ (nåstikya∫),148 he is ener142

See Månavadharma†åstra, 8.352-353. See Månavadharma†åstra, 9.225-226. 144 See Månavadharma†åstra, 9.252-293. What has been said supra cannot replace the reading of this whole section of the treatise. This section is, indeed, full of descriptions that the author gives of the different social factors that he comes into conflict with in order to head off their threats or to stifle their diffusion. 145 See Månavadharma†åstra, 9.273. 146 See Månavadharma†åstra, 10.129. 147 See Månavadharma†åstra, 11.56-57. 148 See Månavadharma†åstra, 11.66-67. 143

V. Violating Tradition and its Boundaries

471

getically fighting the diffusion of visions, practices and customs that he perceives as a challenge to his system of belief. Finally, reasserting the difficulty of bearing comparison with persuasions external to his dharma, the author again indicates the risk of diverting oneself in the study of non-orthodox texts (asacchåstrådhigamana∫).149 Such works, here defined as ‘external to the Veda’ (vedabåhya), are extraneous to the category of ‘revelation’ (†ruti) and therefore –since they are of different nature– not durable (utpadyante vyathante ca yånyato ’nyåni kånicit) and of modern origin (tåny arvåkkålikatayå). They must be stigmatised by those who defend ‘tradition’ (smr¢ti). Studying these works and making reference to ‘errant visions’ –practices which were evidently widespread at the time when the treatise was written– are indicated as causes of doctrinal deviation (kudr¢ß™ayaΔ) and as actions incapable of producing any good result. Effectively, it is stated that: The scriptures that are outside the Veda, as well as every kind of fallacious doctrine –all these bear no fruit after death, for tradition takes them to be founded on Darkness. 96. All those different from the Veda that spring up and then flounder –they are false and bear no fruit, because they belong to recent times. (Månavadharma†åstra, 12.95-96)

The author of the Månavadharma†åstra, driven by the necessity to protect the cultural implications of the forma mentis that shaped him, thus prepares an exemplary treatise (†åstra) that outdoes its predecessors –in terms of its form, content and style–, hoping to assure new and long life to the ideas comprised within it. In view of the results, I can say that he clearly succeeded in reestablishing, and therefore renewing, his tradition.150 8. Texts, traditions and interests: an opportunity for revaluing the relation between objective and cognitive structures Considering intellectual production in the above terms invite us to look at the social reasons of a given text, or of a series of texts belonging to a given ‘tradition’. Searching for such reasons implies looking for a social context which is not only broader and more complex than what the text suggests, but also more tangibly involved in the shaping of the very intellectual contents that we are dealing with. This also implies that it is possible, within a given text, to trace the signs –albeit transfigured– of the objective and historical dimensions which marked out the contours of the cultural sphere at the time when it was written. 149

Cfr. Månavadharma†åstra, 11.66. The Månavadharma†åstra is, actually, an example of the extraordinary innovative power of its author and how a classical legal tradition can transform and re-qualify itself. See, on the primacy of the Månavadharma†åstra, Squarcini (forthcoming): chap. 3.5; 3.6. 150

472

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

It is known, however, that setting an intellectual production within a systemic context –and hence in a competitive and interdependent environment–, is tantamount to giving the intellectual practice a relative autonomy. This needs to be done carefully, so as to avoid lapsing into facile extremism. On one hand, the approach which favours the autonomous conception of the the≠rêin, should not push us to the point of forgetting the objectivity of the agonistic dimension underlying every act of veridiction; on the other hand, it must be clear that while it is undoubtedly useful to look at the intellectual production of a tradition with an awareness that “[t]raditions are governed by norms”, and that in their progress they abide by “[…] standards of fidelity and determinants of truthfulness”,151 such production is often the outcome of an original and innovative creation. In other words, all intellectual productions, emerging from the work of a single agent or of a series of agents, are influenced both by the impact deriving from the measures dictated by his own cognitive structure (which, since it provides the basic theoretical orientation necessary to successfully operate within certain epistemic classifications and categories, demands different degrees of observance, coherence and loyalty) and that deriving from the objective structures of the field of which the production itself is part. And since the objective structures guarantee the social status and the degree of visibility within the specific cultural sphere from which the set of agents acquire the legitimisation and justification for their action –and hence their concrete survival–, these structures expect that the correctness of their ideological and political models will be confirmed and strengthened.152 This is why it is necessary to adopt a multifaceted methodology in order to deal appropriately with such a variety of factors: a complex combination of aspects that influences extra-individual dynamics and also has an effect on the very practical and cognitive experience of those taking part in a dispute. Dissidents and antagonists, therefore, have to be an integral part of the analysis of the arguments through which a given tradition launches itself, for example, into a diatribe on what is dharma and what is adharma. It is clear at this point that, finding ourselves faced with cases similar to those addressed in the previous paragraphs, it is no longer justified to describe them as mere diatribes on theoretical and abstract principles. On the contrary, the concrete bonds and the interests concerning reality and the tangible factors surrounding the ‘authors’ 151

Ganeri 2005: 43. These remarkable implications need to be included among the elements for the interpretation of a text, especially because they have the same influence on the intellectual production as those cognitive functions that Bourdieu attributes to ‘structuring and structured structures’. See Bourdieu 2003; Bourdieu 2000: 96-97; Bourdieu 1999: 164166; Bourdieu 1998a: 103-104; 155; 159; 177-178; 181; 191; Bourdieu 1998b: 1-6; Bourdieu 1995: 151-152; Bourdieu 1991: 1-3; 13-14. 152

V. Violating Tradition and its Boundaries

473

of such argumentations are so strong that they can motivate and affect the very articulation of their thought.153 Every case in which the plane of the existing –with its laws and its rules of play– is the direct object of an intellectual reflection, offers proof of the fact that being an [o]bject of knowledge for the agents who practise it, the economic and social world exerts an action which takes the form, not of a mechanical determination, but of an effect of knowledge.154

So, we need to be able to grasp the consequences of this effect of knowledge, the objective of which is to confirm or re-formulate both the conditions of its origin and the positions historically occupied by its ‘producers’ –thus contributing to hold firm the possibilities of reproduction of the entire system of meaning that it comes from. This is a consequence of the fact that, reversing the terms of the question, “the very structures of the world are present in the structures (or rather the cognitive schemes) that the agents use to understand it […]”.155 It’s difficult to think otherwise, since […] the correspondence between the objective divisions and the schemes of classification, between the objective structures and the mental structures, is at the origin of the original adhesion to the established order.156

And if all this is reasonable, then the analysis of the contents of a text that belongs to a majority tradition cannot disregard the contribution made by the study of agonistic relations that are reflected –in a transfigured and ‘derealised’ manner– in the symbolic systems and in the social representations utilised within it.157 Seen from this perspective, a text itself enables us to grasp fragments of the social world from which it emerged and within which its tradition was entirely merged.

153 A striking example on this matter is the relation –clearly demonstrated in works that go from the tenth century CE on– between a specific technical production, that is legal and political, and the court interests. In such context, most authors of legal epitomes (nibandha) would get explicit commissions from sovereigns or they were sovereigns themselves. See, for some of these works, Rocher 2002; Jayaswal 1936; Jayaswal, Banerji-Sastri 1925: vol. 11; 1925-1926: vol. 12; Dave 1962; Kane 1953-1954; Mådhava, Parå†aramådhavœya; Råjå ~o∂armal, ~o∂aranandam (ed. Vaidya 1948). 154 Bourdieu 1988: 121 (my translation). 155 Bourdieu 1998a: 159 (my translation). 156 Bourdieu 1988: 121 (my translation). 157 See Bourdieu 1983: 311-315.

474

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

BIBLIOGRAPHY Agamben, G. (2003), Stato di eccezione, Bollati Boringhieri, Torino. Aklujkar, A.N. (1989a), The number of pramåñas according to Bhart®hari, in «Wiener Zeitschrift fr die Kunde Sdasiens», 33, pp. 151-158. ________ (1989b), Pråmåñya in the Philosophy of the Grammarians, in A. Kumar, et al. (eds.), Studies in Indology: Prof. Rasik Vihari Joshi Felicitation Volume, Shree, Delhi, pp. 15-28. Allchin, F.R. (1995) (ed.), The Archaeology of Early Historic South Asia. The Emergence of Cities and States, Cambridge University Press, Cambridge. Arnold, D. (2001), Intrinsic Validity Reconsidered: A Sympathetic Study of the Mœmå∫saka Inversion of Buddhist Epistemology, in «Journal of Indian Philosophy», 29, n. 3, pp. 589-675. Bailey, G., Mabbett, I. (2003), The Sociology of Early Buddhism, Cambridge University Press, Cambridge. Balcerowicz, P. (2005), Pramåñas and Language: A Dispute Between Diõnåga, Dharmakœrti and Akalaõka, in «Journal of Indian Philosophy», 33, n. 4, pp. 343-400. Ballanfat, M. (1997), Les matérialistes dans l’Inde ancienne, L’Harmattan, Paris. Bandyopadhyay, N. (1979), The Buddhist theory of relation between pramå and pramåña. A comparative estimate in relation to the Samkhya-Yoga, the Advaita, the Mimamsaka and the Jaina theories, in «Journal of Indian Philosophy», 7, pp. 43-78. Barua, D.L. (1937), A few Evidences on the Age of the Kathåvatthu, in «The Indian Historical Quarterly», 6, n. 2, pp. 367-370. Basham, A.L. (1951), History and Doctrines of the Åjœvikas. A Vanished Indian Religion, Luzac and Company, London. Benveniste, E. (2001), Il vocabolario delle istituzioni indoeuropee, Einaudi, Torino (repr.), vol. 1. Berlinerblau, J. (2001), Towards a Sociology of Heresy, Hortodoxy, and Doxa, in «History of Religions», 40, n. 4, pp. 327-351. Bhattacharya, G.M. (1977), ‡åbdabodha as a separate type of pramåña, in «Journal of Indian Philosophy», 5, n. 1-2, pp. 73-84. Bhattacharya, K. (1973), L’Åtman-Brahman dans le buddhisme ancien, Ecole Francaise d’Extrême-Orient, Paris. Bhattacharya, R. (1997), Origin of Materialism in India: Patrician or Plebeian?, in «Bharatiya Samajik Chintan», 20, n. 1-2, pp. 12-23. ________ (1998a), On lokåyata and lokåyatana in Buddhist Sanskrit, in «Annals of the Bhandarkar Oriental Research Institute», LXXIX, pp. 231-235. ________ (1998b), Så∫khya, Yoga and Lokåyata in the Kau™ilœya Artha†åstra: A Review, in «Bharatiya Samajik Chintan», 21, n. 1-2, pp. 70-74.

V. Violating Tradition and its Boundaries

475

________ (1999), Paurandarasütra Revisited, in «Journal of Indian Philosophy», 27, pp. 485-497. ________ (2000a), Perception and Inference in the Cårvåka Philosophy, in «Journal of the Asiatic Society (of Bengal)», XLII, n. 12, pp. 29-38. ________ (2000b), The Significance of lokåyata in Pali, in «Journal of the Department of Pali Calcutta University», 10, pp. 39-46. ________ (2000c), Yogåcåra against the Cårvåka: a critical survey of Tattvasa∫graha, Chapter 22, in «Anvikßa», 21, pp. 46-55 ________ (2002a), Cårvåka fragments: a new collection, in «Journal of Indian Philosophy», 30, n. 6, pp. 597-640. ________ (2002b), Haribhadra’s ¯a∂dar†anasamuccaya: Verses 81-84: A Study, in «Jain Journal», 36, n. 3, pp. 134-148. ________ (2002c), Jayantabha™™a’s Representation of the Cårvåka: A Critique, in J. Heidrich, H. Rustau, D. Weidemann (eds.), Indian Culture: Continuity and Discontinuity. In Memory of Walter Ruben (1899-1982), Abhandlungen der LeibnizSozietat, Band 9, Trafo Verlag, Berlin, pp. 85-93. Bhattacharyya, S. (1987), Doubt, Belief, and Knowledge, Indian Council of Philosophical Research, Delhi. Bollée, W.B. (1999), Adda or the Oldest Extant Dispute Between Jains and Heretics (Süyaga∂a, 2,6). Part Two, in «Journal of Indian Philosophy», 27, n. 5, pp. 411-437. ________ (2002), The Story of Paesi (Paesi - kahåñaya∫): Soul and Body in Ancient India. A Dialogue on Materialism. Text, Translation, Notes and Glossary, Harrassowitz Verlag, Wiesbaden. Bourdieu, P. (1983), The Field of Cultural Production, or: The Economic World Reversed, in «Poetics», 12, n. 4-5, pp. 311-356. ________ (1988), La parola e il potere. L’economia degli scambi linguistici, Guida, Napoli. ________ (1988), Vive la crise! For heterodoxy in social science, in «Theory and Society», 17, n. 5, pp. 773-788. ________ (1991), Genesis and structure of the religious field, in «Comparative Social Research», 13, pp. 1-44. ________ (1995), Ragioni pratiche, Il Mulino, Bologna. ________ (1998a), Meditazioni pascaliane, Feltrinelli, Milano. ________ (1998b), The State Nobility. Elite Schools in the Field of Power, Polity Press, Cambridge (repr.). ________ (1999), Language and Symbolic Power, Harvard University Press, Cambridge (repr.). ________ (2000), Outline of a Theory of Practice, Cambridge University Press, Cambridge (repr.). ________ (2003), Sul potere simbolico, in A. Boschetti, La rivoluzione simbolica di Pierre Bourdieu, Marsilio, Venezia, pp. 119-129. ________, Passeron, J.-C. (2000), Reproduction in Education, Society and Culture, Sage Publications, London (repr.).

476

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Bronkhorst, J. (1998), The Two Sources of Indian Asceticism, Motilal Banarsidass, Delhi (repr.). ________ (2000a), The Riddle of the Jainas and Åjœvikas in Early Buddhist Literature, in «Journal of Indian Philosophy», 28, pp. 511-529. ________ (2000b), Karma and Teleology: A problem and its solutions in Indian philosophy, The International Institute for Buddhist Studies, Tokyo. Butzenberger, K. (1996), On doubting what there is not: the doctrine of doubt and reference of terms in Indian Grammar, Logic, and Philosophy of Language, in «Journal of Indian Philosophy», 24, n. 4, pp. 363-406. Chakravarti, R. (ed.) (2005), Trade in Early India, Oxford University Press, Delhi (repr.). Champakalakshmi, R. (1996), Trade, ideology, and urbanization: South India 300 BC to AD 1300, Oxford University Press, Delhi. Colas-Chauhan, U. (2002), Umåpati on pråmåñya, in «Journal of Indian Philosophy», 30, pp. 305-338. Colebrooke, H.T. (1837), Miscellaneous Essays, W.H. Allen, London, vol. 1. Collins, R. (2001), The Sociology of Philosophies. A Global Theory of Intellectual Change, Harvard University Press, Harvard (repr.). ________ (2002), On the Acrimoniousness of Intellectual Disputes, in «Common Knowledge», 8, n. 1, pp. 47-70. Collins, S. (1993), The discourse on what is primary (Agga∞∞a-sutta), in «Journal of Indian Philosophy», 21, n. 4, pp. 301-393. Coquet, J.-Cl. (1983), Les discours de la véridiction, in H. Parret (ed.), De la croyance. Approches épistémologiques et sémiotiques/On Believing. Epistemological and Semiotic Approaches, De Gruyter, Berlin-New York, pp. 62-69. Dasgupta, S.N. (1988), A History of Indian Philosophy, Motilal Banarsidass, Delhi (repr.), vol. 3. Dave, J.H. (ed.) (1962), Vyavaharaprakasa of King Prithvichandra, Bharatiya Vidya Bhavan, Bombay, 2 voll. Davis, T.W.R. (1977), Vinaya Texts, Pali Texts Society, Oxford. Deshpande, M.M. (1993), Sanskrit and Prakrit: Sociolinguistic Issues, Motilal Banarsidass Delhi. ________ (1997), Who inspired Påñini. Reconstructing the hindu and buddhist counter-claims, in «Journal of American Oriental Society», 117, n. 3, pp. 444-465. ________ (1999), What to do with the Anåryas? Dharmic Discourses of Inclusion and Exclusion, in J. Bronkhorst, M.M. Deshpande (ed.), Aryan and Non-Aryan in South Asia. Evidence, Interpretation and Ideology, Harvard Oriental Series, Opera Minora vol. 3, Cambridge, pp. 107-127. Dezso, C. (2004), ‘Much Ado About Religion’. A Critical Edition and Annotated Translation of the Ågama∂ambara, a Satirical Play

V. Violating Tradition and its Boundaries

477

by the ninth century Kashmirian philosopher Bha™™a Jayanta, PhD Thesis, Oxford 2004. Doniger, W. (1971), The Origin of Heresy in Hindu Mythology, in «History of Religions», 10, n. 4, pp. 284-287. Douglas, M. (1990), Come pensano le istituzioni, Mulino, Bologna. Dreyfus, G. (1991), Dharmakœrti’s Definition of pramåña and its Interpreters, in E. Steinkellner (eds.), Studies in the Buddhist Epistemological Tradition. Proceedings of the Second International Dharmakœrti Conference Vienna, June 11-16, 1989, Verlag der Osterreichischen Akademie der Wissenschaften, Wien, pp. 19-38. Durante, M. (1976), Sulla preistoria della tradizione poetica greca, Edizioni dell’Ateneo, Roma, vol. 2. Eisenstadt, S.N., Kahane, R., Shulman, D. (ed.) (1984), Orthodoxy, Heterodoxy and Dissent in India, Walter de Gruyter, Leiden. Eltschinger, V. (2000), ‘Caste’ et philosophie bouddhique. Continuité de quelques arguments bouddhiques contre le traitment réaliste des dénominations sociales, Arbeitskreis Fr Tibetische und Buddhistische Studien Universität Wien, Wien. Erdosy, G. (1988), Urbanisation in Early Historic India, BAR International Series, Oxford. Fortier, F. (1997), Les stratégies textuelles de Michel Foucault. Un enjeu de véridiction, Nuit blanche éditeur, coll. Littératures, Québec. Franci, G.R. (1975), Studi sul pensiero indiano, CLUE, Bologna. Franco, E. (1983), Studies in the Tattvopaplavasi∫ha I. The Criterion of Truth, in «Journal of Indian Philosophy», 11, pp. 147-166. ________ (1984), Studies in the Tattvopaplavasi∫ha II. The Theory of Error, in «Journal of Indian Philosophy», 12, pp. 105-137. ________ (1994), Perception, knowledge, and disbelief: a study of Jayarå†i’s scepticism, Motilal Banarsidass, Delhi. ________ (1997), Dharmakœrti on Compassion and Rebirth, Wiener Studien zur Tibetologie und Buddhismuskunde 38, Wien. Frauwallner, E. (1957), Vasubandhu’s Vådavidhi, in «Wiener Zeitschrift fr die Kunde Sd- und Ost- Asiens», 1, pp. 104-134. Freiberger, O. (1998), The ideal sacrifice. Patterns of reinterpreting brahmin sacrifice in early Buddhism, in «Bullettin d’Ètudes Indiennes», 16, pp. 39-49. Ganeri, J. (2001), Argumentation, Dialogue and the Kathåvatthu, in «Journal of Indian Philosophy» 29, pp. 485-493. ________ (2005), Traditions of Truth: Changing Beliefs and the Nature of Inquiry, in «Journal of Indian Philosophy», 33, n. 1, pp. 4354. Gokhale, P.P. (1993), The Cårvåka theory of pramåña, in «Philosophy East and West», 43, pp. 665-682. ________ (ed.) (1993), Vådanyåya of Dharmakœrti. The Logic of Debate, Sri Satguru Publications, Delhi. Goldman, R.P. (1997), Eßa dharmaΔ sanåtanaΔ: Shifting Moral Values

478

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

and the Indian Epics, in P. Bilimoria, J.N. Mohanty (ed.), Relativism, Suffering and Beyond. Essays in Memory of Bimal K. Matilal, Oxford University Press, Delhi, pp. 187-223. Gombrich, R. (1997), How Buddhism Began. The conditioned Genesis of the Early Teachings, Munshiram Manoharlal, Delhi (repr.). Gonda, J. (1975), Vedic Literature, Harrassowitz, Wiesbaden. ________ (1981), Le religioni dell’India. Veda e antico induismo, Jaca Book, Milano. ________ (1984), The Vision of the Vedic Poets, Munshiram Manoharlal, Delhi (repr.) ________ (1997), Change and Continuity in Indian Religion, Munshiram Manoharlal, Delhi (repr.). Gonzalez-Reimann, L. (2002), The Mahåbhårata and the Yugas: India’s Great Epic Poem and the Hindu System of World Ages, Peter Lang, New York. Greimas, A.J. (1980), Le contrat de véridiction, in «Continental Philosophy Review», 13, n. 3-4, pp. 345-355. Gupta, B. (1981), Skepticism: Ancient ‘East’ and Modern ‘West’, in «Indian Philosophycal Quarterly», 9, pp. 29-44. Halbfass, W. (1990), India and Europe. An Essay in Philosophical Understanding, Motilal Banarsidass, Delhi (repr.). ________ (1991), Tradition and Reflection. Exploration in Indian Thought, State University of New York Press, Albany. Hardy, F. (1990), Hinduism, in U. King (ed.), Turning Points in Religious Studies. Essays in Honour of Geoffrey Parrinder, T & T Clark, Edinburgh, pp. 145-155. Heesterman, J.C. (1985), The Inner Conflict of Tradition: Essays in Indian Ritual, Kingship, and Society, University of Chicago Press, Chicago. Hegde, R.D. (1982), The nature and number of pramåñas according to the Lokåyata system, in «Annals of the Bhandharkar Oriental Research Institute», 63, pp. 99-120. Hiltebeitel, A. (2001), Rethinking the Mahåbhårata. A Reader’s Guide to the Education of the Dharma King, Chicago University Press, Chicago. Hopkins, E.W. (1925), Words of defamation in sanskrit legal language, in «Journal of the American Oriental Society», 45, pp. 39-50. Houben, J.E.M. (1999), Why did rationality thrive, but hardly survive in Kapila’s system? On the pramåñas, rationality and irrationality in Så∫khya (part I), in «Asiatische Studien», 53, pp. 491-512. ________ (2001), ‘Verschriftlichung’ and the Relation between the Pramåñas in the History of Så∫khya (Why did Rationality Thrive but Hardly Survive in Kapila’s ‘System’? Part II), in J. Bronkhorst (eds.), Études de Lettres 3: La Rationialité en Asie / Rationality in Asia, Faculté des Lettres, Univ. de Lausanne, Dorigny, pp. 165-194.

V. Violating Tradition and its Boundaries

479

Hudson, D.D. (1998), The Uses of Delusion: Conflicts Between Religions in Ancient India, in «Common Knowledge», 7, n. 2, pp. 57-72. Inden, R. (1992), Changes in the Vedic Priesthood, in A.W. van den Hoek, D.H.A. Kolff, M.S. Oort (ed.), Ritual, State and History in South Asia. Essays in honour of J.C. Heesterman, Brill, Leiden, pp. 556-577. Jayantabha™™a, Agamadåmbara, edited and translated by C. Dezso, ‘Much Ado About Religion’. A Critical Edition and Annotated Translation of the Ågama∂ambara, a Satirical Play by the ninth century Kashmirian philosopher Bha™™a Jayanta, PhD Thesis, Oxford 2004. Jayaswal, K.P. (ed.) (1936), The Råjanœti-Ratnåkara by Chañ∂e†vara, The Bihar and Orissa Research Society, Patna (sec. ed.). ________, Banerji-Sastri, A. (ed.) (1925), Bha™™ha-Svåmin’s Commentary on Kau™ilya’s Artha-‡åstra, in «Journal of the Bihar and Orissa Research Society», vol. 11, parts I-III. ________, Banerji-Sastri, A. (ed.) (1925-1926), Bha™™ha-Svåmin’s Commentary on Kau™ilya’s Artha-‡åstra, in «Journal of the Bihar and Orissa Research Society», vol. 12, parts I-III. Kane, P.V. (1953-1954), King Bhoja and His Works on Dharma†åstra and Astrology, in «Journal of Oriental Research», 23, pp. 94-127. ________ (1993), History of Dharma†åstra, Bhandarkar Oriental Research Institute, Poona (repr.), vol. III. ________ (1997), History of Dharma†åstra, Bhandarkar Oriental Research Institute, Poona (repr.), vol. II.1. Kangle, R.P. (ed.) (1997), The Kau™ilœya Artha†åstra, Motilal Banarsidass, Delhi (repr.), vol. 3. Kataoka, K. (2003), The Mœmåμså Definition of pramåña as a Source of New Information, in «Journal of Indian Philosophy», 31, n. 13, pp. 89-103. Kathåvatthu, edited by A.C. Taylor (for the Pali Text Society Series, n. 48-49), Routledge and Kegan Paul, London 1894, vol. 1; London 1897, vol. 2. Kaviraj, S. (ed.) (1999), Politics in India, Oxford University Press, Delhi. ________ (2005), The Sudden Death of Sanskrit Knowledge, in «Journal of Indian Philosophy», 33, n. 1, pp. 119-142. Kellner, B. (2003), Integrating Negative Knowledge into pramåña Theory: The Development of the d®†yånupalabdhi in Dharmakœrti’s Earlier Works, in «Journal of Indian Philosophy», 31, n. 1-3, pp. 121-159. King, R. (1997), Early Advaita Vedånta and Buddhism, Sri Satguru Publications, Delhi (repr.). Kosambi, D.D. (2002), Early Brahmins and Brahminism, in B.D. Chattopadhyaya (ed.), D.D. Kosambi. Combined Methods in Indology and Other Writings, Oxford University Press, Delhi, pp. 87-97.

480

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Krasser, H. (2001), On Dharmakœrti’s Understanding of pramåñabhüta and His Definition of pramåña, in «Wiener Zeitschrift fr die Kunde Sdasiens», 45, pp. 173-199. Lahiri, N. (1999), The Archaeology of Indian Trade Routes Upto c. 200 BC, Oxford University Press, Delhi (repr.). Landowski, E. (1988), Vérité et véridiction en droit, in «Droit et Société», 8, pp. 47-63 (eng. trans. as E. Landowski, Truth and veridiction in Law, in «International Journal for the Semiotics of Law», 2(1989), n. 1, pp. 29-47). Lindtner, C. (1998), Buddhism as Brahmanism, in «Indologica Taurinensia», XXIII-XXIV[1997-1998], pp. 217-246. Lingat, R. (1999), The Classical Law of India, Oxford University Press, Delhi (repr.). Macdonell, A.A., Keith, A.B. (1982), Vedic Index of Names and Subjects, Motilal Banarsidass, Delhi 1982 (repr.), vol. 2. Masset, D. (2000), Une lecture bouddhiste des mythes hindous: la mise en question des modèles brahmaniques dans le Canon bouddhique tibétain, in M.-L. Barazer, J. Fezas (ed.), Le norme et son application dans le monde indien, Ecole Francaise Extrem Orient, Paris, pp. 155-173. Matilal, B.K. (1985), Scepticism and Mysticism, in «Journal of the American Oriental Society», 105, n. 3, pp. 479-484. ________ (1998), The Character of Logic in India, State University of New York Press, Albany. Menon, M.S. (1987-1988), Caricature of ‘Lokayata Dar†ana’ or materialism in ancient India, in «‡rœ Veõka™e†vara University Oriental Journal», 30-31, pp. 81-86. Minkowski, C.Z. (1991), Priesthood in Ancient India. A Study of the Maitråvaruña Priest, Publications of the De Nobili Research Library, Vienna. Mohanty, J.N. (1992), Reason and Tradition in Indian Thought. An Essay on the Nature of Indian Philosophical Thinking, Oxford University Press, New York. Morrison, K.D. (1995), Trade, Urbanism, and Agricultural Expansion: Buddhist Monastic Institutions and the State in the Early Historic Western Deccan, in «World Archaeology», 27, n. 2, pp. 203-221. Morton Smith, D. (1974), The early heresies in the development of Indian religion, in «Indologica Taurinensia», 2, pp. 149-198. ________ (1977), Addendum to the Early Heresies (Indo. Taur. Vol. II), in «Indologica Taurinensia», 5, pp. 177-178. Nakamura, H. (1955), Upanisadic Tradition and The Early School of Vedånta as Noticed in Buddhist Scripture, in «Harvard Journal of Asiatic Studies», 18, n. 1-2, pp. 74-104. ________ (1961), Indian heterodoxies in comparative light, in «Adyar Library Bullettin» 25, pp. 550-558. Namai, C.M. (1976), A Survey of Bårhaspatya Philosophy, in «Indological Review (Kyoto)», 2, pp. 29-74.

V. Violating Tradition and its Boundaries

481

Oetke, C. (2003), Limitations of Theories of Pramåña, in «Journal of Indian Philosophy», 31, n. 1-3, pp. 199-227. Oguibénine, B. (1985), Essais sur la culture védique et Indo-européenne, Giardini Editori, Pisa. ________ (1998), Essays on Vedic and Indo-european Culture, Motilal Banarsidass, Delhi. Olivelle, P. (1993), The ņrama System. The History and Hermeneutics of a Religious Institution, Oxford University Press, New York. ________ (ed.) (2000), Dharmasütras. The Law Codes of Åpastamba, Gautama, Baudhåyana and Vasiß™ha, Motilal Banarsidass, Delhi. ________ (2002), Structure and Composition of the Månava Dharma†åstra, in «Journal of Indian Philosophy», 30, n. 6, pp. 535-574. ________ (ed.) (2004), Dharma: Studies in Its Semantic, Cultural, and Religious History, (Special issue) in «Journal of Indian Philosophy» 32, n. 5-6, pp. 421-873. ________ (ed.) (2005), Manu’s Code of Law. A Critical Edition and Translation of the Månava-Dharma†åstra, Oxford University Press, New York. ________ (forthcoming), Manu and Gautama. A Study in ‡åstric Intertextuality, in E. Franco, K. Preisendanz (ed.), Halbfass Memorial Volume, Vienna. Osier, J.-P. (2000), Une critique satirique des normes brahmaniques chez les jaina: Dhürtåkhyåna de Haribadra et Dharmaparœkßå de Harißeña, in M.-L. Barazer, J. Fezas (ed.), Le norme et son application dans le monde indien, Ecole Francaise Extrem Orient, Paris, pp. 139-153. Parpola, A. (1975-1976), Sanskrit kala- ‘time’, Dravidian kal ‘leg’, and the mythical cow of the four yugas, in «Indologica Taurinensia», 3-4, pp. 361-378. Patton, L. (ed.) (1994), Authority, Anxiety, and Canon. Essays in Vedic Interpretation, State University of New York Press, Albany. Phuntsho, K. (2005), Shifting Boundaries: pramåña and Ontology in Dharmakœrti’s Epistemology, in «Journal of Indian Philosophy», 33, n. 4, pp. 401-419. Piantelli, M. (1996), Lo Hindüismo, in G. Filoramo (ed.), Storia delle Religioni. 4. Religioni dell’India e dell’Estremo Oriente, Laterza, Roma-Bari, pp. 49-131. Pollock, S. (1990), From Discourse of Ritual to Discourse of Power in Sanskrit Culture, in «Journal of Ritual Studies», 4, n. 2, pp. 291-320. ________ (2005a), The ‘revelation’ of ‘tradition’: ‡ruti, Smr¢ti, and the Sanskrit Discourse of Power, in F. Squarcini (ed.), Boundaries, Dynamics and Construction of Traditions in South Asia, Firenze University Press - Munshiram Manoharlal, Firenze. ________ (2005b), The Ends of Man at the End of Premodernity, Royal

482

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Netherlands Academy of Arts and Sciences, Amsterdam. Potter, K.H. (1992), Does pråmåñya mean truth?, in «Asiatische Studien», 46, n. 1, pp. 352-366. Råjå ~o∂armal, ~o∂aranandam, edited by P.L. Vaidya, Anup Sanskrit Library, Bikaner 1948. Råjårådhåkånta, D. (1961), ‡abdakalpadruma, Chowkhamba Sanskrit Series, Varanasi (repr.), 3 voll. Rangaswami Iyengar, H.R.R. (1953), The Vådavidhi and the Vådavidhåna of Vasubandhu, in «Adyar Library Bulletin», 17, pp. 9-20. Renou, L. (1960), Le destin du Véda dans l’Inde, in «Etudes Védiques et paninéennes», Tomo VI, Editions E. De Boccard, Paris. Rhys Davids, T.W. (ed.) (2000), Dialogues of the Buddha, Motilal Banarsidass, Delhi (repr.), vol. 3. Rocher, L. (ed.) (2002), Jœmütavåhana’s Dåyabhåga. The Hindu Law of Inheritance in Bengal, Oxford University Press, New York. Ronzitti, R. (2001), Testimonianze di ateismo nel ¥gveda, in M. Mariani, R. Ronzitti (ed.), Ricerche di linguistica diacronica, prospettica e retrospettiva, Edizioni dell’Orso, Alessandria, pp. 21-34. Sarma, R.N. (1989-1990), Pramåñas in the Pråbhåkara-mœmåμså, in «Indologica Taurinensia», XV-XVI, pp. 319-324. Scharfe, H. (1989), The State in Indian Tradition, Brill, Leiden. Seyfort Ruegg, D. (1994), Pramåñabhüta, *pramåña(bhüta)-purußa, pratyakßadharman and såkßåtk®tadharman as epithets of the ®ßi, åcårya and tathågata in grammatical, epistemological and Madhyamaka texts, in «Bulletin of the School of Oriental and African Studies», 57, n. 2, pp. 303-320. ________ (1995), Validity and authority or cognitive rightness and pragmatic efficacy? On the concepts of pramåña, pramåñabhüta and pramåña(bhüta)purußa, in «Asiatische Studien», 49, pp. 817-828. Sharma, A. (1979-1980), A note on the nomenclature for materialist in ancient India, in «Sambodhi», 8, pp. 34-38. Sharma, R.S. (1989), Origin of the State in India, University of Bombay, Bombay. ________ (2001), The State and Varna Formation in the Mid-Ganga Plains. An Ethnoarchaeological View, Manohar, Delhi. Silburn, L. (1989), Instant et Cause: Le discontinu dans le pensée philosophique de l’Inde, De Boccard, Paris (repr.). Silk, J.A. (2002), Possible Indian Sources for the Term Tshad Ma’i Skyes BU as pramåñapurußa, in «Journal of Indian Philosophy», 30, n. 1, pp. 111-160. Slaje, W. (1998), Nasti Daive Prabhutvam: Traces of Demythologisation in Indian Epic Thought, in «Journal of Indian Philosophy», 26, n. 1, pp. 27-50. Smith, B.K. (1994a), Classifying the Universe. The Ancient Indian Varña System and the Origins of Caste, Oxford University Press, New York.

V. Violating Tradition and its Boundaries

483

________ (1994b), The Veda and the Authority of Class. Reduplicating Structures of Veda and Varña in Ancient Indian Texts, in L. Patton (ed.), Authority, Anxiety, and Canon. Essays in Vedic Interpretation, State University of New York Press, Albany, pp. 67-93. ________ (1998), Reflections on resemblance, ritual and religion, Motilal Banarsidass, Delhi (repr.). Solomon, E.A. (1976-1978), Indian Dialectics. Methods of Philosophical Discussion, Gujarat Vidya Sabha, Ahmedabad, 2 voll. Squarcini, F. (2004), Testi senza autore, autori senza testa. Appunti rispetto all’odierno dibattito sulla condizione autoriale nel campo letterario sudasiatico classico, in «Rivista degli Studi Orientali», 78 [2005], pp. 1-22. ________ (forthcoming), Violenza, norma, immaginativa politica. Fra costruzione identitaria e violenza simbolica nel Månavadharma†åstra, Firenze University Press, Firenze. Stcherbatsky, T. (2000), Buddhist Logic, Motilal Banarsidass, Delhi (repr.). Taber, J. (1992), What Did Kumårila Bha™™a Mean by svataΔ pramåñya?, in «Journal of the American Oriental Society», 112, n. 2, pp. 204-221. ________ (2001), Much Ado About Nothing: Kumårila, ‡åntarakßita, and Dharmakœrti on the Cognition of Non-being, in «Journal of the American Oriental Society», 121, n. 1, pp. 72-88. Thapar, R. (2000a), Dissent and Protest in the Early Indian Tradition, in R. Thapar, Cultural Pasts. Essays in Early Indian History, Oxford University Press, Delhi, pp. 213-234. ________ (2000b), Ethics, Religion and Social Protest in the First Millenium BC in Norther India, in R. Thapar, Cultural Pasts. Essays in Early Indian History, Oxford University Press, Delhi, pp. 856-875. ________ (2000c), The Image of the Barbarian in Early India, in R. Thapar, Cultural Pasts. Essays in Early Indian History, Oxford University Press, Delhi, pp. 234-270. ________ (2002), Early India. From the origins to AD 1300, University of California Press, Berkeley. Thompson, G. (1998), On Truth-Acts in Vedic, in «Indo-Iranian Journal», 41(1998), pp. 125-143. Tillemans, T.J.F. (1993), Persons of Authority. The sTon pa tshad ma’i skyes bur sgrub pa’i gtam of A lag sha Ngag dbang bstan dar, A Tibetan Work on the Central Religious Questions in Buddhist Epistemology, Tibetan and Indo-Tibetan Studies 5, Franz Steiner Verlag, Stuttgart. Tucci, G. (1928), Vådavidhi, in «Indian Historical Quarterly», 4, pp. 630-636. ________ (1971), Linee di una storia del materialismo indiano, in G. Tucci, Opera Minora, Bardi Editore, Roma (repr.), vol. 1.

484

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Upadhyay, G.P. (1979), Bråhmañas in Ancient India. A Study in the Role of the Bråhmaña Class from c. 200 BC to c. AD 500, Munshiram Manoharlal, Delhi. Varadacharya, K.S. (ed.) (1969), Nyåyama∞jarœ of Jayanta Bha™™a, Oriental Research Institute, Mysore, vol. 1. Wezler, A. (1976), Zur Proklamation religiös-weltanschaulicher Toleranz bei dem indischen Philosophen Jayantabha™™a, in «Saeculum», 27, pp. 329-347. ________ (1999), Medhåtithi on såmånyato d®ß™am [anumånam], in «Journal of Indian Philosophy», 27, pp. 139-157. ________ (2004), Dharma in the Veda and the Dharma†åstras, in «Journal of Indian Philosophy», 32, n. 5-6, pp. 629-654. Witzel, M. (1995), Early Sanskritization. Origins and Development of the Kuru State, in «Electronic Journal of Vedic Studies», 1-4, pp. 1-26. ________ (1997a), ¥gvedic history: poets, chieftains and polities, in G. Erdosy (ed.), The Indo-Aryans of Ancient South Asia, Munshiram Manoharlal, Delhi (repr.), pp. 307-338. ________ (1997b), Macrocosm, mesocosm, and microcosm: the persistent nature of ‘Hindu’ beliefs and symbolic forms, in «International Journal of Hindu Studies», 1, n. 3, pp. 501-553. ________ (1997c), The Development of the Vedic Canon and its Schools: The Social and Political Milieu. (Materials on Vedic ‡åkhås, 8), in M. Witzel (ed.), Inside the Texts, Beyond the Texts. New Approaches to the Study of the Vedas, Harvard Oriental Series, Opera Minora, vol. 2, Cambridge, pp. 266-313. ________ (2003), Das Alte Indien, Verlag C.H. Beck, Mnchen. Yang Gyu An (2003), The Buddha’s Last Days, Pali Texts Society, Oxford. Zavos, J. (2001), Defending Hindu Tradition: Sanatana Dharma as a Symbol of Orthodoxy in Colonial India, in «Religion», 31, pp. 109-123. Zimmermann, M. (2000), A Mahåyånist Criticism of Artha†åstra: The Chapter on Royal Ethics in the Bodhisattva-gocaropåya-vißayavikurvaña-nirde†a-sütra, in «Annual Report of the International Research Institute for Advance Buddhology at Soka University», pp. 177-211.

ANTONIO RIGOPOULOS The Nonconformity to Tradition of the Mahånubhåvs

In this paper I wish to present the historical case of the ascetic, devotional movement of the Mahånubhåvs, in an effort to highlight an exemplary case of a religious group distancing itself and becoming autonomous from bråhmañical models of transmission and tradition, which also characterized various other South Asian sects between the thirteenth and sixteenth centuries. Indeed, the doctrines, beliefs, and practices of the Mahånubhåvs are a significant testimony of the anticonventional spirit which animated their founders. The sect of the Mahånubhåvs –‘Those of the great experience’– arose, like the much more popular bhakti movement of the Vårkarœs centered in Pañ∂harpur, in thirteenth century Mahåråß™ra. These two movements, which were seminal in the origin and development of Marå™hœ literature, remained separate and independent, never coming into any significant contact with one another.1 The Mahånubhåvs nonconformity with respect to mainstream Hinduism appears evident at a first glance: the sect rejects the caste system and the entire varñå†ramadharma ideology as well as the Vedas and all bråhmañical scriptural authority; in order to safeguard their identity and avoid bråhmañical persecution Mahånubhåvs had to go underground and develop a secret script to preserve their scriptures; they accept on equal 1 The Marå™hœ scholar V.B. Kolte suggested that the founder of the Vårkarœ movement, the great J∞åndev (d. 1296), might have written his J∞åne†varœ as a direct counterresponse to Mahånubhåv doctrine (see Kolte 1950). This hypothesis, however, seems far-fetched. Even R.D. Ranade argued that the Mahånubhåvs made current certain Yoga practices which might have influenced some of J∞åndev’s writings. Nonetheless, he observed that J∞åndev owed almost nothing or very little to this tradition (Ranade 1982: pp. 27-29). Though according to the Mahånubhåv Smr¢tistha¬ (chap. 244) it would have been a Mahånubhåv to turn the thoughts of the Vårkarœ saint-poet Nåmdev (1270-1350) to Kr¢ßña, inspiring his song of repentance My days have passed to no purpose, this is most probably a hagiographic invention.

486

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

terms both untouchables and women and created an order of female renouncers alongside one of men; they compound asceticism and devotion in a rigorous and at the same time original way, which reinforces their sectarian, elitist character; they are strict monotheists and devalue the entire Hindü pantheon (except Kr¢ßña and Dattåtreya) repudiating the bråhmañical ritual apparatus and the worship of gods (devatåpüjå); philosophically, they appear to be the sole bhakti group to embrace a Vedåntic dualism (dvaita), opposite to the non-dualist devotionalism (advaita bhakti) dominant among the Vårkarœs and in the whole of the Marå™hœ cultural area; their temples are famous as healing centers, to which people flock in hopes of being exhorcized and freed from malevolent spirits and demons (bhüts); finally, for some particular aspect of their doctrine and practice, the influence upon them of other religions such as Jainism and even Islåm has been postulated. Here, in order to offer a first historical overview,2 I will focus attention on the origins and main religious and doctrinal characteristics of the movement, discussing those aspects which appear especially revealing of their nonconformist attitude. 1. Origin and History of the Movement At the origin of the Mahånubhåvs there are some key figures who moulded the destiny of the entire movement. These, according to Måhånubhav doctrine, are the five manifestations (avatårs) of the One God Parame†var (‘Supreme Lord’), the sole source of isolation (kaivalya) or liberation (mokßa) to whom is directed exclusive devotion. These five manifestations are collectively called the ‘five Kr¢ßñas’ (pa∞cakr¢ßñas), comprising two deities –Kr¢ßña himself and Dattåtreya– and three sect figures: Cakradhar (d. 1274), the founder of the sect, his predecessor Guñ∂am Rå¬ (d. 1287-1288), Cakradhar’s guru, and Cåõgdev Rå¬, Guñ∂am Rå¬’s guru. The early period of the sect is dominated by the figures of Cakradhar, Guñ∂am Rå¬, and Cakradhar’s successor Någdev, also known as Bha™obås (d. 1312-1313). If, in the beginning, the Mahånubhåvs knew a fairly rapid expansion, especially in the northern and eastern regions of Mahåråß™ra –the old districts of Khånde† and Någpur, and especially the Varhå∂ or Vidarbha/Berår area, in which they have always been strongest– around the end of the fourteenth century their movement had already split into thirteen ‘sub-sects’ (åmnåya, a term often associated with ‡åkta Tantrism). The Mahånubhåvs went silently underground aiming at a defensive isolation from the larger Hindü context. They never became a popular movement and always centered themselves in remote areas, gathering in monasteries (ma™hs) situated in decayed and removed villages. To this day, the sect’s main 2 For a fuller, more exhaustive account on the Mahånubhåvs, see my forthcoming monograph The Mahånubhåvs.

V. Violating Tradition and its Boundaries

487

cult center is R¢ddhipur (modern Rœtpur) in Varhå∂, a small tumbledown village north of Amraoti: this is the town where Cakradhar met his master Guñ∂am Rå¬ and attained enlightenment from him. Although the prominent leaders among the early Mahånubhåvs were all bråhmañs (often converts from the prevailing advaita vaißñavism), their followers were and are mostly non-bråhmañs, that is, low caste people and even untouchables. A clear aversion toward the Mahånubhåvs became evident as early as the latter half of the fourteenth century. Paradigmatic of the disfavor with which they came to be looked upon by Hindüs and of their willingness to separate themselves from bråhmañical orthodoxy so as to protect their distinctiveness, was the transcribing of their sacred works, originally written in Old Marå™hœ, into various ciphers or lettersubstitution codes which they themselves invented. The most common among these ciphers and the first to be introduced around the middle of the fourteenth century was the saka¬a lipœ, the cipher of ‘all’ (saka¬a) as it was used throughout the sect, traditionally ascribed to Rava¬obås.3 In those days, the Mahånubhåvs’ adoption of a secret script was not devised out of fear of Muslim persecution, but rather out of fear of orthodox bråhmañism, which became all the more rigid in its violent opposition and persecution of the sect. Mahånubhåvs were so successful in their secretive attitude that they remained practically unknown for about five hundred years, that is, until the end of the nineteenth century. Actually, they were even able to expand beyond the borders of the Marå™hœ cultural area. Around the sixteenth century an offshoot of the Mahånubhåvs, known as the Jai Kr¢ßñi panth, developed in Pa∞jåb and as far as in what is now Pakistån, with monasteries in Lahore and Peshåwår. The Mahånubhåvs’ link with this offshoot was strongly maintained until partition. Outside of their own closed circles, and precisely because of being perceived as separate from mainstream Hinduism, the Mahånubhåvs were met with prejudices and distrust by common people, especially by the bråhmañs of the districts in which they flourished. According to D.D. Kosambi, the Mahånubhåv ‘protest group’ would go back to the ideals of a tribal, communal life: Black garments, absolute rejection of the caste system, organization into clan-like sub-groups, sharing among members, and a greatly simplified marriage ritual (ga∂a-ba∂a-guñ∂å) prove this, though a few leaders of the sect later accumulated some property, with a concomitant thirst for Hindu respectability. (Kosambi 1962: 33)4 3 The script was first deciphered in 1910 by V.K. Rajwade (see the Bhårata Itihåsa Sa∫†odhaka Mañ∂a¬a Reports, Poona, †aka 1832, p. 78; †aka 1835, pp. 58-59). For an explanatory presentation of this important cipher, invented as all other most-commonly used ciphers by members of the Upådhye sub-sect, see Raeside 1970: 328-334. 4 Kosambi also argues that “Mahånubhåvas take Så∫dœpani as Kr¢ßña’s guru” (Kosambi 1962: 24). Så∫dœpani is the name of a sage (muni) and a master-at-arms who

488

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Although this hypothesis of a sort of tribal, egalitarian ideal seems untenable, Mahånubhåvs were certainly never entirely accepted by the Marå™hœ people, being perceived as different and strange. Indeed, there are proverbs and idiomatic sayings in Marå™hœ which are derogatory of the ‘Månbhåvs’: they are said to be hypocritical and two-faced, immoral and lustful beggars who carry their sticks upside down, thieves, and cunning people in general. Bråhmañical enmity and hatred towards the sect, both in Mahåråß™ra as well as in Gujaråt, comes out very clearly in the following decree promulgated in 1782 by Mådhavråo Pe†vå: “The Manbhaus are entirely to be condemned. They are to be entirely outcasted. They have no connection with the four castes nor with the six Dar†anas. No caste should listen to their teachings. If they do, then they are to be put out of caste. (in Farquhar 1984: 322)”. For centuries Mahånubhåvs suffered in silence such offences.5 Still in 1885-1887, Sir William Wilson Hunter in The Imperial Gazetteer of India (vol. 12, p. 58) presented an account of the Mahånubhåvs which was both inaccurate and filled with popular misconceptions. In it, we read that its supposed founder, one Kishen Bhat6 said to be the spiritual guide of a king ruling in Pai™hañ around the middle of the fourteenth century, was made an outcaste because of his marriage with a woman of the lowest of Mahåråß™ra’s three untouchable castes i.e. that of the rope-makers Måtåõgas or Måõgs: the very name Månbhåv/Månbhåu is said to be derived from it.7 The professed celibacy of the male and female members of the sect –who all have their heads shaved (men also their faces) and typically wear black or ash-grey clothes perhaps in Kr¢ßña’s honor8– is also called into quesinstructed Kr¢ßña and Balaråma according to the Vißñupuråña. In my reading of Mahånubhåv literature, however, I have never come across such belief. 5 Still in the nineteenth-century, a Muslim from Ellichpur noted that there was bitter enmity between the Mahånubhåvs and the bråhmañs of the district and that, even though many people oppressed them, they never complained (Kolte 1962: 148). 6 In other ethnographic accounts, his name is given as Arjun Bhat or Krishna Bhat. 7 In the 1881 Berår Census Report, E.J. Kitts wrote (p. 62): “The Bråhmans hate the Månbhaos […]. The Bråhmans represent them as descended from one Krishna Bhat, a Bråhman who was outcasted for keeping a beautiful Mång woman as his mistress. His four sons were called the Mång-bhaos or Mång brothers” (in Russell 1916: 181). This article on the Månbhao (pp. 176-183), reporting various popular stories documenting bråhmañical hatred toward the sect, is said to have been compiled by combining three sources: notes on the caste drawn up by Colonel Mackenzie and contributed to the Pioneer newspaper by Mrs. Horsburgh; Captain Mackintosh’s Account of the Manbhaos (India Office Tracts); a paper by one Pyåre Lål Misra, ethnographic clerk. On the Måõgs, a term derived from Sanskrit måta∫ga, see Karve 1968: 33. For another short but useful account on the Mahånubhåvs see Farquhar 1984: 247-249. See also Gonda 1963: 177. 8 Another derogatory story put forward by bråhmañs concerning the origin of the sects’ clothing is the following: “Krishna Bhat’s followers, refusing to believe the aspersions cast on their leader by the Bråhmans, but knowing that some one among them had been guilty of the sin imputed to him, determined to decide the matter by the ordeal of fire. Having made a fire, they cast into it their own clothes and those of their guru, each man having previously written his name on his garments. The sacred fire made short work of all the clothes except those of Krishna Bhat, which it rejected and refused to burn, thereby

V. Violating Tradition and its Boundaries

489

tion, suggesting a situation of promiscuity and of illicit sexual misconduct.9 This is due to the fact that the order allows women as well as men to become ascetic renouncers10 and that Mahånubhåv monasteries even nowadays house both men and women under the same roof, though living in separate quarters. Suspicions of sexual misconduct, though unfounded, can be traced in the sacred narrative of the founder’s deeds, the Lœ¬åcaritra, since they remount to the times of Cakradhar himself (Tulpule 1996: 201-211). In 1907, the account of the Imperial Gazetteer was utilized in a court case at the Bombay High Court as evidence to acquit an important figure of the Vårkarœ movement who had been charged with having spoken offensively about the Mahånubhåvs. It was precisely this case which brought some Mahånubhåv heads of monasteries (mahants) to interrupt their long, self-imposed silence and publicly defend their order. Thus, they decided to reveal their secret scriptures to the scholar R.G. Bhandarkar –as testified in an article which he wrote in the Times of India, dated 15 November, 1907– and successfully petitioned for a thorough revision of the Imperial Gazetteer article. The revised article which appeared in the 1907-1909 version of the Imperial Gazetteer (vol. 21, p. 302) retracted the erroneous connection of the Mahånubhåvs with the Måõg caste, correctly named Cakradhar as the founder of the sect and highlighted that even though celibacy is viewed as the perfect life, the weaker brethren are allowed to marry. In another article which appeared in 1909, W. Crooke wrote that, besides their celibate section (bairågœ), householder Mahånubhåvs –called gharvåsœ– are divided into nominal adherents following caste rules (bholå) and those who ignore caste distinctions (Crooke 1909: 504).11 In the 1920s R.E. Enthoven also noted that there are forcing the unwilling disciples to believe that the finger of God pointed to their revered guru as the sinner” (Russell 1916: 181-182). The Mahånubhåvs’ wearing of dark clothes in Kr¢ßña’s honor is mentioned by various authors: for instance, Ranade observed that “it is probably due to the recognition of this deity [Kr¢ßña] that they wear dark-blue clothes” (Ranade 1982: 28). Kr¢ßña literally means black and, in iconography, he as well as Vißñu are typically represented bearing a dark-blue complexion, recalling the nocturnal sky or the dark monsoon cloud. 9 W. Crooke, noticing how Mahånubhåvs like other vaißñava sects have been accused of immorality, wrote: “In former times it is said that marriage between a monk and a nun was symbolized by the pair laying their wallets close together – a practice now denied by the members” (Crooke 1909: 504). Nonetheless, Crooke himself wrote how Mahånubhåvs “are a quiet, thrifty, orderly people” and that, although “their rejection of the manifold saints and orthodox gods has brought them into conflict with Brahmans”, yet “they are held in much respect by lower caste Hindüs” (ibidem). Also Russell, in his account of 1916, wrote the following: “The Månbhaos are intelligent and generally literate, and they lead a simple and pure life […]. Their honesty and humility are proverbial among the Kunbis, and are in pleasing contrast to the character of many of the Hindu mendicant orders” (Russell 1916: 176). 10 For a comparison with contemporary forms of female asceticism in the Hindü context, based upon a field-research conducted in Vårañåsœ between 1976 and 1981 see Denton 1991: 211-231. 11 On these divisions within the order, bearing slightly different names, see Russell 1916: 178-179.

490

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

householder Mahånubhåvs, called angvanshils or gharbårœs (the same as gharvåsœ), who marry by the gåndharva or love marriage form and, at the same time, wear the dress of the order and live in monasteries (Enthoven 1922: 430).12 The 1907-1909 events marked the renewed contact of the Mahånubhåvs with the outside world and the end of their long isolation. The heretical, even orgiastic nature attributed to the Mahånubhåvs and their writings was proved to be totally unfounded. Meanwhile, their thirteen åmnåyas or sub-sects were reduced to just two: the Upådhye and the Kavœ†var, with minimal doctrinal differences between them. Besides the emergence of a new attitude, almost a kind of missionary spirit among Mahånubhåv leaders, the coming into the open of their sacred texts stimulated a great interest among scholars. As I.M.P. Raeside puts it: “Marå™hœ scholars were astonished to find themselves presented with a whole corpus of literature much of which dated from the fourteenth century and was contemporary with the oldest works of Marå™hœ literature known up to that time”. (Raeside 1976: 586) Among the first Marå™hœ scholars who rediscovered the Mahånubhåvs in the early years of the twentieth century was V.L. Bhave. To be sure, despite the sect’s marginality these documents are most precious, being the earliest extant sources of the very beginning of Marå™hœ language. Many of their early works are in prose, not in verse, and thus provide almost the only important corpus of prose writing in Marå™hœ before the seventeenth century.13 Moreover, the Old Marå™hœ language of these early texts was to a large extent preserved, being ‘frozen’ at the stage it had reached at the time when they came to be enciphered. Thus they were not subject to modifications and modernization along the centuries. Already in 1899, B.G. Tilak, in an article published in the journal Kesarœ about his research on Marå™hœ traditions, had underlined the historical and literary importance of the Mahånubhåv sect. But the Marå™hœ scholar who in the twentieth century made the most significant contribution to the study of Mahånubhåv literature was V.B. Kolte. Besides Kolte and the abovementioned Bhave, mention should be made of S.G. Tulpule, who also wrote extensively in English (Tulpule 1979), as well as of N.B. Bhavalkar, V.N. Deshpande, Y.K. Deshpande, S.K. Joshi, N.G. Kalelkar,14 H.N. Nene, and V.K. Rajwade. Among contemporary Western scholars, the greatest authorities on the Mahånubhåvs are I.M.P. Raeside and A. Feldhaus, to whom we owe fundamental contributions. 12 Enthoven obtained all information for his article on ‘Manbhavs’ (pp. 427-433) from R.G. Bhandarkar. 13 For an overview concerning the historical emergence and development of the Marå™hœ language, see Pacquement 2000: 741-763 and Armelin 1980. 14 His French unpublished doctoral thesis, titled La secte Manbhav (Paris, 1950), appears as the earliest scholarly work in a Western language. Unfortunately, I was not able to see it.

V. Violating Tradition and its Boundaries

491

Such scholarly interest also contributed to push Mahånubhåvs out of their secretive, closed milieux. The principal Mahånubhåv leaders who opened themselves and their libraries to the outside world were Pa∞jåbœs. Starting in the 1920s, scholars have emerged even among their adepts and a few personalities among them have recreated some of the lost åmnåyas, such as the Yakßadev åmnåya. Between the two World Wars, the mahants of the Devadeve†var monastery at Måhür (the old Måtåpur) and of the Gopiråj temple at Rœtpur have played a prime role in collecting and studying Mahånubhåv works and also in helping outside scholars to understand them. Their successors, however, have not been so active and collaborative. In the mid-1970s Raeside observed: The position today is that many mahantas within the pantha are happy to take their doctrinal difficulties to Professor Kolte to be settled, for he has devoted more study to the Mahånubhåva philosophy and ritual (vicåra and åcåra) than anyone within the sect. The other half of the sect are strictly orthodox still, and refuse to disclose or even discuss Mahånubhåva beliefs with outsiders. (Raeside 1976: 589)

Nowadays, it is quite difficult to estimate the total number of Mahånubhåvs, most of whom belong to the Marå™hå caste of agriculturists. The Census of India has always counted them as Hindüs and never as a separate ‘religion’. In 1901, Enthoven estimated their number as around 22.000 (Enthoven 1922: 427-433). For the same year Crooke, quoting the 1881 Berår Census Report of E.J. Kitts, said that in Berår they numbered 2.566. Crooke added that their numbers are decreasing “perhaps due to the fact that in the present day fewer join the celibate section” (Crooke 1909: 504). In R. V. Russell’s report it is stated that in 1911 the Månbhao’s religious sect, now become a caste, counted only 10.000 members, of whom the Central Provinces and Berår contained 4.000 (Russell 1916: 176). Feldhaus has more recently suggested that “a figure of 100.000 to 200.000 today seems likely, although the numbers at pilgrimage places and one’s subjective impressions indicate more” (Feldhaus 1988: 279, n. 18). 2. Cakradhar’s Nonconformity and the Mahånubhåvs’ Difference as Reflected in their Foundational Texts In Cakradhar’s times, the kingdom in power was that of the Yådavs. Their capital was Devgirœ (=Daulatåbåd, near Auraõgåbåd) in the Marå™hvå∂å region. The kingdom’s heartland was the agricultural area of the Godåvarœ basin in northern Mahåråß™ra, which came to be extended: from here, the Yådavs tried to expand their rule to much of the rest of the actual State of Mahåråß™ra as well as to other parts of the Deccan. Although the Yådav kingdom was small, it considered itself as a great force from the Arabian sea to the central regions of the subcontinent. The Yådav army fought against Gujaråt

492

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

and Må¬vå to the north and against the Hoysa¬a empire to the south. Under the leadership of King Siõghaña early in the thirteenth century, the Yådavs invaded Gujaråt and other regions, establishing a dominion which, however, lasted only a few years.15 Presenting themselves as orthodox bråhmañs intent upon the restoration of the sacred Vedas, the Yådav rulers claimed to govern society strictly following the precepts of the caste system and of the bråhmañical orders and stages of life (varñå†ramadharma, caturvarñya).16 Most importantly, the Yådav kings spoke Marå™hœ and patronized the Marå™hœ language in a variety of ways. The kingdom played a crucial role in the rise of the Marå™hœ language and in the emergence of the self-awareness of an entire region as a cultural whole. Significantly, J∞åndev explicitly mentions King Råmcandra as “the delight of the Yådav race” (J∞åne†varœ, 18.1783) and refers to the kingdom’s capital Devgirœ as “the city of the Marå™hœ language” upon which he prays that “the blessed day of the knowledge of the Absolute” may dawn (J∞åne†varœ, 12.16).17 Hemådrœ was the prime minister or chancellor (mahåkarañådhipa) of King Mahådev (reigned 1261-1270), and was also associated with Mahådev’s successor Råmcandra or Råmdev (reigned 12711311). He was the author of various works on dharma†åstra and of the famous Caturvargacintåmañi, a kind of encyclopaedia of religious rites and observances reflecting the orthodox view concerning the four legitimate aims of human life (the purußårthas: dharma, artha, kåma, and mokßa). The Caturvargacintåmañi was meant as a guideline for a society to be administered according to the rules of dharma and caste, that is, jåti and varña prescriptions and regulations. Hemådrœ is also credited for the building of many temples in what is known as the hemå∂pantœ style, which he would have invented or favored, and for creating a rapid, cursive script for Marå™hœ, the mo∂œ script. In contrast to the institutional, ritualistic orthodoxy of the Yådavs, stood the two great promoters of devotional movements of thirteenth century Mahåråß™ra: J∞åndev and Cakradhar. Their religiosity of love, aimed at cultivating a direct, intimate relationship with one’s chosen deity, met the aspirations of a vast majority of the population and especially of women and low caste people, †üdras as well as untouchables. However, whereas J∞åndev through the development of his lay Vårkarœ movement didn’t constitute any serious threat to the bråhmañical socio-religious order, Cakradhar, through the development of his ascetic movement, was less careful in his dealings with social rules, especially for what concerned ritual purity and his liberal attitudes toward women. Though he did not 15 For a historical overview of the Yådav kingdom of Devgirœ, see Bhandarkar 1957: 116-131; Ganguly 19662: 185-197; Verma 1970. 16 On the general religious conditions during Yådav rule, see Verma 1970: 294-333. 17 For the English translation of the J∞åne†varœ, see Swami Kripananda 1989.

V. Violating Tradition and its Boundaries

493

actively rebel against bråhmañism –being rather indifferent to it– Cakradhar inevitably tended to be anti-conventional and heterodox. He devalued the Vedas and his sect developed as an anti-bråhmañical, anti-Vedic group. His religiosity was opposite to that of Hemådrœ’s, as when in the Lœ¬åcaritra (uttarårdha 585) he advises his followers not to distinguish between commandments and prohibitions (vidhinißedha), a mahåtmå being beyond both. In the same teaching concerning the dharma to be observed by his disciples in his absence (asannidhån), he recommends a life of detachment and asceticism (sannyås): one should have no likes and dislikes and nothing to do with buying or selling; one should avoid staying in towns and cities and even avoid visiting popular pilgrimage places and going to fairs and festivals (kßetras and jåtras); above all, one should avoid violence in any form. The rejection of the need to visit any pilgrimage place or sacred ford (tœrth), as well as his condemnation of the practice of making vows and gifts (vrat, dån), is another sign of his ascetic nature and of the transcendence of established ritual and religion, be it bråhmañical or even folk (Lœ¬åcaritra, uttarårdha 316). Cakradhar emphatically underlined that one should abandon faith in tœrth, kßetra, vrat, and dån, and that going to such ‘sacred’ places such as Dvårkå, Vårañåsœ, etc. would not lead one to liberation. Rather, he taught that his very presence and any service (dåsya) rendered to him or to God with full faith (bhåv) would grant mokßa. Again, such a teaching was diametrically opposed to the ritualist religion upheld by Hemådrœ, as is evidenced in the latter’s Caturvargacintåmañi. Moreover, Cakradhar did not observe the traditional rules of pollution in his dealings with women –for instance, at the time of their menstrual period– and with untouchables such as Måõgs or Camårs (the traditional cobbler caste). He was always accessible to them, never erecting any ‘walls’ or barriers and often miraculously proving his spiritual ‘oneness’ with them, saving them from the wrath of people and even from legal expiation (pråya†citta) and capital punishment when they themselves came to break pollution rules (see, for instance, Lœ¬åcaritra, pürvårdha 27 and Lœ¬åcaritra, uttarårdha 72, 102, 384). The Muslim invasion of the Deccan began in 1294 and had widely spread by 1318, upsetting the whole of Mahåråß™rian culture. Passing with his army of apparently only 8,000 men near Elichpur (=A¬ajpur), the capital of Varhå∂ less than twenty miles away from R¢ddhipur, ‘Alå-ud-dœn Khaljœ (d. 1316), nephew of the Delhi Sultan, was able to make a surprise attack on Devgirœ in 1294. He utterly overpowered the Yådav’s army, pillaging and carrying off a great treasure, and forcing King Råmdev –the last independent sovereign of the Deccan– to negotiate peace: Råmdev was allowed to remain in power though having to pay heavy annual tributes. It was from Devgirœ that ‘Alå-ud-dœn Khaljœ launched his successful campaign for the throne in Delhi in 1296, after arranging for the killing of his uncle

494

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

the Sultan. Several raids of the Deccan by the Muslim armies followed. Sometime around 1310, King Råmdev was captured and sent to Delhi as prisoner by Malik Kåphür, an officer of ‘Alå-ud-dœn Khaljœ. The Yådav king was nonetheless allowed to return to the Deccan, again as a tribute-paying vassal. He died soon afterwards and some years later Råmdev’s son-in-law, Harapål, having revolted against the Muslim foreigner, paid his offence by being flayed alive in 1318. With his tragic death, the Yådav’s dynasty came definitely to an end. From now on, Persian became the court language of the Muslim rulers in the Deccan, though the administrative system remained that which Hemådrœ had established. In 1327, sovereign Muhammad ibnTughlaq (reigned 1325-1351) –the Tughlaqs having supplanted the Khaljœ rulers in Delhi– made Daulatåbåd the second capital of his empire: it even appears that some Mahånubhåv elders had meetings with him. Muhammad ibn-Tughlaq’s conquests, however, were shortlived and paved the way for local Muslim dynasties. By 1350, the Bahåmanœ kingdom with its capital outside the Marå™hœ-speaking area was in full power. Though the Muslim kings were no cruel rulers, Marå™hœ ethos and literature was impeded or altogether halted for about two centuries. The founder of the Mahånubhåvs, Cakradhar, left behind no writings. In the earliest period, Cakradhar’s charisma was thought to supplant all scriptural authority. The only source of valid, authoritative knowledge was believed to come directly from Parame†var or one of his manifestations. There is clearly an anti-intellectual tendency, especially critical of bråhmañical, orthodox learning and opposed to the use of Sanskrit. The sole exception is represented by the Bhagavadgœtå which the Mahånubhåvs, like the Vårkarœs (J∞åndev’s magnum opus is the J∞åne†varœ, a commentary on the Bhagavadgœtå) hold in special reverence. Mahånubhåvs believe that the Bhagavadgœtå was spoken by the avatår ‡rœ Kr¢ßña in person, whereas all else is attributed to Vyås (Sütrapå™h, 11.108).18 Thus, all texts of the Hindü tradition were and are radically devalued.19 However, as all new religious movements necessitating recognition in order to stress their difference and supe18 For Sütrapå™h’s verses on the Bhagavadgœtå, see Sütrapå™h, 10.85-90. Sütrapå™h, 13.153 is a commentary of Bhagavadgœtå, 7,16 ( “Of these the possessor of knowledge, constantly disciplined, of single devotion, is the best; for extremely dear to the possessor of knowledge am I, and he is dear to Me” [trans. from Edgerton 1964: 39]). The Sütrapå™h’s commentary, in Feldhaus’s translation (Feldhaus 1983), reads: “One who longs [for Parame†var] has his mind troubled by sorrow; one who desires knowledge is the best with respect to me; one who aims at the goal is expectant of an eternal abode; the believer goes to the highest by means of faith”. 19 Sütrapå™h, 11.109 declares that Puråñas and Ågamas are merely related to the devatåcakra, the ‘wheel of deities’ comprising the whole of the Hindu pantheon: herein, nine groups of gods are arranged in hierarchical order. The Puråñas are linked to the fourth level of the hierarchy and the Ågamas –the authoritative texts of Tantric, †aiva schools such as the Nåth sect– to the third level of the eight Bhairavas (aß™abhairavas). Even the authority of the Vedas is indirectly minimized and ultimately rejected since it does not afford access to Parame†var. Sütrapå™h, 10.14 says that some portions of the Vedas,

V. Violating Tradition and its Boundaries

495

riority, after Cakradhar’s times Mahånubhåvs developed their own body of authoritative scriptures which came to be bound by secret (as the Sütrapå™h, treasuring the holy words and teachings of Cakradhar, itself intimates: 12.155-157, 199; 13.27, 198).20 These scriptures were jealously guarded within the sect. As Feldhaus writes: “Having rejected the elitism of the learned by composing their scriptures in Marå™hœ rather than Sanskrit, the early Mahånubhåvas then hid the scriptures from public view. They thereby created a new elite: the Mahånubhåvas themselves” (Feldhaus 1978: 308). Cakradhar’s biography is the Lœ¬åcaritra, a prose collection of short chapters or anecdotes (Lœ¬ås: 920 in S.G. Tulpule’s edition; 1.237 in V.B. Kolte’s edition).21 These were painstakingly collected from a variety of people who had direct knowledge of the events of Cakradhar’s life by one of his disciples, Mhåï∫bha™, who subjected these stories to Någdev’s scrutiny for final approval. The lœ¬ås of the Lœ¬åcaritra are divided into two halves: these are the pürvårdha section which is the earlier section dealing mainly with Cakradhar’s early life and his period as a solitary ascetic (relying upon disciples’ accounts of Cakradhar’s own reports), and the uttarårdha section which deals with the later part of Cakradhar’s life. The dividing point between the two halves is Någdev’s becoming a follower of Cakradhar. Every word uttered by Cakradhar was written down by Mhåï∫bha™ as he could remember them, after consulting with the other disciples. Dated 1278 and thus composed soon after Cakradhar’s death, it has an authentic ring given the simplicity and antiquity of its language, reflecting the sect’s early events as well as the social and religious conditions of the Yådav kingdom. According to tradition, the original work of Mhåï∫bha™ was lost or stolen in the political upheaval (the dhå∂a) which took place during the Deccan raids of the Muslim army in either 1307-1308 or 1310, when the Yådav king Råmdev was captured and sent to Delhi as prisoner by Malik Kåphür. As a consequence, the Lœ¬åcari™ra had to be reconstructed from the memory of the disciples (Hœråœsa, Kavœ†var, Para†aråm). This reconstruction process of the Lœ¬åcaritra, which brought to the elaboration of different versions of the text, continued all during the fourteenth century. Soon afterwards and still during the lifetime of Någdev, Kesobås (=Ke†ava vyåsa), who probably joined the sect after Guñ∂am Rå¬’s death, produced an epitome of Cakradhar’s teachings: this is the possibly the Upanißads, know of the existence of Caitanya. But Caitanya for Mahånubhåvs is simply the highest level in the hierarchy of relative, non-ultimate devatås which are not Parame†var and which are qualitatively separate and, therefore, not conducive to him. 20 Besides the impure ones loaded with rajas and tamas, whose karman is bad and who are devoted to devatås, among the ones who are specifically mentioned as not fit to receive the teachings contained in the scriptures are the very old (Sütrapå™h, 11.257), the mad and possessed ones (Sütrapå™h, 10.258; 13.150), as well as the blind, the deaf, and the mute (Sütrapå™h, 13.150). 21 Herein, I follow Kolte’s 1982 edition which subdivides the text in 592 (pürvårdha) and 645 (uttarårdha) lœ¬ås.

496

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Sütrapå™h, a collection of ‘aphorisms’ (sütras) culled from the Lœ¬åcaritra under Någdev’s direction and guidance.22 Apparently, the Sütrapå™h was also lost or stolen during the Muslim invasion of Malik Kåphür and subsequently rewritten/reconstructed by the disciples Kavœ†var, Para†aråm, and Råme†var. Consisting as we now have it in nine short chapters or navprakarañ (anyavyåvr¢tti, yugdharma, vidyåmårg, sa∫hår, sa∫sarañ, mahåvåkya, nirvacan, uddharañ, asatœparœ) and four longer ones (åcår, åcår målikå, vicår, vicår målikå), the Sütrapå™h is the veritable ‘Bible’ for all Mahånubhåvs, their most important doctrinal text, originally datable between 1287-1288 (Guñ∂am Rå¬’s death) and 1302 (Någdev’s traditional year of death). Some manuscripts also add three other short chapters to the navprakarañ: the pürvœ, the pa∞cakr¢ßña, and the pa∞canåm. These three, however, as well as the åcår målikå and vicår målikå, are taken to be later additions. The sayings attributed to Cakradhar follow no coherent order, being connected together in an unsystematic fashion. Kesobås, basing himself upon Någdev’s memories, also wrote a sort of appendix to the Sütrapå™h, that is, the Dr¢ß™åntapå™h, which is a collection of all the stories that Cakradhar utilized in order to illustrate his teachings. Certainly both the Lœ¬åcaritra and the Sütrapå™h were for some time floating, oral ‘texts’ before becoming literary documents. These two seminal sources grew up together, depending upon the various oral traditions which influenced each other and which had their origin in the memories of the first disciples. Thus both the Lœ¬åcaritra and the Sütrapå™h were built up and finally composed by different authors over a considerable length of time within the different, even rival, sub-sects of the Mahånubhåv order.23 A significant number of later Mahånubhåv works are in the form of commentaries or philosophical elaborations of these foundational texts. The events which took place between Cakradhar’s death and Någdev’s death can be unveiled thanks to the hagiography of Cakradhar’s predecessor Guñ∂am Rå¬, and the hagiography of Cakradhar’s successor Någdev. After Guñ∂am Rå¬’s death in 12871288, Någdev became the leader of the Mahånubhåv fold until his own death, which most probably took place about twenty-five years later (in 1312-1313). The hagiography of Guñ∂am Rå¬, the R¢ddhipurlœ¬å or R¢ddhipurcaritra24 (also known as ‡rœgovindaprabhucaritra, Govindaprabhu25 being the Sanskritized version of his Marå™hœ name), is traditionally ascribed, as all prose biographies of 22 For the Sütrapå™h, see A. Feldhaus’s authoritative edition (Feldhaus 1983). All quotes are taken from her translation. 23 On the relationship between the Sütrapå™h and the Lœ¬åcaritra, see Feldhaus 1983: 16-20. 24 For a fine English translation, see Feldhaus 1984. 25 Prabhu is a name of the Supreme Lord, often identifying Parame†var, meaning ‘the mighty or powerful one’. Besides Govindaprabhu, in Mahånubhåv literature it is usually added as a suffix to the name of Dattåtreya.

V. Violating Tradition and its Boundaries

497

the pa∞cakr¢ßñas are, to Mhåï∫bha™, who wrote it within six months of Guñ∂am Rå¬’s death, utilizing the testimony of Någdev and other adepts who had personally known Guñ∂am Rå¬. It comprises 323 episodes of his life and is traditionally dated 1287. This text, however, in not the prototype of the manuscripts we have. This is due to the fact that even this text is said to have been lost or stolen by highway robbers when Mahånubhåvs were crossing the mountains on their way to the Koõkañ area with their precious holy books. Again, this might have happened during the Deccan raids of the Muslim army in either 1307-1308 or 1310. As a consequence, the hagiography had to be reconstructed from the disciples’ memory. Kolte has arrived at the conclusion that the R¢ddhipurlœ¬å we now have dates to 1392 and was prepared by one Dattobås of Ta¬egåv, being the composite reconstruction of several people’s memories of Mhåï∫bha™’s original text. Coming back to Cakradhar, the Lœ¬åcaritra (pürvårdha 20) reports the crucial encounter between him and Guñ∂am Rå¬, in which Cakradhar received a food offering (prasåd) and both j∞ån†akti (‘the power of knowledge’) as well as paråvara†akti (‘the high-and-low powers’) from the divine lunatic who placed his hand on Cakradhar’s mouth. This meeting is said to have taken place in the bazaar at R¢ddhipur and lasted just a few minutes. Cakradhar then left him and started wandering around in solitude, totally indifferent to the world. In Mahånubhåv sources, Cakradhar always magnifies Guñ∂am Rå¬’s life as a pure divine ‘play’ (lœ¬å): the latter is revered as one who ‘delights in the Self’ (åtmåråmu), identical with the ‘Absolute beyond attributes’ (nœrguña brahma), the ‘eternally liberated Reality’ (nœtyamukta vastu). Apparently, though Cakradhar later sent his disciples to R¢ddhipur to serve Guñ∂am Rå¬ even for long periods, he himself returned to visit Guñ∂am Rå¬ only twice, the second time receiving a quite rough reception (Lœ¬åcaritra, pürvårdha, 33-37): Guñ∂am Rå¬, being choleric and unpredictable, most often received people with curses and blows. Cakradhar wandered twelve years in the wilderness before he ‘reemerged’ and started attracting disciples. He led an itinerant and solitary life, always begging his food, never staying in one place for more than a few days. His temporary abodes were out-of-the-way places at the foot of trees or dilapidated temples. His insights were grounded in self-discipline and the pursuit of ethical values, among which non-violence (ahi∫så) was paramount. The twelve-year-period of itinerant life in the forest (vanavåsa) is often recommended in Hindü asceticism, number twelve being symbolic of perfection and completion of a cycle (Schimmel 1993). Cakradhar is then reported to have spent some time at Warangal in Åndhra Prade†, where he strangely married with the daughter of a rich merchant. By this paradoxical, antinomian behaviour he pos-

498

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

sibly wished to prove the freedom of his divine, avadhüt nature, beyond attachment and non-attachment, and thus beyond all caste and family rules and restrictions. He was unpredictable and, on occasions, acted in bizarre and apparently non-dharmic ways. The name of his wife was Ha∫såmbå. However, he was really married only to renunciation and soon left her and started living again as an avadhüt in the wilderness. Cakradhar’s mastery of alchemy is also mentioned: in this early period, he is said to have received the power of arresting the aging process from a Nåth adept, a råjguru called Udhalinåth. Later on (Lœ¬åcaritra, pürvårdha 315), in a meeting with another Nåth yogin by name Vi†vanåth, Cakradhar will declare his ability to turn copper into gold, that is, of being a master at alchemical transubstantiation. His itinerant life of aimless wandering ended when he came to the ancient city of Pai™hañ on the Godåvarœ, which at that time was the center of learning and bråhmañ orthodoxy and which Cakradhar made his headquarters (Feldhaus 1991). He settled here at the instance of one Båïså: this woman was his first follower, whom he probably met in 1266. This marks the end of the ekåõka period in Cakradhar’s life, that is, the period in which he was alone. It should be noticed that, although he came from Gujaråt, the Lœ¬åcaritra (uttarårdha 133) says that he spoke Marå™hœ fluently. This has led to hypothesize that he and his family, though residents in Gujaråt, actually came from Mahåråß™ra. This opinion, however, is contradicted by the tradition that his father didn’t want him to go on pilgrimage to Råmtek in Mahåråß™ra, said to be a foreign land. The last eight years of Cakradhar’s life, from 1266 to 1274, are meticulously reported in the Lœ¬åcaritra. This is the period of his association with a group of followers in which women, especially elderly and poor women to whom he gave shelter, outnumbered men. In the Lœ¬åcaritra, the memories of the disciples who were witness to his teachings, his many miracles, and all the various events which took place in those early days are presented in a very realistic way. Countering the obscurity of his origins and early life, here detailed accounts are offered of even the most minute incidents in Cakradhar’s everyday life and of his wanderings in the various locales along the Godåvarœ. All stories are written with freshness, in a naïve, terse, and popular style reflecting the language and culture of village Mahåråß™ra of the time. About half-way through these eight years, around 1270, Cakradhar was joined by Någdev who soon became his closest male disciple and inseparable companion. Cakradhar is always revered as an incarnation of Parame†var. He himself was perfectly self-conscious of his divine identity (Sütrapå™h, 11.19). Apparently, in these last years his fame spread so much that even court ministers and the Yådav rulers Kr¢ßñadev and Mahådev wished to meet him. King Kr¢ßñadev (1245-1261), in the Lœ¬åcaritra better known as Kånherdev or Kånhadev, is reported to have once

V. Violating Tradition and its Boundaries

499

met with Cakradhar. Then the Gosåvœ,26 as Cakradhar was commonly called, gave proof of his supreme indifference, when the king offered him a quantity of gold coins (Lœ¬åcaritra, pürvårdhaå 61). We are also told that Kr¢ßñadev’s son and successor, i.e. King Mahadev (1261-1270), made more than one attempt to have dar†an of Cakradhar, the latter’s fame having spread to such an extent that the king once found his court empty. Cakradhar, however, is reported to have slipped away knowing that the king might hand over his kingdom to him (Lœ¬åcaritra, pürvårdha 225-227). The Gosåvœ’s refusal highlights his ascetic aloofness from the worldly domain, as well as his willingness to give himself completely to his followers and especially to the poor and downtrodden. In the Lœ¬åcaritra, uttarårdha 564, Cakradhar says of himself: “a walking and speaking god is hard to get in the world”. On various occasions, physical contact with Cakradhar, or his mere touch or glance, is said to have had healing power (Lœ¬åcaritra, pürvårdha 54, 92, 116, 232) or even the power of ‘deifying’ a person and rendering him/her invulnerable (Lœ¬åcaritra, purvardha 27). Though Cakradhar led his followers with great firmness and required adherence to strict discipline, still he was known as a vedhåcårya, a ‘master of attraction’, like Kr¢ßña. Apparently, the Gosåvœ was quite handsome and extremely charismatic. Women were especially attracted by him, and this, in turn, attracted criticism.27 In fact, Cakradhar was not a womaniser at all. He was a staunch renouncer who, adhering to traditional male ascetic views, regarded woman as the chief ‘intoxicating substance’ and always taught his male followers to be on guard.28 On the other hand, Cakradhar recognized equal religious rights for women, granting them initiation into renunciation (sannyås dœkßå, also known as the ‘mendicant life’, bhikßå), their soul being recognized as the same as that of men (Lœ¬åcaritra, uttarårdha 102). He especially taught his disciples to transcend the conventional ideas of bodily purity and pollution (Lœ¬åcaritra, uttarårdha 384, 537). In the main, Cakradhar followed general ethical principles (sådhårañadharma) and there are many examples in the holy texts purporting his love, kindness, patience, forgiveness, courage, forbearance, knowledge, modesty, obedience to his guru Guñ∂am Rå¬, etc. Moreover, when confronted with his lay followers he did not underestimate the importance of family life and of the gr¢hasthadharma, and often took interest in their problems: he would offer solutions to property claims (Lœ¬åcaritra, pürvårdha 387), and once granted a son 26 This term identifies a person who has renounced all worldly ties and pleasures, an ascetic. 27 On women’s attraction for Cakradhar, see Lœ¬åcaritra, pürvårdha 56 and 97, uttarårdha 80 and 266. A criticism of this female presence around him can be found in Lœ¬åcaritra, uttarårdha 474. 28 On the intoxicating properties of women, who intoxicate just by being seen, see Sütrapå™h, 12.9-12; 13.252.

500

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

through his miraculous powers to a childless –though polygamous– wealthy landlord (Lœ¬åcaritra, pürvårdha 63). To be sure, the granting of a male son has always been one of the most sought after graces, precisely because of the wish (and dharmic obligation) to continue one’s life in the world through progeny. Cakradhar is even reported to have sent Någdev, who had already become a renunciant, back to his village so as to organize the maintenance of his destitute (ex-)wife and to beget a son for helping her in life! When taking sannyås, one should first settle all primary, worldly duties (Lœ¬åcaritra, uttarårdha 32). Cakradhar, through his powers, is also said to have cared for the fate of sonless widows (Lœ¬åcaritra, pürvårdha 217). Precisely because he was a renouncer of a radical type beyond the worldly domain (sa∫sår) and all social norms, he would often act as moved primarily by compassion rather than justice. Cakradhar noticed how theft and adultery were most common and indeed the dharma of this kali age. Once he protected an adulterer (Lœ¬åcaritra, uttarårdha 271). As for the universal duty of asteya or abstention from theft, Cakradhar is said to have more than once protected thieves. Here we find a clear conflict between Cakradhar’s free, transcendent status –a ‘law’ into himself– and the common rules of law administered by the State, according to which it is the prime duty of the king to administer justice through punishment (dañ∂a) (Sontheimer 1982). For instance, Cakradhar condoned and took moral blame upon himself of a theft of chick-pea bundles done by his followers, which he then had roasted and ate with gusto (Lœ¬åcaritra, uttarårdha 15). On another occasion, he had a village officer (adhikårœ) free a thief who had stolen his horses: he defended him calling him ‘my companion’, forcing the officer to worship and feed him even before himself. He then helped this same thief to escape by making a folded rope out of clothes (Lœ¬åcaritra, pürvårdha 57). Although Cakradhar seems to have been conscious that people might object to the numerous presence of women among his followers (Lœ¬åcaritra, uttarårdha 266), he got especially into trouble when Dematœ, the wife of Hemådrœ, the famous chancellor of the Yådav king, became attracted to him and wished to become his disciple. As a wife, Dematœ was unhappy being much neglected by Hemådrœ. Her dar†an of Cakradhar in Pai™hañ strongly affected her, and, surprisingly, Hemådrœ came to be newly attracted to her (Lœ¬åcaritra, uttarårdha 509). When he wondered about the reason for his renewed affection towards Dematœ, she told him of her dar†an of Cakradhar and attributed the ‘miracle’ to him. This naturally caused the jealousy of the chancellor, and apparently even Cakradhar resented the disclosure by Dematœ, knowing that the orthodox bråhmañ Hemådrœ, upholder of a strict adherence to varñå†ramadharma, was against him and his followers. Moreover, one Såraõgpañ∂it, a celebrated bråhmañ scholar of Pai™hañ, was also jealous of his young wife’s devotion for Cakradhar.

V. Violating Tradition and its Boundaries

501

Despite Cakradhar’s holy character, bråhmañical opposition to him and his movement grew. Some though not all versions of the Lœ¬åcaritra narrate that, due to the hatred of some bråhmañ ministers of the Yådav king Råmdev, who was possibly a Vårkarœ devoted to god Vi™™ha¬ in Pañ∂harpur,29 Cakradhar was arrested and taken to Pai™hañ. Events reached a climax when a general court (sabhå) was instituted, in which Cakradhar –confronted by all the leading bråhmåñs of the town, the important people (mahåjanas) of all castes, as well as by Jains and members of the Nåth sect– was accused of having a special attraction to women (Lœ¬åcaritra, uttarårdha 536). The formal accusation was that of living immorally with his female disciples, and the Lœ¬åcaritra, signalling the whispering which took place in the assembly, suggests that the whole case was manipulated. At large majority (with only two persons dissociating themselves from the judgement, Måyitå Harœ and Praj∞åsågar), the decision (samaya) of the gathered assembly was to punish Cakradhar with ‘the püjå of the nose’, that is, the cutting off of his nose. Apparently, this was done right on the spot. Another source reports that even his ears were cut off. According to Mahånubhåvs, however, Cakradhar had his nose (and ears) restored miraculously and was able to continue his preaching activity. In about 1274, on the command of the Yådav king or of Hemådrœ himself, Cakradhar was arrested a second time and finally assassinated by being beheaded (Lœ¬åcaritra, uttarårdha 641). Besides Cakradhar’s anti-bråhmañism and the formal accusation of living immorally with his female adepts, some scholars speculate that he might have been put to death also for political reasons. As the son of a Gujaråtœ prince, Cakradhar alias Haripå¬ might have been involved in the conflict against the Marå™hå army of Sœõghañ Yådav (1200-1247): Les motifs de cette ‘exécution’ sont mystérieux: sa doctrine aurait dérangé l’ordre établi et la morale, mail la raison était peut-être d’ordre politique, dans la mesure où Harpaldev, en tant que prince de la famille de Broach, avait combattu les armées marathes de Singhana Yadav. (Kshirsagar, Pacquement 1999: 201, n. 12) 29 A Pañ∂harpur stone inscription (col. 1, l. 31), dated between 1273 and 1277, mentions King Råmdev as “chief of the pha∂s (=groups) in Påñ∂arœ (=Pañ∂harpur)” (Tulpule 1963: 179). Historical events of the Yådav period are mentioned here and there in Mahånubhåv literature. Useful references are especially found in the Smr¢tistha¬, chaps. 77; 83-87; 92-93; 145-146; 148-150 (Feldhaus, Tulpule 1992). Interesting is the case of queen Kåmåïså, the wife of the Yådav king Råmdev. She was very devoted to Någdev. Någdev, however, refused to initiate her as an ascetic since the land in which Cakradhar commanded his followers to stay, Mahåråß™ra (as per Sütrapå™h, 12.24), belonged to her husband. Since she could not leave/renounce her husband’s land, she was considered not fit for initiation. Någdev also instructed her not to enter the fire, that is, commit satœ of her own accord: if she did, he told her, she would certainly end up in hell since this is the destiny of all those who commit suicide. Eventually, at the time of Råmdev’s death in 1311, she was forced to commit satœ by her stepson, who cruelly had her thrown into the fire despite her cries and her implorations to spare her. For an overview of these various incidents, see Joshi 1976.

502

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Before being beheaded, Cakradhar forecasted two dreadful events: the invasions of the Muslim ‘barbarians’ (mleccha) from the north, and the coming of a most terrible famine (Lœ¬åcaritra, uttarårdha 603). Both events took place: the first Muslim invasions led by ‘Alå-ud-dœn Khaljœ started twenty years after Cakradhar’s death, in 1294, whereas the devastating famine is identified with the so-called durgådevœ famine, which took place almost a century later, in the years 1396-1407.30 The Mahånubhåvs, however, have not accepted Cakradhar’s violent death and believe that he miraculously got his head back and left ‘by the northern way’ (uttåra panth), being last seen in the holy city of Ujjain in Madhya Prade† (Lœ¬åcaritra, uttarårdha 645). The North is the direction of immortality and Cakradhar, on analogy with Dattåtreya and the mythical Nåths (all believed to be masters of the alchemical science, rasåyan), is still thought to be living incognito as a splendid immortal somewhere in a Himålayan cave.31 Though twice in the Smr¢tistha¬ (chaps. 132, 232) Cakradhar is said to have been seen in Ujjain, Någdev discouraged his disciples to rush there remembering his master’s final words: ‘Now we will meet anew’ (Sütrapå™h, 11.139; 12.262; Lœ¬åcaritra, uttarårdha 646, 655). These words are interpreted as a prohibition against searching for him, meaning that the encounter with him will take place in the next rebirth. Following Cakradhar’s ‘disappearance’ sometime in 1274 (=‡aka 1196, possibly in the month of mågh),32 the disciples decided to move to R¢ddhipur and gathered around Guñ∂am Rå¬. They stayed with him up until his death, said to have been caused by an attack of diarrhea around 1287, that is, about thirteen years after Cakradhar’s ‘departure.’ In the R¢ddhipurlœ¬å (chap. 322), when Guñ∂am Rå¬ is dying, Någdev asks him: “Lord, King ‡rœ Cakradhar entrusted us to you. Now you are leaving, Gosåvœ. So to whom have you entrusted us?” To this, Guñ∂am Rå¬ answers: “I have entrusted all these others to you, and I have entrusted you to ‡rœ Dattåtreyaprabhu”. From then onwards, in the absence of a recognizable avatår of Parame†var, the leadership of the group was taken up by the ‘charismatic’ (vedhavantœ) Någdev, whom the Mahånubhåvs understand to have been appointed as successor or ‘deputy’ (adhikarañ) by both Cakradhar and Guñ∂am Rå¬. He administered initiation to all new 30 In the Lœ¬åcaritra, uttarårdha 436, Cakradhar also predicted that in the kali age Vißñu and ‡iva temples would come to decay and that kßetrapålas such as Mairå¬a would replace them. With the fall of the Yådav empire and the decay of its hemå∂pantœ style temples even this prediction came to fulfillment. 31 For an assessment of Cakradhar’s whole life and ‘final event’ (†eva™ace∫ prakaraña), see Kolte 1952. On the immortal Dattåtreya meeting ‡aõkara at Badarœnåth and taking him by the hand to a cavern from which they were never seen to come out, see Rigopoulos 1998: 95. 32 In the past, scholars such as K.M. Munshi, S.G. Tulpule and others have privileged 1272 as the year of Cakradhar’s death. Kolte, in his ‡rœcakradhar Caritra proposed the date of 1276. Nowadays, however, most authorities agree on 1274 as the most probable date.

V. Violating Tradition and its Boundaries

503

disciples and expounded the fundamental teachings in the same way to everybody. His authority depended upon his knowledge, based upon memory, of Cakradhar’s precepts and was tempered by the frequent religious discussions and debates (dharmavårtå) which took place especially with the Mahånubhåv elders. Någdev distinguished himself from Cakradhar and the other avatårs of Parame†var, saying that he was not God but simply a jiva and a devotee of God (Smr¢tistha¬, chaps. 126, 198). People, therefore, should not worship him as God. Nonetheless, Någdev’s unique position and role inevitably brought the followers to pay him special homage. He guided the Mahånubhåvs for about twenty-five years till his own death, which probably occurred in 1312, the same year in which the Yådav kingdom fell to the Muslims. For Någdev, the Smr¢tistha¬ (chap. 260) reports that the disciples officiated the solemn rites appropriate for a yogœ. 3. Following the Example of One Who doesn’t Follow any Example: The Case of Guñ∂am Rå¬ According to Mahånubhåv pothœ tradition, Cakradhar was born in 1194 at Bharvås/Broach, in Gujaråt. He was the son of one Vœsaldev, a royal minister –perhaps a chief minister (pradhån) of King Malladev– and his original name was Haripå¬ or Harapå¬. He was married to a woman called Kama¬åïså, and he was much addicted to gambling. Apparently, he died young around 1221 but just before his body was to be cremated it was reanimated by the spirit of Cåõgdev Rå¬, also known as Cakrapåñi (another name of Vißñu, significantly being indistinguishable in meaning from Cakradhar).33 According to Raeside, this story of Cåõgdev Rå¬ being reborn as Cakradhar may have developed between Cakradhar’s death and the time of Kavœ†var. Cåõgdev Rå¬, almost certainly a Nåth yogœ believed to possess great powers (the entire fifty-two siddhis),34 lived most of his life and also voluntarily died in Dvåråvatœ or Dvårkå, the holy Kr¢ßñaite city of Kå™hiyåvår¢, in Gujaråt, believed to have been founded by Kr¢ßña himself. The Lœ¬åcaritra (pürvårdha 16) reports why Cåõgdev Rå¬ had decided to drop dead. There was a ha™hayoginœ called Kåmåkßœ who tried to lure him away from his asceticism, as she had done with many other yogœs. She wanted to have sex with him and, although he would 33 It should be noted that in the Sütrapå™h Cåõgdev Rå¬ is referred to in only four sütras from the vicår målikå section (11.51-54), most probably a later accretion and not the earliest record of Cakradhar’s own words. 34 There are many Cåõgdevs in the early period of Marå™hœ literature, the most famous being the yogœ to whom J∞åndev addressed his Cåõgdevpåsaß™œ, a poem in sixty-five quatrains magnifying the experience of non-duality. On the various Cåõgdevs and their possible links with Cåõgdev Rå¬, who has also been identified as a †aiva På†upata, see ˘here 1977. The Sütrapå™h (11.51, 54) states that Cåõgdev Rå¬ gave an infinite number of powers and also bestowed knowledge to fifty-two men or siddhas. He was probably credited with the teaching of the fifty-two siddhis. These are supposed to comprise all possible powers, derived from the fifty-two syllables (måtr¢kås) of the Sanskrit alphabet and their associated mantras.

504

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

refuse, she stubbornly sat for seven days and nights at his cell’s doorway. In the end Cåõgdev Rå¬, evidently tired of such a situation and resolute to keep to his vow of continence, decided to discard his body thanks to his extraordinary yogic powers.35 Although he was no more Haripå¬, the young prince went back to live with his wife Kama¬åïså and had a son from her. He also continued to gamble and on one occasion he lost heavily and so asked his wife to give him her golden ornaments so that he might pay off his debts. She, however, angrily refused to give him her gold. His affectionate father then intervened and generously paid off all his debts: he ordered his treasurer to go and pay the creditors the total sum of five hundred gold coins (åsus). This incident was the turning point in Harip嬒s life. He became disgusted with worldly life and decided to leave on a pilgrimage to Råm™ek, northeast of Någpür, the most important Råmaite pilgrimage center in Mahåråß™ra, in which Råm is venerated primarily as a sovereign king (Bakker 1990). Haripå¬ was never to return home. Apparently, his father didn’t want him to leave on pilgrimage, Mahåråß™ra being viewed as a foreign land. Also, the father seems to have objected to his departure by saying that there was a state of war between the Gurjars of Gujaråt and the Yådav kingdom and, Haripå¬ belonging to the kßatriya caste (råje), a bråhmañ priest should better be sent on pilgrimage in his place. However, separating himself from his retinue, he never got to Råm™ek but eventually stopped in R¢ddhipur, a small town in Varhå∂ which saw the presence of a rich variety of †aiva, vaißñava, and even Nåth temples and monasteries and the passing by of many ascetics and pilgrims on their way to Råm™ek and other holy places. Here, he finally met with the anti-conventional Guñ∂am (lit.: ‘rounded stone’) or Guñ∂am Rå¬, a paradigm of the crazy (unmatta) ascetic who was a bråhmåñ by birth and who initiated him into spiritual life and renunciation (sannyås), renaming him Cakradhar. The latter’s biography (Lœ¬åcaritra, pürvårdha 15) says that Guñ∂am Rå¬ in his early years had been formally initiated as a da†anåmœ renouncer in the order of ‡aõkara (788-820) by one Kama¬årañya. His name as a renunciant would have been Vibudhårañya (˘here 1977: 227-230). Nonetheless, as ˘here has concluded, Guñ∂am Rå¬, like Cåõgdev Rå¬, was most probably a Nåth yogœ, and his name appears in several lists of the eighty-four siddhas. Lœ¬åcaritra, pürvårdha 15, narrates that Guñ∂am Rå¬ received the ‘high-and-low powers’ (paråvara†akti) –the most complete of the three †aktis, making him a full manifestation of Parame†var–36 from 35 Despite obvious differences, this story of Cåõgdev Rå¬ calls to mind ‡aõkara’s temporary death and ‘entrance’ in the corpse of King Amarüka, which he reanimated in order to master all the secrets of erotic pleasure (kåma). See, on this episode, Piantelli 1974: 68-70. 36 The Lœ¬åcaritra, however, adds that while Guñ∂am Rå¬’s higher power was manifested, his lower power was hidden. The other two types of †akti which may characterize

V. Violating Tradition and its Boundaries

505

Cåõgdev Rå¬ when, on the bank of the Gomatœ river, the latter placed his winnowing fan on his head and hit him with his broom. Sütrapå™h, 11.52 states that the eccentric Cåõgdev Rå¬ showed the way with a winnowing fan, and knowledge with a broom. The broom or brush, which the yogœ Cåõgdev Rå¬ habitually used to sweep the streets of Dvåråvatœ with, and especially his winnowing-fan, in which he collected all dust and rubbish, might be interpreted symbolically as magic, shamanic tools. Guñ∂am Rå¬ is the most strange and unorthodox of the manifestations of Parame†var. Mahånubhåvs’ nonconformity reflects the unconventional character of their avatårs: indeed, how could it be otherwise? Guñ∂am Rå¬ is gluttonous, childish (båla), breaks pollution rules, disrupts rituals, even treats deities with disrespect (Lœ¬åcaritra, pürvårdha 168, reports that he often played with the images of deities, for instance putting his fingers into their mouth, nose, ears, eyes, etc.). He is altogether mad (ve∂å). The townspeople of R¢ddhipur,37 the out-of-the-way and yet prosperous village in which Guñ∂am Rå¬ spent most of his life and which is equated with Dvårkå/Dvåråvatœ by Mahånubhåvs,38 would occasionally say: “The Rå¬ is mad, the Rå¬ is possessed” (rå¬ ve∂å, rå¬ pœså). It is precisely his madness,39 manifested through his often rude, bizarre, aberrant behaviour, which is understood to be an essential sign of his divinity, transcending all varñas and å†ramas, all social and ritual norms. Guñ∂am Rå¬’s madness is paradigmatic of the transcendent otherness of Parame†var. The early Mahånubhåvs never denied his madness, but rather understood it as a sign of his divinity. In other words, his madness and his divinity are not thought of as contradictory but rather as complementary, madness being the special illustration of his divinity. He moves freely in society as a capricious egalitarian, utterly disregarding caste and sex barriers. For instance, he has a very special affection for Måõgs, the lowest of Mahåråß™ra’s untouchable manifestations of Parame†var are, according to Mahånubhåv theology, the paradr¢†ya type, which can ‘see the high’, and the avaradr¢†ya type, ‘which can see the low’. See Sütrapå™h, 11.20. 37 On R¢ddhipur, nowadays in the Mor†œ taluk of the Amråvatœ district, see Feldhaus 1987: 68-91. In the R¢ddhipurmåhåtmyas, it is reported that the original name of the place was Rucikå†rama or Rucikapur, the forest hermitage of a r¢ßi named Rucika. Through his ascetic powers, Rucika won the wish-granting cow Kåmadhenu, the wish-granting tree Kalpataru, and the wish-granting gem Cintåmañi. These three being called r¢ddhis, the town came to be known as R¢ddhipur. It is narrated that Rucika, in order to win his bride, had to present her parents with five hundred black-eared horses. After searching everywhere in vain –even in heaven– he finally obtained the horses from Dattåtreya and his mother at Måhür, that is, only about a hundred miles to the south. Rucika’s son was Jamadagni who later married Reñukå, the chief devœ of Måhür. The celebrated son of Reñukå and Jamadagni was Par†uråm. 38 On the identification of R¢ddhipur with Kr¢ßña’s capital, found in chap. 213 of the R¢ddhipurlœ¬å, see Feldhaus 1987: 76. 39 On Guñ∂am Rå¬’s madness, see Feldhaus 1982 and 1984:3-29. Useful for a summary as well as for a comparison with other radical saints is Zelliot 1987. On the behaviour of saints as if mad, see Kinsley 1974; McDaniel 1989; Feuerstein 1992.

506

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

castes, and he scrubs women’s back and even toys with a woman’s breast (!), though the conventions he breaks are not primarily of a sexual kind. Guñ∂am Rå¬ exhibits the perfect freedom of the Supreme, who does not know or care about any of the categories in which humans divide themselves. In his unlimited freedom, why shouldn’t or couldn’t he be mad? Guñ∂am Rå¬’s madness does not arise from any loving or devotional attitude, as is common among many saints, especially vaißñavas. Guñ∂am Rå¬ is never portrayed as a devotee: in fact, he is not devoted to anyone since he is God. Others, eventually, are devoted to him. Similar to other gurus of bhakti movements which tended to iconoclasm, Guñ∂am Rå¬ taught the paramount importance of an intimate, direct relationship with God. All exterior rituals and images (Padoux 1990; Waghorne, Cutler 1996), mürtis and temples were unimportant to him and finally to be rejected diverting attention from the sole necessary practice of inner, spiritual search. For Guñ∂am Rå¬ definitions and categories simply did not apply, he being utterly beyond them. As he gloriously sang (R¢ddhipurlœ¬å, 281): I am not a man, nor a god or Yakßa [= a semi-divine being], nor a bråhmañ, a kßatriya, a vai†ya or a †üdra. I am not a celibate; I am not a householder or a forest hermit; neither am I a mendicant, I who am innate knowledge. (Feldhaus 1984)40

Exhibiting a crazy or foolish behaviour, a sort of ‘divine intoxication’, is thought to be appropriate for a manifestation of Parame†var. As Sütrapå™h, 10.106 declares: God becomes a tortoise, he becomes a fish; he descends among the gods, he descends among men, he descends among animals. When he has descended among men, God becomes a madman, he becomes a possessed man, he becomes a mute; but a walking, talking God is rare. (Feldhaus 1983: 184)

Of course, such divine madness and possession, even such divine muteness, are thought of as altogether different from ordinary human madness and possession, which Sütrapå™h, 10.258 and 13.150 views as an evil, negative condition: people affected by such mental illnesses –as well as by blindness, deafness, and muteness– are said to be unfit to receive the Mahånubhåv teaching. Guñ∂am Rå¬’s divine madness, on the other hand, is not unique and certainly not to be interpreted as a Mahånubhåvs’ specialty: such a typology is often found among a variety of extreme renouncers, especially Nåth and Tantric yogœs. As de40 This is the only verse which is also found in the Sütrapå™h (11.a61). Guñ∂am Rå¬, like all manifestations of Parame†var, is revered as an omniscient. As it is said in the Sütrapå™h, 10.215: “My children, there is nothing he does not know. Even though he knows everything, he [acts as if he] is ignorant”.

V. Violating Tradition and its Boundaries

507

monstrating the madness of God Himself, Guñ∂am Rå¬’s crazy character resembles the madness of ‡iva, the mad god par excellence in the Hindü context. But here the exemplary model of Guñ∂am Rå¬’s behavior as a lunatic and as one possessed by demons is no doubt Dattåtreya in his well known bålonmattapi†åca characterization. In order to stress Guñ∂am Rå¬’s divinity –he was a ‘womb incarnation’ (garbhœcå) of Parame†var according to Mahånubhåv theology41– his biography highlights his precocious genius and extraordinary talent, his innate, superhuman qualities. The R¢ddhipurlœ¬å magnifies his omniscience and power over nature and all sorts of disease: among the many miracles attributed to him, we find such extraordinary feats such as raising humans from the dead (and animals as well: for instance, a donkey!). There are many other aspects of the ‘nonconformist model’ inaugurated by the pa∞cakr¢ßñas which can be detected in both Mahånubhåv doctrines and practices.42 Moreover, the disruptive force of this general attitude can be appreciated by looking at the more ample context in which the Mahånubhåvs were (and are) situated. Especially in their seminal phase the Mahånubhåvs clearly appropriated a variety of nonconformist elements from other religious groups. In turn, the Mahånubhåv sect certainly appeared as a ‘paradigm of nonconformity’ to other sampradåyas of the time, thus determining a complex set of mutual influences and borrowings. 4. Mahånubhåvs and Other Religions Scholars have noted similarities between certain aspects of Mahånubhåv theology and practice and other religious systems. For instance, I. Karve noted that, according to some, Mahånubhåvs are nearer to Jain philosophy than to any other form of Hinduism (Karve 1968: 190-191). Bråhmañs have often derogatorily linked the sect to other religions, especially Jainism and Islåm, encouraging the identification of Mahånubhåvs with these heterodox and hated mlecchas, precisely because of their perceived unorthodoxy and nonconformity. In the Yådav period, Jainism was the dominant religion of Karñå™ak, having its stronghold among merchant and trading castes 41 The Sütrapå™h (10.104-105) names three ways in which the One Parame†var takes on a måyå-body, that is, manifests himself in the realm of illusion: he takes it on in a womb (as in Kr¢ßña’s case); he raises a corpse (as in Cakradhar’s case); he pushes out a soul (jœva) from a living body (dava∂añe avatår; neither the Lœ¬åcaritra nor the Sütrapå™h give examples of this third type). Concerning Guñ∂am Rå¬, some authorities such as V.B. Kolte (1975: 199) have argued that he was an incarnation of the third type. In other words, he would have replaced a soul already in his mother’s womb. By favoring such interpretation, many present-day Mahånubhåvs argue that Guñ∂am Rå¬’s madness was not originally his own. Rather, he would have inherited it from the jœva he replaced. This interpretation, however, is unfounded if one looks at early Mahånubhåv literature. 42 On these other aspects and on the special role of the Dattåtreya unconventional icon, see Rigopoulos, forthcoming.

508

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

and receiving an ongoing royal patronage from the various dynasties in power. This is precisely the reason which has led to suggest a Jain influence over Vœra†aivism. At the time of the Yådav King Mahådev, it appears that even one of the royal preceptors was a Jain (one Bha™™årkadev). In the Marå™hœ-speaking area, one of the most important Jain centers was Kolhåpur (Verma 1970: 309-313). Feldhaus has noted that there are partially excavated Jain ruins in R¢ddhipur and that there are mentions of Jain temples (vasais) in both the Lœ¬åcaritra and the Sthånpothœ.43 Though G.B. Sardar has argued that Jains and Mahånubhåvs were never on friendly terms, he cannot fail to notice similarities, especially in the realm of ethics, venturing to suggest that Mahånubhåvs copied their monastic system from the Jains (Sardar 1969: 36-49; 133). Raeside has even observed that the Mahånubhåv monastic structure ‘has Buddhist echoes’ (Raeside 1989: vii). Certainly, it is a striking similarity that Mahånubhåvs, like the Jains, emphasize the rule of non-violence as the first and foremost dharma. Mahånubhåvs believe that in our kali age the anti-dharmic conduct par excellence is ‘violence’ (hi∫så), its practice leading straight to hell (Sütrapå™h, 2.18, 20). Still, the doctrine of an eternal damnation, which is found only in Madhva and in Jainism, seems not to have been adopted by Mahånubhåvs. Moreover, both Jain and Mahånubhåv ascetics encourage extreme forms of renunciation and self-mortification, adopt vegetarianism, and preach the religious and social equality of both males and females. The Mahånubhåvs’ insistence on the emaciation of the true ascetic (Sütrapå™h, 12.60-61; 7778), and on the practice of perpetual fasting (Sütrapå™h, 12.94) to the point of becoming ‘rattlingly thin’ (Sütrapå™h, 12.78) are cases in point. In the Lœ¬åcaritra (uttarårdha 532-534), Cakradhar is once reported to have ordered his female disciple Åüså to embrace Digambar Jain monks without her clothes on! On the other hand, we also find rules of moderation in the Sütrapa™h: the ascetic is not to starve or go naked (Sütrapå™h, 13.72; 210; 229), nor is he or she to subject himself/herself to extremes of heat and cold (Sütrapå™h, 13.68). Ph. Granoff has contrasted Jain monks –who exhibit clear limits in their interaction with the lay community and secular life as such– with the Mahånubhåv avatår Guñ∂am Rå¬, who, on the contrary, enters intimately into the domestic realm of even his female devotees, offering them concrete assistance and guidance (Granoff 2001 : 114-116). Whereas Jain monks tend to separate from the worldly sphere, presenting a minimal interaction, Guñ∂am Rå¬ would represent a model of freedom in interaction, irrespective of social and religious boundaries. As already noted, however, Guñ∂am Rå¬’s freedom is part and parcel of his characterization as an antin43 In a Mahånubhåv narrative about Haragarva/Hayagrœva (vr¢ddhåcåra 16), we find mention of the town of ‡œrpur, about forty miles south of Akolå in Varhå∂: this is a Jain pilgrimage site with a temple of Pår†vanåth.

V. Violating Tradition and its Boundaries

509

omian avadhüt. In fact, the Mahånubhåv order as a whole has also tended to separate itself sharply from society, even adopting a secret script from early times. Interestingly, in Jain Mahåpuråñas both Kr¢ßña and Datta are listed among the nine Våsudevas: Jain heroes who bravely engage in war killing their enemies. Having gone against the precept of ahi∫så, the Våsudevas are believed to be reborn in hell. However, as ‘illustrious beings’ (†alåkåpurußas), Jain scriptures say that they will remanifest in the next time-cycle as Jinas or Tœrthaõkars, the glorious ‘ford-makers’ (on analogy with Mahåvœr, the twenty-fourth and final ‘ford-maker’ of the current world-era) (Jaini 1993). In particular, via Nåthism the ubiquitous Dattåtreya appears to be linked with Jain asceticism (Rigopoulos 1998: 98). ‘King Dattåtri’ is said to have been the first convert of the twenty-second Tœrthaõkar Neminåth, and we have testimonies that Dattåtreya has come to be worshipped as Neminåth.44 On the highest peak of Guru Shikhar, at Mount Abu –one of the holiest places of Jainism from the eleventh century onwards– one can find a small cell where the foot-prints (pådukås) of Dattåtreya are venerated. On another north-western peak, there is a shrine dedicated to Anasüyå, Dattåtreya’s mother. The Dattåtreya shrines of Nåth adepts amid Jain sanctuaries around Mount Abu as well as in the Girnår area –coupled with Dattåtreya’s characterization as a digambara or ‘clad in space’– must have favored his assimilation or identification as a Jain saint.45 All in all, Mahånubhåvs’ asceticism is certainly to be ascribed to Nåth influence. Nonetheless, the strong presence of Jainism in central and southern India, coupled with the peculiar contiguity of Nåthism and Jainism, as evidenced in Dattåtreya’s case, must have led Mahånubhåvs to appropriate certain Jain features, first of all their emphasis on the practice of ahi∫så. Coming to an assessment of the Mahånubhåvs’ resemblance to Islåm, one cannot fail to recognize the following similarities: their radical monotheism, their aniconism and rejection of image worship, the supreme authority of the Mahånubhåvs’ holy book, the Sütrapå™h, over all human preceptors (on analogy with the Qur’an or the Ådigranth of the Sikhs), and also their custom of burying the dead. As K.M. Munshi has argued in the Foreword to The History and Culture of the Indian People: Hindu and Muslim saints, not unoften, had a common appeal to both the communities, and the sects of both the religions, by way of action and re-action, and sometimes by challenge, influenced each other. The Mahånubhåva sect, a non-idolatrous Krishna cult, founded by 44 Appropriations work both ways. In the Bhågavatapuråña (5.3-7), R¢ßabha, the first of the twenty-four Tœrthaõkars of the Jains, is cast in the role of partial (a∫†a) avatår of Vißñu in order to re-establish the †ramañadharma of the naked ascetics (Jaini 2000a). 45 See, on the Jain assimilation of Hindü deities in the Girnår area, Jaini 2000b: 275-276.

510

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Chakradharasvåmœ […] about the time the first Süfœ saints settled in Aurangåbåd, is an instance in point. (Munshi 1957: xviii)

A. Ahmad, in his Studies in Islamic Culture in the Indian Environment, paraphrased Munshi in total agreement with him (Ahmad 1964: 130). Even Vaudeville –following ˘here’s Musalmån marå™hœ sa∫takavœ– noted that Mahånubhåvs are “suspected of having later on imbibed a fair number of Muslim views and practices” (Vaudeville 1987: 219). Feldhaus has pointed at the significant Muslim presence in R¢ddhipur, with five mosques and two saints’ shrines (dargås). One mosque is located on the site of the Nr¢si∫ha temple which served as Guñ∂am Rå¬’s abode: it so happens that the most important Mahånubhåv pilgrimage place in R¢ddhipur is now the site of a mosque! Over the centuries, some kind of mutual exchange between Mahånubhåvs and Muslims undoubtedly took place, especially with Süfœ mystics and holy men. The case of Shåh Muni, an eighteenthcentury Muslim bhakti poet who knew Hindü theologies well and had a bent for Mahånubhåv doctrine precisely because of its monotheism, is quite famous (˘here 1967: 127-129). As N. H. Kulkarnee has written: Shah Muni’s tomb is maintained by five Muslim families who are vegetarians […] worship Kr¢ßña and regard Mahånubhåvas as their intimates […]. Since Siddhånta Bodha [Shåh Muni’s celebrated work] has a place of honour among the Mahånubhåvas […] Shah Muni also must be looked upon as a Mahånubhåva […] [The Siddhånta Bodha] contains a mixture of Mahånubhåva teachings, advaita philosophy and Puråñic stories in lyrical language. (Kulkarnee 1989: 224)

Similarities in doctrine and practice, coupled with the social proximity of the heterodox Mahånubhåvs to the Muslim mlecchas, determined curious links as well as identifications. In the Marå™hœ cultural area, Dattåtreya was appropriated by Süfœ circles at least from the time of Eknåth. It is often the case that Muslim faqœrs are popularly identified with Dattåtreya.46 By the same token, one finds revered saints within the Dattasampradåya who are clearly Muslims (Rigopoulos 1998: 135-168, 237). The paradigm of this synthetic mysticism aiming at Hindü-Muslim unity, even in Mahåråß™ra, was Kabœr, the fifteenthcentury sant of Vårañåsœ. Despite all affinities and ‘integrative encounters’, a direct, foundational Islåmic influence over Mahånubhåv theology appears untenable. Besides the Mahånubhåvs, we know of other medieval bhakti sects which are monotheist, yet there is no reason to suppose any Islåmic borrowing or dependence for any of them. Mutatis mutandis, I 46 As it has happened with the famous Såœ Båbå of ‡ir∂œ (d. 1918), nowadays the most beloved saint all across the Indian subcontinent; see Rigopoulos 1993 and Warren 1999.

V. Violating Tradition and its Boundaries

511

think that the following remarks of S. Siauve with regard to the alleged influences of Islåm on Madhva’s theology, may equally well apply to the Mahånubhåvs’ case: Mais il est impossible […] de parler d’influence de l’Islam sur la pensée de Madhva. Il y a là une certaine convergence de doctrine sur un point certes essentiel, celui de la transcendance de Dieu, mai ce point unique est enchassé chez Madhva dans un environnement si spécifiquement hindou qu’aucun musulman ne pourrait y apercevoir un reflet de sa religion. Et d’autres part cette thèse centrale prend toutes ses références à l’intérieur de sa propre tradition, dans le Veda et les Upanißad, et plus encore dans des textes du Pa∞ca-råtra qui sont premièrement des textes de rituel, le plus hindou et le plus traditionaliste qui soit. (Siauve 1971: 110-111)

Finally, although Mahånubhåvs have recurrently been accused by Hindüs of siding with the Muslim occupants and of being on friendly terms with them, this is surely not true. Mahånubhåvs never received any favors from Muslim rulers.

BIBLIOGRAPHY Ahmad, A. (1964), Studies in Islamic Culture in the Indian Environment, Clarendon Press, Oxford. Armelin, I. (1980), Lettres indiennes contemporaines, Principales caractéristiques de la littérature des Mahanubhava. Notices bibliographiques, Librairie orientaliste Paul Geuthner, Paris. Bakker, H. (1990), Ramtek: An Ancient Centre of Vißñu Devotion in Maharashtra, in Id. (ed.), The History of Sacred Places in India as Reflected in Traditional Literature (Panels of the VIIth World Sanskrit Conference), E.J. Brill, Leiden, vol. 3, pp. 62-85. Bhandarkar, R.G. (1957), Early History of the Dekkan down to the Mahomedan Conquest, Reproduced from the Bombay Gazetter, Susil Gupta, Calcutta. Crooke, W. (1909), Berar, in Hastings, J. (ed.), Encyclopaedia of Religion and Ethics, T. & T. Clark, Edinburgh, vol. 2, p. 504. Denton, L.T. (1991), Varieties of Hindu Female Asceticism, in Leslie, J. (ed.), Roles and Rituals for Hindu Women, Pinter, London, pp. 211-231. ˘here, R.C. (1967), Musalmån Marå™hœ Sa∫takavœ, J∞ånråj Prakå†an, Puñe. ______ (1977), Cakrapåñi, Vi†vakarmå Såhityålaya, Puñe. Edgerton, F. (1964), The Bhagavad Gœtå, Harper Torchbook, New York. Enthoven, R.E. (1922), The Tribes and Castes of Bombay, Government Central Press, Bombay, vol. 2, pp. 427-433.

512

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Farquhar, J.N. (1984) [1920], An Outline of the Religious Literature of India, Motilal Banarsidass, Delhi. Feldhaus, A. (1978), The Mahånubhåvas and Scripture, in «Journal of Dharma», 3, pp. 295-308. ______ (1982), God and Madman: Guñ∂am Råu¬, in «School of Oriental and African Studies», 45, pp. 74-83. ______ (1983), The Religious System of the Mahånubhåva Sect: The Mahånubhåva Sü™ra-på™ha, Edited and Translated with an Introduction by Anne Feldhaus, Manohar, New Delhi. ______ (1984), The Deeds of God in R¢ddhipur. Translated from the Marå™hœ and Annotated by Anne Feldhaus. With Introductory Essays by Anne Feldhaus and Eleanor Zelliot, Oxford University Press, New York-Oxford. ______ (1987), The Religious Significance of R¢ddhipur, in Israel, M., Wagle, N.K. (eds.), Religion and Society in Maharashtra, Centre for South Asian Studies, University of Toronto, pp. 68-91. ______ (1988), The Orthodoxy of the Mahanubhavs, in Zelliot, E., Berntsen, M. (eds.), The Experience of Hinduism: Essays on Religion in Maharashtra, State University of New York Press, Albany, pp. 264-279. ______ (1991), Pai™hañ and the Någas, in Eck, D.L., Mallison, F. (eds.), Devotion Divine: Bhakti Traditions from the Regions of India. Studies in Honour of Charlotte Vaudeville, Egbert Forsten, Groningen, École Française d’Extrême-Orient, Paris, pp. 91-111. ______, Tulpule, S.G. (1992), In the Absence of God: the Early Years of an Indian Sect. A Translation of Smr¢tishtha¬ with an Introduction, University of Hawaii Press, Honolulu. Feuerstein, G. (1992), Holy Madness: The Shock Tactics and Radical Teachings of Crazy-Wise Adepts, Holy Fools, and Rascal Gurus, Arkana Books, New York. Ganguly, D.C. (19662), The Yådavas of Devagiri, in Majumdar, R.C. (ed.), The Struggle for Empire. The History and Culture of the Indian People, Bharatiya Vidya Bhavan, Bombay, vol. 5, pp. 185-197. Gonda, J. (1963), Die Religionen Indiens, vol. 2, Der jngere Hinduismus, Kohlhammer, Stuttgart. Granoff, Ph. (2001), From Detachment to Engagement: The Construction of the Holy Man in Medieval ‡vetåmbara Jain Literature, in Mallison, F. (ed.), Constructions hagiographiques dans le monde indien. Entre mythe et histoire, Bibliothèque de l’École des Hautes Études- Sciences Historiques et Philologiques, Paris, Tome 338, pp. 97-121. Jaini, P.S. (1993), Jaina Puråñas: A Puråñic Counter Tradition, in Doniger, W. (ed.), Puråña Perennis: Reciprocity and Transformation in Hindu and Jaina Texts, State University of New York Press, Albany, pp. 207-249.

V. Violating Tradition and its Boundaries

513

______ (2000a), Jina R¢ßabha as an Avatåra of Vißñu, in Id. (ed.), Collected Papers on Jaina Studies, Motilal Banarsidass, Delhi, pp. 325-349. ______ (2000b), Is There a Popular Jainism? in Id. (ed.), Collected Papers on Jaina Studies, Motilal Banarsidass, Delhi, pp. 267-279. Joshi, P.M. (1976), Glimpses of History in Mahånubhåva Texts, in «Indica» (Bombay), 13, n. 1-2, March-September, pp. 67-74. Karve, I. (1968), Maharashtra: Land and Its People, Maharashtra State Gazetteer, Directorate of Government Printing, Stationery and Publications, Bombay. Kinsley, D. (1974), ‘Through the Looking Glass’: Divine Madness in the Hindu Religious Tradition, in «History of Religions», vol. 13, n. 4, pp. 270-305. Kolte, V.B. (1950), Cakradhar åñi J∞åndev, Bombay. ______ (1952), ‡rœcakradhar Caritra, Aruñ Prakå†an, Malkåpür. ______ (1962), Mahånubhåv Sa∫†odhan 1, Aruñ Prakå†an, Malkåpür. ______ (19754), Mahånubhåv Tattvaj∞ån, Aruñ Prakå†an, Malkåpür. ______ (19822), Mhåï∫bha™ Sa∫kalit ‡rœcakradhar Lœ¬åcaritra, Mahåråß™ra Råjya Såhitya-Sa∫skr¢ti Mañ∂a¬, Bombay. Kosambi, D.D. (1962), Myth and Reality. Studies in the Formation of Indian Culture, Popular Prakashan, Bombay. Kshirsagar, A., Pacquement, J. (1999), Parlons Marathi. Langue, histoire et vie quotidienne du pays marathe, L’Harmattan, Paris. Kulkarnee, N.H. (1989), Medieval Maharashtra and Muslim Saint-Poets, in Bhattacharyya, N.N. (ed.), Medieval Bhakti Movements in India. ‡rœ Caitanya Quincentenary Commemoration Volume, Munshiram Manoharlal, New Delhi, pp. 198-231. McDaniel, J. (1989), The Madness of the Saints: Ecstatic Religion in Bengal, University of Chicago Press, Chicago. Munshi, K.M. (1957), The Struggle for Empire. The History and Culture of the Indian People, Bharatiya Vidya Bhavan, Bombay, vol. 5. Pacquement, J. (2000), Linguistique historique et histoire de la langue 80 ans après ‘La formation de la langue marathe’ de Jules Bloch, in «Bulletin de l’École Française d’Extrême-Orient», 87, pp. 741-763. Padoux, A. (ed.) (1990), L’image divine. Culte et méditation dans l’Hindouisme, Centre National de la Recherche Scientifique, Paris. Piantelli, M. (1974), ‡aõkara e la rinascita del bråhmañesimo, Esperienze, Fossano. Raeside, I.M.P. (1970), The Mahånubhåva saka¬a lipœ, in «Bulletin of the School of Oriental and African Studies», 33. ______ (1976), The Mahånubhåvas, in «Bulletin of the School of Oriental and African Studies», 39. ______ (1989), Gadyaråja: A Fourteenth Century Marå™hœ Version of the Kr¢ßña Legend. Translated from the Marå™hœ with Annotations, Popular Prakashan and School of Oriental and African Studies, Bombay-London.

514

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Ranade, R.D. (1982) [1933], Mysticism in Maharashtra, Motilal Banarsidass, Delhi. Rigopoulos, A. (1993), The Life and Teachings of Sai Baba of Shirdi, State University of New York Press, Albany. ______ (1998), Dattåtreya: The Immortal Guru, Yogin, and Avatåra. A Study of the Transformative and Inclusive Character of a Multi-Faceted Hindu Deity, State University of New York Press, Albany. ______ (forthcoming), The Mahånubhåvs, Firenze University Press, Florence. Russell, R.V. assisted by Rai Bahadur Hœra Lål (1916), The Tribes and Castes of the Central Provinces of India, Macmillan and Co., London, part 2, vol. 4, pp. 176-183. Sardar, G.B. (1969), The Saint-Poets of Maharashtra, Orient Longmans, Bombay. Schimmel, A. (1993), The Mystery of Numbers, Oxford University Press, Oxford-New York. Siauve, S. (1971), Les hiérarchies spirituelles selon l’Anuvyåkhyåna de Madhva, Institut Français d’Indologie, Pondichéry. Sontheimer, G.-D. (1982), God, dharma and Society in the Yådava Kingdom of Devagirœ according to the Lœ¬åcaritra of Cakradhar, in Id., Aithal, P.K. (eds.), Indology and Law. Studies in Honour of Professor J. Duncan M. Derrett, Franz Steiner Verlag, Wiesbaden, pp. 329-358. Swami Kripananda (1989), Jnaneshwar’s Gita: A Rendering of the Jnaneshwari, State University of New York Press, Albany. Tulpule, S.G. (ed.) (1963), Pråcœn Marå™hœ Korœv Lekh, Poona University Press, Poona. ______ (1979), Classical Marå™hœ Literature from the Beginning to A.D. 1818, vol. 9, fasc. 4, in Gonda, J. (ed.), A History of Indian Literature, Otto Harrassowitz, Wiesbaden. ______ (1996), Cakradhar and Women, in Feldhaus, A. (ed.), Images of Women in Maharashtrian Literature and Religion, State University of New York Press, Albany, pp. 201-211. Vaudeville, Ch. (1987), The Shaiva-Vaishnava Synthesis in Maharashtrian Santism, in Schomer, K., McLeod, W.H. (eds.), The Sants: Studies in a Devotional Tradition of India, Berkeley Religious Studies Series and Motilal Banarsidass, Berkele, pp. 215-228. Verma, O.P. (1970), The Yådavas and Their Times, Vidarbha Samshodhan Mandal, Nagpur. Waghorne, J.P., Cutler, N. (with V. Narayan) (1996) [1985], Gods of Flesh, Gods of Stone: The Embodiment of Divinity in India, Columbia University Press, New York. Warren, M. (1999), Unravelling the Enigma: Shirdi Sai Baba in the Light of Sufism, Sterling Paperbacks, New Delhi. Zelliot, E. (1987), Four Radical Saints in Maharashtra, in Israel, M., Wagle, N.K. (eds.), Religion and Society in Maharashtra, Centre for South Asian Studies, University of Toronto, pp. 131-144.

FABRIZIA BALDISSERA Traditions of Protest: the Development of Ritual Suicide from Religious Act to Political Statement 1

1. The Evolution and Implications of a Ritual Stance A number of Sanskrit works from medieval Kashmir, including the first known historical work, deal repeatedly with both single and mass suicides as a means of political action. In India the dharma texts generally prohibit suicide, yet these instances of suicides to obtain redress enjoyed great prestige, as they were seen to share the attributes of the holy death of ascetics or of heroes. Though the phenomenon is particularly frequent in the texts examined, it is by no means confined to Kashmir nor to medieval times, but in fact covers the whole of the subcontinent and a span of time that stretches from the Upanißads to Gandhi. The present study investigates the common cultural background that gave this particular stand popular appeal and political support, so that it could be used as an effective weapon against abuses of authority. A lexical enquiry on the evolving meaning of pråya and pråyopave†a compared to later terms such as dharña and tråga shows how this practice, that in the dharma texts seems confined to the claiming of a debt, in fact extended to cover a whole range of issues. The first instance of this sitting in fast, found in the Kaußitakyopanißad,2 has actually an ascetic fast against a village that had failed to give him food for alms. Attitudes towards all types of suicide were investigated in the dharma texts, in the epics and in some Buddhist and Jain cases. Although dharma prescriptions usually forbade suicide, the belief in 1 All translations are by Baldissera, except for those from the Råjataraõgiñœ –where the original Stein’s translations (1989) were maintained– and those from Månavadharma†åstra by Olivelle (2005), as well as for some of Hopkins (1900). 2 See infra, § 3.

516

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

reincarnation is seen to allow for a different perspective on life and death, compared for instance with that of the Christian faith. In India several types of suicide appear to stem from a shared awareness of a superior duty owed to one’s guiding principles. Thus the holy deaths in restraint of old ascetics and Brahmins, the suicides by fast in meditation of wounded or vanquished warriors, the self-burning of loyal widows as well as the self sacrifice of heroic devotees to goddesses, or again the self sacrifice of generous people to help others, a rather frequent instance in the Buddhist Jåtakas, are seen to belong to a similar sphere. This includes the particular sets of duties honoured by the different groups, such as the correct performance of rites (including those of bodily purifications) for old ascetics and Brahmins, the preservation of honour for all types of heroes and faithful wives, and the adherence to a generous ideal for those who gave their life for the sake of others. In all these instances there is a definite belief in the righteousness of such voluntary deaths, that are supposed to lead to a positive reincarnation. Personal and political suicide for redress was perceived to belong in this wider context. It enjoyed first the prestige of the saintly death of old ascetics and of Brahmins in the last stage of their lifecycle, and secondly the general admiration for the selfless courage of the hero, vœra, warrior, satœ or devotee. An outward similarity with the holy deaths can be seen in the ritual stance adopted by the faster. This involved a formal declaration of intent, sa∫kalpa, that preceded touching water and sitting on darbha grass in silence and meditation. Underlying the first motive for popular support is the interdiction to kill Brahmins and ascetics, or especially being responsible for their death, which gave them a strong position from which they could take political action against injustice. As for the second motive, the accent is put on the courage and generosity of the persons who are ready to pledge their own lives to achieve a higher end, be it honour or the redress of a wrong. The latter stand is generous in as much as the redressing in itself, even when it concerns personal matters, promotes the upholding of justice from the executive power, as this ultimately benefits the whole community. Both religious prestige and the appeal of heroic courage against unjust authority made for popular favour at all levels of society, and helped to turn what had started as single instances of solitary protests into the crowded demonstrations of medieval times and of modern civil disobedience campaigns. 1.1. Some Kashmiri accounts of fasting to death and the questions they raise Somadeva’s Kathåsaritsågara [Kåthå.], Kßemendra’s Narmamålå [Narm.] and De†opade†a [De†.], and Kalhaña’s Råjataraõgiñœ

V. Violating Tradition and its Boundaries

517

[Råj.], respectively a collection of tales, two satires and a historical work present several instances of ritual suicide. The three former works are coheval, while the latter comes approximatively 150 years later. The historical work in places seems to borrow terms, attitudes and ideas from the satirical ones, while the tales, in their composite origin, being a compilation culled from different sources composed in different periods, present links with other cultural milieus and times. The presentation of the material on pråya in the three different contexts however is rather homogeneous, especially in the three latter works, concerned with actual Kashmiri reality. The overall picture reveals a condition of frequent misgovernment, that often drew unfortunate groups of people to voluntarily kill themselves through despair. The frequency of cases of pråyopave†a in these three works should be seen against this particular background. Both the historical text and the works of fiction present cases of pråya undertaken either by single persons or by large groups, by men and women, the old and the young, by Brahmins, kßatriyas, merchants, common people, and even by a cat. Recourse to the particular attitude of pråya in these works is in most cases motivated by the wish to redress a wrong, though sometimes there are also occurrences of pråya undertaken on false pretences, or to influence important political decisions, or merely to obtain riches. In some of the law books pråyopave†a was one of the customary ways for a debtor to reclaim his deposit. The creditor would fast against the debtor, while the one so fasted upon was of course obliged to fast back, “so that the practice, as far as fasting went, resolved itself into a sort of stomach-duel”.3 While considering all the different cases presented in these texts a pattern of relations began to emerge. This resulted also in a set of questions that it seemed relevant to address to other works, some older and some more recent than those here considered, in order to better interpret the large mass of data collected.4 What was the origin of the practice? What tradition did its instances conform to? And what tradition, on the contrary, they eventually opposed? What did the law books say about it? Which social groups were most involved in it? And what did the public think about it? Was it deemed a prestigious practice, or was that controversial? And lastly, what did it seem comparable to? Coming to terms with even some of these questions proved rather challenging, first because the ancient texts that mentioned pråya often showed ambiguous views. Yet some of the more recent instances 3

Hopkins 1900: 158. To claim a complete survey of all the Sanskrit texts dealing with pråya would have exceeded the scope of this study, so here is presented only the selection which seems relevant. 4

518

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

of pråya (the practice was abolished officially around 1870)5 seemed to have changed the original meaning of the term, or to have fixed it as one directional. So, to start with, dictionaries and lexicons, then philology, literature and history were resorted to. The usual acceptance of pråyopave†a, (the most usual term found in Råj., where the highest number of occurrences is recorded), is “sitting in restraint [awaiting] death (by inanition)”. In most of the examples found in the Kashmiri texts, this practice, which in its formal aspects looked very much like a sacred ritual, was employed as the last means to oppose a wrong decision from a powerful agent, whom it would have been impossible to oppose in any other manner. Even so, its similarity with an extreme form of penance, involving fasting, as well as the lexically similar initial part of the word made it probable that it could be related to ‘expiation’ (pråya†citta). That would connect it with the world of Brahmanical lore, as well as showing also its closeness to the outlook of the renouncers, whose attitude resembles that of initiation (temporary or life-long), that involves a certain restraint, notably concerning food and sex. Prestige and influence are attached to this particular attitude, for “the highest value is renunciation, not purity”.6 In this perspective it is obvious that the prestige of the sannyåsin would entail popular support of his eventual demands. At the same time, the bold stance of the person who was prepared to die by fasting at the door of those responsible for the wrong revealed that it was also a case of extreme courage. This then could be seen as similar to the concept of honour of warriors, who were ready to adopt extreme measures in their avoidance of the worst dishonour or loss of face. This similarity could extend also, in some cases, to heroic self sacrifice for the good of others or, in a different context, for one’s own salvation. And the fact that starvation is a slow and painful voluntary death shows its similarity with the religious deaths of old ascetics, or to the satœ 7 of widows. The intriguing multiple facets of the concept required further investigation. In fact it is precisely the web of interconnections between different beliefs and the development of the different themes from an organic background, in which ascetic values are connected with worldly heroism and the preservation of honour –both for oneself as well as for one’s extended family, for both males and females, and eventually for everyone’s survival–, that seems worthy of investigation. These topics will emerge in the course of the discussion of the materials assembled. These include all the instances of suicide, prayopave†a and satœ found in Råj., all instances of pråya and collective 5

See Renou 1943-1945: 117, note 1. Heestermann 1985: 155. 7 Throughout this essay the term is used to indicate both the person who commits the act, and the act itself. 6

V. Violating Tradition and its Boundaries

519

suicides in the two satires of Kßemendra, as well as most examples of suicide, prayopave†a and satœ told in Kathå. Selected cases drawn from the Alexander historiography, from the epics and from different law books, as well as a couple of examples from the fifteenth and sixteenth century, Gandhi’s experience and some contemporary issues shall provide further illustration of the complex evolution of this ritual stance. The material is however elusive, because in most suicides or attempted suicides several different motifs tend to overlap. People who envisage to kill themselves due to sorrow or despair put the blame, or sometimes an explicit threat, on those they deem responsible for their situation, like Kausalyå does on Råma, or a matter of honour can turn into a battle of wills, like in the case of Kaikeyœ and Da†aratha.8 1.2 Lexical Aspects The term pråya comes from pra-i, a root meaning ‘to go’, ‘proceed’, ‘advance’, and further ‘to die’. In the Sa∫hitas pra-i and its derivatives mean only ‘to advance’, ‘to proceed’, eventually ‘to escape’. “The primary meaning of pråya seems to be ‘going forward’, or, figuratively, ‘continuity, the holding (of a certain attitude)’”.9 To Renou’s remark could be added also the term preta (‘the departed’), which immediately acquires the meaning of ‘the dead’. In the classical lexicons, in fact, as already observed by Renou (194345: 118), pråya is explained as ‘abstention from food’ (ana†ana), as well as ‘death’, due to a reinterpretation of the locution pråyam upavi†-. In order to have a complete picture of both the notion and its application, it is important to remember in fact that in the different texts pråya, as an absolutive of the ñamul type, has been usually joined with a verb meaning ‘sitting’. The usual verbal roots cited in the dictionaries are either upavi†-, ås-, upås-, upe-, åsthå-, samåsthå-, or kr¢- and karay- so that at times one of those verbs could stand for the whole expression, and particularly the verb upavi†, found throughout the Råj. In the Narm. and De†., on the contrary, pråya is always joined with the root sthå-. upaviß- means first ‘entering upon’, then ‘sitting (in restraint) [for death]’, where pråya from pra- joined with the root œ, ‘to go’, means ‘proceeding, going forth’, and ‘death’, as found already in the Bråhmañas. Renou recalls that the term is often glossed in the lexicons by “[…] observance of fast connected with the state of renouncer, (samnyåsavant or similar term)”.10 pra-i in its primary meaning of ‘moving forward’ is at the origin of the expression pråya†citta (‘expiation’), literally “[…] the intention of pushing forSee infra, § 2.4. See Renou 1943-1945: 118. 10 Renou 1943-45: 117-118, 123, note 1. 8

9

520

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

ward, discarding (the fault by the appropriate magical action)”. 11 While discussing these terms, Renou mentions also the possibly related term upavas ‘to live at’, ‘spend the night at’, ‘spend the night in [alimentary] restraint’,12 or the upavasatha-uposatha of the Buddhist tradition.13 1.2.1. pråya and dharña/dharañå There is a particular relation between the terms and practices of pråya/pråyopave†a and the later dhårña: both have in common the notion of ‘sitting firmly’, ‘staying put’ in a place, usually in front of the door of the person who has caused or allowed the offence or the injustice. A similar notion is still present in many countries in the contemporary sit-ins staged during peaceful demonstrations to protest against some specific issue. In present times this is considered a non violent way of holding –literally– one’s ground by refusing to move. The first meaning of Sanskrit dhårañå, according to the Monier Williams, is “[…] the act of holding, bearing, wearing, supporting, maintaining; retaining, keeping back; collection or concentration of the mind (joined with the retention of breath)”. In the Kathå. it can be a fiery concentration, born of fire, given to human devotees by a goddess, so that they may cast off their mortal body. Kathå., 1.5.140141a recounts what happened to the scholar Vararuci, who had gone to the hermitage of Badarœ: With great devotion, desirous of freeing himself of his mortal condition, he resorted to the refuge of goddess ‡arañyå. The goddess, after revealing to him her own form, gave him spontaneously the concentration [technique] (dharañå) arising from fire for freeing oneself of one’s body. And after having burnt his body through that concentration, Vararuci reached his own heavenly abode.14

Here this (mentally and physically) self-induced combustion provides a perfect ascetic death in restraint and meditation. The following is the definition of dharña found in the Shorter Oxford Dictionary, where the entrance states the year 1793 as the first English occurrence of the term: Dharna, dhurna (1793) [Hindi dharnå ‘placing’, ‘act of sitting in restraint’, from Skr. dhr¢ ‘to place’]. A mode of compelling payment or 11

Renou 1943-45: 118, note 1. See Aitareyabråhmaña, 2.15.4; 2.36.1. Also ‡atapathabråhmaña, 1.1.1.7. 13 See Renou 1943-1945: 124-130. 14 Kathå., 1.5.140-141a: atha sa nibi∂abhaktyå tatra devœμ †arañyåμ †arañam upagato’sau martyabhåva∫ mumukßuΔ Ù praka™itanijamürtiΔ såpi tasmai †a†aμsa svayam analasamutthå∫ dhårañå∫ dehamuktyai ÙÙ140ÙÙ dagdhvå †arœ^ram atha dhårañayå tayå taddivyaμ gati∫ vararucir yayauÙ 12

V. Violating Tradition and its Boundaries

521

compliance with a demand, by sitting at the debtor’s door, and there remaining without tasting food till the demand shall be complied with; this action is called ‘sitting (in) dharnå’.

This definition, according to Renou and especially Hopkins (see infra) would be more appropriate to describe pråyopave†a, as fasting is not necessarily one of the characteristics of dharña. Explaining the lexical/semantic connection between pråyasthå/prayopavi†- and modern dharña, dharña bai™hnå, Renou (1943-1945: 118) declares that the latter is merely the literal translation of pråyam upavi† ‘to sit down while keeping (a certain attitude)’, or ‘while besieging’, as it also maintains the use of an absolutive of the ñamul type, just like pråya. Hopkins is more precise, in that he takes dharña as ‘obstruction’: “The practice of dharña today includes not only ‘door-sitting’ but also any form of obstruction, for example, obstructing a water course”.15 Fasting is not, therefore, a necessary concomitant of dharña though it is of ‘door-sitting’ (dvåropave†anam). Hopkins provides a further clarification: “dharña (= dhåraña), literally a holding, capio, or in English slang a ‘hold up’, which is restricted to a priest, and as already stated may be any form of obstruction, like obstructing the door or obstructing a water course”.16 1.2.2. Further developments: from dharña to tråga and takåzå There is another modern form of suicide akin to dharña known as tråga, which bears a further connotation of self punishment: “That is suicide simply as a self inflicted punishment for disgrace or failure to carry out what has been solemnly agreed to”.17 As an example, Hopkins refers to Hopkins (1894: 480) where he relates an actual fact that happened in 1894: “The man who had made himself responsible for a payment, on finding that the debtor would not pay, to expiate the disgrace slew his own mother in the presence of the defaulter, who in turn as his only expiation slew himself”.18 Renou interprets tråga as a modern extreme form of dharña: “The creditor, instead of fasting himself, can send his son, or one of his people, to reinforce the pressure by threatening to torture his own son, wife or cattle”.19 This is similar to the old explanation given 15 Hopkins 1900: 147, note 1. Looking for ancient examples of such magical obstruction, of dharñå as ‘obstructing [the water]’, and especially when it is done without good cause one could be reminded of the ancient fight found in Vedic literature between Indra and Vr¢tra, who was obstructing the waters. For there is of course memory of fierce tapas done for evil ends. To find instances of false pretences in pråya see, infra, the appointment of special officers in Råj., installed to watch over cases of prayopave†ana, as well as episodes in the Råj. and in Narm. 16 Hopkins 1900: 157. 17 Ibidem. 18 Ibidem. 19 Renou 1943-1945: 119.

522

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

by the Br¢haspatismr¢ti (circa 500 CE) of the term åcarita found in Månavadharma†åstra, 8.49.20 Br¢haspatismr¢ti, 11.58 says that åcarita is a mode of exacting payment which consists either in seizing the debtor’s wife, son, and cattle, or in ‘performing the door-sitting’ (kr¢två dvåropave†anam).21 This is glossed also as ‘siege of the house’ (gr¢hasa∫rodha), which Renou explains as: “consisting in distraining the son, the wife or the cattle of the debtor and in sitting at his door”.22 Here then it would seem that the creditor simply blocks, besieges the family and cattle of the debtor, without actively harming them (but the cattle, for instance, cannot go to the pasture). On this substitution of the creditor by one of his people, see also a related South Indian practice, called takåzå. This appears to be a looser form of dharña, which […] permits the creditor to institute by proxy a regular siege of the debtor’s house. Here the creditor, instead of acting for himself, hires a band of ruffians to obstruct, besiege, annoy, and threaten the life of the debtor. Some premonition of this substitution is found in the interpretation by a medieval commentary of Nårada’s law –which, i. 122, on this subject coincides with Manu’s law– whereby a son or slave may act for the creditor. The Southern takåzå is of course without any religious significance, for the debtor is simply bulldozed into paying.23

It is interesting to note how different practices apparently stemmed from each other, borrowing some elements while changing others. In the last type of practice, takåzå, not only the personal attendance but the moral stance of the creditor are completely absent. 2. Different attitudes towards suicide: the dharma texts The fact that pråyopave†a, when it does not obtain its end, i. e. the redress of the wrong, amounts to suicide, has also brought to light one of the many ambiguities found in the law texts. In Månavadharma†åstra pråya as such is not mentioned explicitly (though his commentator Medhatithi thought that the term åcarita in 8.49 referred to it), but the usage is known already in an earlier text, the Apastambadharmasütra, where a list of people with whom it is forbidden to eat starts by: the drunk, the madman, he who has served time in jail, the creditor who is sitting against(the debtor) as well as (the debtor) against whom 20 The Månavadharma†åstra states åcarita as ‘the customary practice’, when he prescribes modes of reclaiming debts. 21 Quoted in Hopkins 1900: 146. 22 Renou 1943-1945: 122. 23 Hopkins 1900: 157-158.

V. Violating Tradition and its Boundaries

523

(the creditor) is sitting, for as long as (this practice lasts) (matto unmatto baddhoñikaΔ pratyupaviß™o ya† ca pratyupave†ayate tåvantaμ kålam).24

Suicide is forbidden in the dharma†åstras,25 but there were exceptions, not merely in the practice, but also contemplated in some of the actual texts. The suicide by fire or inanition of old ascetics, for instance, is well known from ancient times, and has even been witnessed and recorded by foreign sources, starting with the Greek accounts of the Alexander campaign. For the self-burning of Calanus and Zarmarus or Zarmanochegas, as well as for the first instances of satœ and possibly jauhar, see Thakur (1963: xii [especially on jauhar]), Michaels (1990: 21-34), where the oldest inscription concerning widow burning is found to be from Nepal in ‡aka 386, i.e. 464 CE (Michaels 1990: 22-23), Karttunen (1997: 64-67, and note 281), Garzilli (1997: 205-243; 339365), Arrien (2002: 21). What is interesting in these first foreign accounts is the extraordinary impression produced among the Greeks and Romans by the self-burning to death of singular gymnosophistes, probably because they were so calm and collected in the face of such a slow and painful death. The widows burning was seen in Greece and Rome as a noble example of faithfulness, as well as a preemptive means to prevent disaffected wives from poisoning their husbands. The dharma†åstras present contradictory statements, as even the Månavadharma†åstra –previously quoted in 5.89 as condemning suicide– allows it in the contexts of practices to be followed by the vanaprasthas: Or he may set out in a north-easterly direction and, subsisting on water and air, walk straight on steadfastly until his body drops dead. When a Brahmin has discarded his body through any one of these means employed by the great seers, freed from sorrow and fear, he will be exalted in the world of Brahman.26 (Månavadharma†åstra, 6.31-32)

A similar statement is found also in Atri, as quoted by Kane: Atri did not condemn suicide in certain cases. He states (verses 218219):27 ‘if one be old (beyond 70), if one cannot observe the rules of 24 Åpastambadharmasütra, 1.6.19.1. According to Renou (1943-1945: 122), in place of baddhoñikaΔ one should read baddho’ñikaΔ, a prakritic form for r¢ñikaΔ. 25 Notably in Månavadharma†åstra, 5.89, that enjoins that no libations of water should be offered to a series of dead people, including those who have committed suicide. Other authors who forbid suicide are Yaj∞avalkyasmr¢ti, 3.6, Apastambadharmasütra, 1.28.17, and Gautamadharmasütra, 14.12, who mentions as suicides those who kill themselves by starvation, pråya, weapons, fire, poison, hanging or jumping (from precipices). 26 Transl. by Olivelle 2005. In this regard Bhler (1969: 204, note 32) explains: “‘By one of those modes’, i.e. ‘drowning oneself in a river, precipitating oneself from a mount, burning oneself or starving oneself to death’ (Medatithi ad loc.) […]”. 27 Atri, 218-219: vr¢ddaΔ †aucasmr¢ter luptaΔ pratyåkhyåtabhißakkr¢iyaΔ Ù åtmåna∫ ghåtayedhyastu bhr¢gvagnyana†anåmbubhiΔ ÙÙ tasya triråtram å†auca∫ dvitœye tv asthisaμcayamÙtr¢tœye tüdaka∫ kr¢två caturthe †råddham åcaret ÙÙ (quoted by Medhatithi on Månavadharma†åstra, 5.89).

524

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

bodily purification (owing to extreme weakness), if one is so ill that all medical help is discarded, and if one in these circumstances kills himself by throwing himself from a precipice or into fire or water or by fasting,28 mourning should be observed for him for three days and †råddha should be performed for him’.29

The debate continued, and a general tendency of the commentators to become more lenient with the passing of time was perceived. 2.1. Similarity of pråyopave†a with the attitude of the renouncer: some fasting pråya†cittas in Månavadharma†åstra, and the danger posed by the wrong penitent In assessing ascetic practices, one should consider the original importance attributed to fasting austerities. They once were expiations, pråya†citta, for mistakes and omissions in ritual. Thus in Månavadharma†åstra: [11.204] For neglecting the daily rites prescribed by the Veda and for breaking the vow of a bath-graduate, the penance is fasting; […] [11.212] A twice-born practicing the Pråjåpatya penance should eat in the morning for three days and in the evening for three days, eat what is received unasked for three days, and abstain from food during the final three days. [11.213] Subsisting on cows’s urine, cow dung, milk, curd, ghee, and water boiled with Kußa grass, and fasting during one day –tradition calls this the Såntapana penance. [11.214] A twice-born practicing the Atikr¢cchra (very arduous) penance should eat as before one mouthful a day during the three three-day periods and fast during the final three days; […] [11.216] When a man, controlled and vigilant, abstains from food for twelve days, it is called the Paråka penance, which removes all sins.30

This last type of expiation involves a very long fast, and is said to remove all faults. By the force of such injunctions, the stance of the pråyopave†in then would seem to be unassailable, even from within. In fact, whatever his ends and intentions, by the painful undertaking of a self imposed penance the person engaged in this fast is already clearing oneself of all evil.31 This is why the practice of extreme tapas could become a real weapon in the wrong hands and be used even against dharma, or become a battle of dharma against dharma. This is the case of the many Here expressed with anå†ana-. Kane 1946, 3: 958-959. In the following paragraph Kane quotes other authorities who hold similar views, and refers to Kane (1941, 2: 926-927) for further details. 30 Transl. by Olivelle 2005. 31 This is actually a controversial issue, because already in Vaikhånasagr¢hyasütra, 5.11, it is stated that: “He who undertakes pråya without reason shall not be incinerated”. And Råj., 6.14, mentions that there were special officers appointed to investigate and report to the king cases of pråya, and presumably also to assess their adherence to truth. 28

29

V. Violating Tradition and its Boundaries

525

demons who, helped by their superhuman powers, did tapas to such an extent that they were able to obtain –usually from Brahmå– boons that proved very dangerous to the dharmic functioning of the world. The Narm. has a similar instance of a demon doing tapas to achieve evil ends. A few verses give this imaginary account of the previous birth of the formidable government officer: the ∂ivira is said to have been originally the home accountant of the demons; after the demons’ defeat by the gods,32 he undertook a severe fast lasting one thousand years, during which he subsisted only on handfuls of his own urine,33 and thus obtained to return to earth in order to harass the gods –through spoliating the Brahmins. In earlier times, in the upanißadic milieu, renunciation had come to equal prestige, because the renouncers were capable of restraining their senses to the utmost of their endurance. This involves also the abandonment of material things in order to acquire a higher status. It would seem that this attitude bears some resemblance to the stance of the ancient South Indian kings, who weakened their armies and depleted their treasuries by granting extraordinary largesse, abandoning wealth in order to gain greater prestige. In this display of extreme liberality there is also a dramatic aspect, in the turning of politics into a public spectacle, as well as into a publicized exhibition of piety. 2.2. Suicide in the epics: similarity of the wounded hero’s death with the religious death of the renouncer In the epics there are different views on suicide, usually expressed in the course of different episodes. The most frequent occurrences have to do with severely wounded heroes. Their attitude is compared to that of the old ascetics, for whose pråya is found provision in Månavadharma†åstra, Atri, etc. (see supra). In fact in some instances in Mahåbhårata these heroes are said to be ‘entering yoga’ as well as ‘entering pråya’ (see, for instance, Mahåbhårata, 7.192.46; 7.198.2931). Often wounded or vanquished heroes resort to voluntary death to gain heaven, like the kßatriya Bhüri†ravas in Mahåbhårata, 7.198.29-31: Bhüri†ravas sits silent, muniΔ, on the field of battle, ‘having entered upon pråya’, pråyagataΔ, devoting himself to death. In this state he ‘withdraws his breath’ and meditates upon holy texts, fixing his eye upon the sun, desirous of going to the Brahma-world. In other words he acts just like a Yogin, and the terms used of his act are indifferently yogayuktaΔ (abhavan muniΔ) and pråyopagataΔ or pråyam upåvi†at (above and Mahåbhårata, 143.33-35).34 32 Here Narm., 1.16, offers also a fictional etimology of the term divira, ‘officer, bureaucrat’ (originally a Persian word), saying that it comes from divi roditam, ‘crying in the sky’, for the demons, his masters, who had been vanquished by the gods. 33 Narm., 1.10b: svamütraculakåhåraΔ. See also Baldissera 2005. 34 Hopkins 1900: 151.

526

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

This withdrawing of breath was mentioned as part of the ritual of dharña in the Oxford Dictionary, and is found several times in the Kashmiri medieval literature describing the attitude of the pråyopave†ins. What is remarkable is that even some young kßatriyas undertake similar staunch ascetic practices. In the Råmåyaña Bharata first tries to persuade his brother Råma to come back and reign by sitting in pråya in front of his hut.35 After Råma tells him that such a behaviour is only proper for a Brahmin, he desists but makes a formal declaration (saμkalpa) and vows that he shall adopt a particular mode of life, which happens to be the stance of the renouncer. Bharata in fact declares that for fourteen years he shall live outside the town, wearing tree bark and matted hair, surviving on fruits and roots, while Råma’s sandals shall sit on his throne. If at the end of the fourteenth year he should not see Råma, on the following day he shall enter the fire.36 An interesting corollary of such a view, that sees the kßatriya epic hero as an ascetic or a Brahmin, is the observation of Biardeau (1976: 132), that vedic sacrifice in the epics, and especially in the Mahåbhårata, is transformed into the sacrifice of war, yuddhayaj∞a, where kßatriyas participate in the triple role of patrons, sacrificers and eventual sacrificial victims. Another epic passage, speaking of a warrior, says: “A pråyopave†in, O king, always obtains bliss” (Mahåbhårata, 13.7.16 [pråyopave†ino råjan sarvatra sukha∫ ucyate]), and then a telling comparison follows: “like an ascetic who lives on grass”. Such passages, though, are contradicted by others which hold the opposite view, namely that suicides go to the nether world. 2.3. Suicide in the epics: heroes and the avoidance of dishonour In Råmåyaña even monkey heroes resort to pråya as a proper form of suicide when in absolute despair. At times monkey-warriors undertake it as an expiation for failing to accomplish something which had been promised, as in Råmåyaña, 4.53.12-13: “We must die now, for we have failed in our attempt, and hence to enter upon death (pråyopave†ana) is better”. Here the thing to avoid is loss of face in front of their ally Råma. Later also the monkey hero Aõgada, thinking about his future life as a prisoner of Sugrœva, exclaims that pråyopave†ana is better than being in jail (Råmåyaña, 4.55.11). So, sitting on darbha grass, Aõgada declares that he will give up his life in the last fast. All his warriors, monkeys like him, prepare to follow his example. Facing East and sitting on darbha grass whose tips are pointing South (the direction of Yama) they touch water and undertake the solemn fast sitting around him (Råmåyaña, 4.55.15 ff). 35 36

Råmåyaña, 2.111.14-17. See also infra, § 2.5, pråya of the claimant. Råmåyaña, 2.112.20-23.

V. Violating Tradition and its Boundaries

527

Despair connotes also the musings of Hanumån, who conceives of a sort of domino-like falling to voluntary death of Råma and of his family, entraining then the subsequent death of all the monkey population which would seek death either by poison, by hanging, by fire, by fast unto death or by weapons. Hanumån himself at this point thinks of committing suicide by various means, including fasting in the proper way, so that birds and jackals may eat his remains (Råmåyaña, 5.13.22-42). An oath involving pråya is pronounced both by Arjuna, when he vows to kill Jayadratha within night or undergo pråya (Mahåbhårata, 7.146.105 ff), and by Duryodhana after vowing to get rid of his rivals (Mahåbhårata, 3.7.5).37 Similar oaths, involving fasts to the death or until the attainment of one’s martial purpose, occur also in Råj. See for instance the oath pronounced by King Harßa’s general Kandarpa, who had been slighted by the king (Råj., 7.972-991): “Vexed by these reproaches, the latter took a vow that he would fast till he had conquered Råjapurœ, and then set out, though he had no supplies”.38 After six days of fasting on march, Kandarpa took the city. In another occasion Duryodhana announces that he intends to undertake pråya as he thinks he has been slighted (Mahåbhårata, 3.249.11, 20 ff): He then touches water, sits down upon sacrificial grass, and clothed in rags, and silent, collecting his thoughts, prepares to die of starvation; though his friends attempt to dissuade him by telling him that he is foolish and that ‘a suicide goes to hell’ (kasmåt pråyopave†ana åtmatyågœ hy adhoyåti).39

The first reason given by his friends is the ridicule he might incur, as the cause who prompted his pråya is insufficient: “After undertaking pråya, o king, you shall be a laughing stock for the [other] monarchs” (pråyopaviß™as tu nr¢pa råj∞åμ håsyo bhavißyasi).40 Two important points emerge from the epic passages: first, the possibility of ridicule, the worst fate for a warrior. Loss of face (or presumed loss of face) is the cause of Duryodhana’s grim decision, but this may entail an even worse loss of face. And for kßatriyas honour is the highest value, to be seen also in the beliefs that a heroic death in battle is the coveted goal (that leads to Heaven), and that for their women suicide is better than being taken by the enemies. This notion is very popular throughout the Far East as well, 37 Mahåbhårata, 3.7.5: punaΔ †oßa∫ gamißyåmi nirambur niravagrahaΔ Ù vißam udbandhanaμ caiva †astram agniprave†anaμ karißye ÙÙ. ‘I will dry myself up, without water’ amounts to declaring the start of a fast unto death. 38 Råj., 7.972: kr¢tapratij∞onåhåratayå råjapurœjaye Ù upålambhårditaΔ sotha niΔsa∫gryo’py avåcalat ÙÙ. 39 Mahåbhårata, 251.19; 252.2. Sanskrit verse quoted in Hopkins 1900: 152-153. 40 Mahåbhårata, 250.12.

528

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

although the manner of dying is different –but still very painful, requiring great self possession and restraint: compare for instance the many cases of suicide by sword for loss of face described in ancient Chinese literature, as well as both ancient and recent seppuku and harakiri in Japan.41 For a Japanese warrior the worst fate was to be made prisoner or to be killed by the enemy. It was considered much more honourable to put a voluntary end to one’s life, sometimes helped by a friend who would take care to shorten the agony of the disembowelled warrior by cutting his head after the ritual stabbing.42 The second point, concerning the dharmic attitude towards suicide, is also interesting, in that it shows that in the same epic there were controversial opinions, according to the different possible interpretations of dharma. Here two contrasting views are expressed in the Mahåbhårata: in 13.7.16 the action of the pråyopave†in is exalted, as he is said to obtain bliss, while in 251.19; 252.2 his lot is to go to hell. 2.4. Suicide in the epics: heroines and the avoidance of dishonour Women of kßatriya descent, while enjoying a relative freedom in choosing their spouses (see infra, § 3.2), had to adhere to a very strict code of honour. They were not generally required to lead their troupes in battle, like the glorious Rani of Jhansi,43 but were expected to burn themselves if they risked to fall in the hands of their enemies, or at the death of their lords. The practice of satœ is already known in the epics, for instance in Råmåyaña, 2.45.36-37, where Sœtå says that should she remain without Råma, she would throw herself into the Godåvarœ, or hang herself, or jump from a cliff, or drink poison, or enter fire, rather than touch another man. There are several instances of women ascending the pyre also in Mahåbhårata, for instance in 12.148.9-10, and there is also a rather explicit general statement in Mahåbhårata, 1.74.46): “A good woman (sådhvœ), follows after her husband who has died before her”. But the Mahåbhårata, in those that are considered its older parts, “fully recognizes the survival of widows, cases of suttee being mentioned only in the later added books”.44 In the Månavadharma†åstra, for instance, there is no trace of such a practice. In the Råmåyaña on the other hand there are passages where women threaten suicide to preserve their honour. Queen Kaikeyi in 41 Very popular among warriors especially from the middle of the twelfth/thirteenth century, with the rising of the military class. 42 Compare also the two instances of wives of heros in Råj., who killed themselves when their husbands were imprisoned: see infra, § 5.1, and especially the suggestions of Harßa’s ministers and his reply, see infra, § 4.2. 43 In 1858 the Rånœ of Jhansi died heroically at the head of her army in the attempt to defend Gwalior from British troops. 44 Hopkins 1900: 149.

V. Violating Tradition and its Boundaries

529

Råmåyaña, 3.47.8-9 threatens the king with the following words: “If Råma is consecrated (against my will and thy promise) I shall not eat, nor sleep, nor drink, from this day on forever, and this will be the end of my life”. Hopkins observes: In R. ii.11.21 (compare 9.59 ff.) this is represented as being a threat of death because of the disgrace attaching to the queen if her husband breaks his promise to her: ‘despised by thee, I will die today’; ib. 12.47, ‘I will drink poison in thy presence’.45

It is a different case when a similar threat is uttered by Queen Kausalyå when she tells Råma: “If thou abandonest me, my son, I will sit to death, being unable to live, and then thou shalt go to hell world renowned”.46 Here the desire to end her life would be occasioned by affection and despair, though, and not by the possibility of wounded pride, or the avoidance of dishonour. 2.5. Suicide in the epics: pråya of the claimant Only in the last two epic instances of pråya did the fasters claim anything of others: this is the case, later publicly undertaken for upholding political views, when the fast is done specifically against someone perceived to have done some wrong to the faster. Here is when a definite threat (and eventual blackmail) is posed. Another of the first instances of such a pråya in the epics occurs when young Bharata, ashamed at his mother’s doings, prostrates himself in front of Råma, beseeching him to go back and rule. Bharata orders his süta to lay ku†a grass on the floor, saying: “I will besiege (beseech) the prince until he grants me his favour. Without food, not averting my eyes, like a priest that has been robbed of his possessions, I will lie before his hut until he (yields or) returns home”.47 To which Råma replies: “Why wilt thou besiege me? For only a Brahman (priest) has the right to obstruct men, and the observance in regard to besieging is not for annointed (kings)”.48 Here Hopkins (1900: 157) observes that the verb rudh-, ‘to obstruct’, is employed in technical application, as the only case in epics. And, one could note, in the same meaning he attributes to later dharña. This besieging and obstruction, considered proper only for Brahmins, is resorted to also by women, already in the epics (and the 45

Hopkins 1900: 155. Råmåyaña, 2.21.27-28. Transl. by Hopkins 1900: 156. 47 Råmåyaña, 2.111.14-17. Transl. by Hopkins 1900: 156-157. 48 Råmåyaña, 2.111.14-17: årya∫ pratyupavekßyåmi yåvan me samprasœdati Ù niråhåro niråloko dhanahœno yathå dvijaΔ Ù †aye puraståc chålåyåμ yåvan må∫ pratiyåsyati Ù ki∫ måm […] pratyupavekßyase Ù bråhmaño hy ekapår†vena narån roddhum ihå’rhati Ù na tu mürdhåbhißiktånåμ vidhiΔ pratyupave†ane ÙÙ (Transl. by Hopkins 1900). 46

530

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

example shown here is of a kßatriya woman).49 It may be that the norm that regulates its use is ‘the customary behaviour’, the åcarita mentioned by Månavadharma†åstra, 8.49. In Mahåbhårata, 1.2.304, it is said of Draupadœ: “what time she, resolved upon not eating, besieged her husbands” (kr¢tåna†anasaμkalpå yatra bhartr¢n upåvi†at).50 (The queen desired from her husbands a reprisal against the enemies). Here there is no mere threat, Draupadœ actually sits and fasts. 3. A question of outlook: was prestige attached to pråya? And why? If one tries to look into the origin of the popular prestige enjoyed by pråya, a form of suicide that was often deemed to abide within dharma, one quickly perceives that it rests with the respect given by people to extreme personal courage and to renouncing ascetics, as well as with the self-attributed superiority of the Brahmins. The outward look of the practice, in fact, from the initial formal declaration, sa∫kalpa, to the touching of water, to the sacred grass on which they sit, to the attitude of mental restraint, symbolized by the silence and breath control of the fasters, recall such ascetic tapas as was performed also by non renouncing Brahmins during dœkßå period, or in the last stage of their lives. The socio-juridical changes that around the end of the Bråhmaña period shifted from a rather flexible ensemble of group judgments to a situation that saw the Brahmins setting themselves up as the sole agents responsible for the interpretation of dharma has been aptly summarized by Falk (2001: 131-136). At the same time, with the fixation of the varña classification, the Brahmins promoted the notion of regality with the king as the executive guardian of dharma, but under their own (the Brahmins’) guidance. The changes outlined above also led to a reworked definition of dharma: it is now a subject of the Brahmins learned in the Vedas, who by their influence demand four forms of respect: reverence, gifts, the respect of not being plundered and the respect of not being put to death.51 (Falk 2001: 134)

The Brahmins needed a king in order to impose their law on others, but in the case of faults committed by Brahmins the latter revert49 Most likely here the underlying notion is that pråya as a mode of compelling another to act in one’s favour is a behaviour considered right for people who should not resort to strength or weapons, such as Brahmins, ascetics, women and children –categories of people whom it is forbidden (and/or dishonourable) to kill. As the undertaking of pråya puts the claimant in the humble, beseeching position of one who belongs to these categories, this is why Råma thinks it a wrong or, rather, a demeaning ritual for a warrior. 50 Transl. by Hopkins 1900: 154. 51 See ‡atapathabråhmaña, 11.5.71.1: lokaΔ pacyamåna† caturbhir dharmair bråhmaña∫ bhunakty arcayå ca dånena cåjyeyatayå cåvadhyatayå ca (quoted in Falk 2001: 134, note 6).

V. Violating Tradition and its Boundaries

531

ed to the old habit of inflicting to their group purificatory rites, even to the extent of ordering the culprit to kill himself. Expiatory suicide was considered as a type of punishment, but it was supposed to be undertaken voluntarily.52 Falk53 identifies two different sets of provisions even in the sole text of Åpastambadharmasütra, where the punishment for the killer of a Brahmin (presumably himself a Brahmin) is either mere expulsion from the community and reintegration after 12 years of penance in the woods (Åpastambadharmasütra, 1.9.24, 6-11), expulsion without reintegration (Åpastambadharmasütra, 1.10, 21) or, “following ÅpDhS 1.9,11 […] he has to go to a battlefield, and there he has to place himself in between the lines and try to be killed by a passing warrior. Alternatively, he can also burn himself on a pyre”.54 From the above it is evident that the rule tends to emphasize the voluntary character of the penance. Even in the case of a thief, the culprit has to go himself to the king and ask to be punished in order to go to heaven after his death. This way, a double end is achieved: no evil attaches itself to the punisher, and the absolute power of the king is somehow mitigated, because the punishment is perceived as a voluntary expiation not decreed by the sovereign. It is interesting to see that the same type of penance, already found in Mahåbhårata for the culprit of a different crime (see infra), in the late official law treatise of Nepal, the Muluki Ain (1854 CE), is used for restoring to their caste people who had lost it through some crime or break of dharma. Axel Michaels kindly gave me an excerpt from his yet unpublished study of the Nepali code.55 His data come from the Muluki Ain (from now on MA) of 1854 and 1888 and concern people who want to restore their loss of caste (the patita):56 3.2.4 Persons travelling to a foreign country (especially soldiers): MA 1854/89/70,71; MA 1888/5.32/7,8. These cases mostly concerned traders and soldiers (but not pilgrims), for whom it was at times difficult to stay separate and cook for themselves. They had to get rehabilitation after their return, which was granted when they reported independently. However, these regulations were added to the MA 1854 of V.S. 1910 on V.S. 1922 Bai-Åkha badi 1, i.e. 12 years after the promulgation of the first Ain. In the MA 1888, the rules for soldiers are again clarified since it is their duty (dharma) to fight in war and to join (foreign) armies, e.g. the English Company or British Government (MA 1854/89/70). Soldiers did not easily loose their caste status but only if sufficient evidence for an illegitimate commensal behaviour was given. The burden of proof was 52

See also, for a similar attitude in Nepal, Lingat 1967: 61; Fezas 2000: 28-30. Falk 2001: 135-136. 54 Falk 2001: 135-136. 55 See Michaels forthcoming. 56 It was important to expiate loss of caste even at the cost of one’s life, because the fall would affect also the members of one’s [extended] family, and in particular one’s direct descendants. Such expiation on the other hand would restore the initial status to the whole family. 53

532

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

somehow reversed, but the soldiers seemed to have been morally requested to get rehabilitation, a thing that even J. Bahådur Råñå received after his return from Europe. Such regulations indicate once again the privileged treatment of members belonging to the kßatriya varña. The same holds true for a kind of probation called death absolution (dehånta pråya†cit[ta]), which was granted to men who were willing to go to war for the king and to fight unto death, a kind of ritual suicide, in which courage and bravoure had an expiative effect (MA 1854/89/25-26). Similar regulations are found in the Mahåbhårata (12.165.46), according to which a killer of an embryo should go to war ready to die in order to expiate his sins. Traditionally, persons travelling over the ocean had to get rehabilitation after their return: atha patanyåni. samudrasåyånam. One of the kalivarja sins causing loss of caste is undertaking a sea voyage (Baudhåyanadharmasütra, 2.2.1-2).57

Even outside Nepal, as the greatest crime in a brahmanically guided society was ‘the murder of a Brahmin’ (brahmahatyå), especially from the end of the Bråhmañas period onwards, Brahmins were virtually immune from capital punishment (at least as meted out by nonBrahmins). This made them virtually58 eligible to take to task and threat people in power without risking their lives for it. When they happened to complain, in such a public way as a declaration of pråya, of a situation which, beside affecting them personally, would also negatively affect the rest of the community, they were certainly perceived as generous benefactors. Fear of brahmahatyå, from very early on, had a very powerful effect, and a similar strong interdiction applied also to the category of ascetics, whom it was equally forbidden to kill. Moreover, the ascetics were both admired and feared, because it was believed that in their wrath they could throw dramatically dangerous curses, that never failed to hit their target. In fact, the earliest mention of pråya recorded so far,59 in the Kaußœtakiyopañißad, seems to refer to a begging ascetic:60 Here, as a simile in a metaphysical discussion, is introduced a case: As if one, after begging a village and getting nothing, should fast (on the village) saying ‘I would not eat now even if (the village) should give’, and then those same (villagers) who previously should repulse him come and urge him saying ‘Permit us to give to thee’61 […] The reason for the sudden and insistent generosity pictured here can be only that the vil57

Personal communication. I insist on the virtual nature of this injunction because there have been cases where the people in power did not hesitate to kill Brahmins. There are numerous examples just in the Kashmirian texts examined. In particular, when describing an extremely corrupt officer, Narm., 1.57, says: “Massacres of Brahmins / do not worry him, / so why should he care / about slaughtering cows? / Through his loyalty to his master / people are thoroughly uprooted”, (brahmahatyå na gañayante govadheßu kathaiva kå Ù prabhubhaktikr¢tå yena mülåd unmülyate janaΔ ÙÙ). 59 Its first discovery is due to Hopkins (1900: 158-159). 60 Kaußœtakiyopañißad, 2.1. Transl. by Hopkins 1900. 61 In a note Hopkins remarks that the term used (’pavi†en) is the same as that used in the epics and gives this text: yathå gråma∫ bhikßitvå’labdhvo’pavi†en nå’ham ato dattam 58

V. Violating Tradition and its Boundaries

533

lagers fear that the beggar will starve himself to death out of revenge, and that they will suffer the usual consequences of pråyopave†anå. (Hopkins 1900: 159)

The term bhikßitvå seems to refer here to a wandering ascetic, rather than to a begging brahmacårin. In this passage a whole village is held responsible for the eventual death of the suppliant. The belief here is in a future catastrophe, a period of bad luck for the entire community, as if the death acted as a silent curse, or black magic. In later cases, notably in the epics or in the most ancient law books, a suppliant sat in pråya to protest against a specific person, be it the one who was immediately responsible for the wrong, or the king who failed to prevent it and to protect the right of the protester. In spite of enjoying popular support, at times the fasters stance was nevertheless considered in a negative light. When the faster is not in good faith, or desires to obtain too much, this can become a case of adharma, as already found in Vaikhånasagr¢hyasütra, 5.11 (see § 2.1, note 31). In medieval Kashmir there are numerous examples of excessive recourse to pråya, that at times has been considered a political evil. 3.1. A question of outlook: how an ancient story of brahmahatyå could travel from a brahmanical into a Buddhist environment, where it becomes a case of cåñ∂ålahatyå, while much later it is known again as a case of brahmahatyå. In an ancient Buddhist text we have a similar case of pråya that could influence negatively a largish group of people, but here the social status of the faster is at the opposite end of the social range. The person fasting is a cañ∂åla, of the despised Måtaõga type (this is actually said to be his name), and his death on someone’s premises could be extremely polluting. Here also decent, self-respecting people must avoid his death at any cost. The episode is narrated in a Påli commentary, the Papa∞casüdanœ, to the Upålisuttam,62 that recounts of a pråyopave†a done in order to obtain one’s desire. The text was first translated in 1888 by Féer (1888: 131-152) and it is the third story narrated by the commentary, about King Mejjha and r¢ßi Måtaõga. In this particular birth the future Buddha was born in a family of cañ∂ålas, he was 16 years old, was called Måtaõga and lived in a house made of hides. One day he was seen in the street by Dittamaõgalikå, the a†nœyam iti ya evaina∫ puraståt pratyåcakßœraõs ta evainam upamantrayante dadåma ta iti (Kaußœtakiyopañißad, 2.1). 62 Féer (1887: 309-340), had introduced the Upålisuttam (Le sütra d’Upali), one of the 150 sütras of the Majjhimanikåya, the second section of the Suttapitaka. Féer (1888) introduces also the Papa∞casüdanœ (Le commentaire de l’Upåli-suttam).

534

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

beautiful, proud only daughter of a rich merchant. Angry at the inauspicious (or simply polluting) sight she immediately went back into her house to purify her eyes with water. The citizens, disappointed in their eagerness to see her, beat Måtaõga and threw him half dead on a heap of rubbish. But after a while he partially recovered and went to lay down in the courtyard in front of the house of Dittamaõgalikå, saying to himself that if he could obtain Dittamaõgalikå he would get up, and if he did not, he would let himself die there. Her father heard about him and of the insulting behaviour of his daughter, and tried to appease him with money. But he obstinately claimed the girl. 63

Féer then translates:64 In those times it was a rule in Jambudvœpa that, if a cåñ∂åla died of wrath at the door of one’s private quarters, all those who lived in those quarters would become cåñ∂ålas; if he died in the middle of the house, all the inhabitants of the house would become cåñ∂ålas; if he died at the door, those who lived inside on both sides would become cåñ∂ålas; if he died in the courtyard, here seven, there seven, fourteen people would become cåñ∂ålas. The father then posted his men outside in order to protect Måtaõga, fearing that somebody might try to kill him. After five days of his laying there the other people who lived in that house, seven and seven, asked the parents to grant their daughter to Måtaõga (so that they would not become cåñ∂ålas). The parents refused, but on the seventh day of his fast the same people forced their way into their house, took hold of the girl and after changing her rich ornaments into ornaments fit for a cåñ∂ålœ, gave her to Måtaõga. The future Buddha then humiliated her by having her carry him throughout the city on her back, and finally introduced her into his house made of hides.65

This story makes several interesting points, first in that here cañ∂ålahatyå seems to carry almost the same negative weight as brahmahatyå. It then shows how popular sentiment and fear of pollution –and we may presume that the class to whom the other inhabitants of the house belonged was also a merchant class, or to that of some lower artisans and menial workers, but still higher than that of cåñ∂ålas– in the end prevails over manners, so that a sort of minor popular uprising takes place in help of the claimant, not for altruism but for sheer self protection. Finally the wider scope of the story shows that, given the impermanent nature of everything, pride over one’s social standing is ridiculous, because it is absolutely relative: in that particular birth, the Buddha himself was a despised outcast. There is a pervading humorous flavour to the whole episode, and part of it must derive from the fact of being a parody in reversal of some older brahmanical story, traces of which seem to survive in an 63

Here I paraphrase the translation of Féer 1887: 133-135. Here I only translate Féer’s French into English. 65 This is not the place for it, but it would be interesting to trace the subsequent story of this theme on purely social grounds, as traces of it remain in the story that provided also the subject matter of the Shakespearian ‘The Taming of the Shrew’. 64

V. Violating Tradition and its Boundaries

535

episode of Kathå., where the people involved are Brahmins and kßatriyas. It is the story of the beautiful A†okamålå of kßatriya descent and of her suitor, the ugly and rich Brahmin Ha™a†arman, (Kathå., 9.2.30-91). Ha™a†arman asks the girl in marriage of her father, but she refuses him and declares to her father that if she is given to him, she will run away. The Brahmin, though hearing this, starts to fast in her house, until her father, fearing brahmahatyå, gives her to him (Kathå., 9.2.37 [taccrutvåpy akarot tåvaddha™ha†armå gr¢he pituΔ Ù pråya∫ yåvad aha∫ dattå tenåsmai vadhabhœruñå]). The story from here on is different, as the young woman, true to her word, runs away after the marriage, to the great humiliation and anger of the bridegroom. Its first portion, however, is very similar to the Buddhist story. In both cases pråya is resorted to by a male in order to impose his will on an unwilling female. There is humiliation involved in both stories, though in A†okamålå’s it is the abandoned husband who suffers it. In the beginning this suitor’s only fault is his ugliness, so perhaps one could say that even this woman was too proud, for this would be a normal eligible situation, with the man, the prospective bridegroom, belonging to a more prestigious varña than that of the woman, and rich besides. On the other hand, from at least the times of Mahåbhårata,66 kßatriya women, thanks to the custom of svaya∫vara,67 have been able to have some say in the choice of their spouses, so it was only natural that A†okamålå not only express her wishes to her father, but also that she acts in accordance with her feelings and courageous words when she runs off. The main differences are in the religious background and social class of the protagonists. The religious atmosphere in fact is what puts the whole situation in a different perspective: the fact that Måtañga is the Buddha in a previous life changes everything from within, and this is the real innovation in a probably lost original story, which the Kathå. may have reproduced. This story is also instructive in that it shows a different Buddhist perspective on pråyopave†a. Usually, especially in the Jåtakas, one reads of the future Buddha as a being who sacrifices his body for the good of the creatures, or of old Buddhist monks relinquishing their worn out bodies through voluntary starvation. 3.1.1. A question of outlook. The consistency of påpas: cañ∂ålahatyå versus brahmahatyå In the Buddhist story, the result of what could be called cañ∂ålahatyå, or rather what happens when a cañ∂åla dies of anger, seemed 66

See for instance the famous story of Nåla and Dåmayantœ. Obviously not when this was a tournament, in which the best warrior would take the bride, but in those cases where the young women were encouraged to choose among some previously selected candidates. 67

536

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

to acquire a substantial consistency, that would attach itself in a particularly obnoxious way to a given number of people. The same is true of brahmahatyå as exemplified in the first instance of pråyopave†a in the Råj. (4.82-116), which occurs in an episode relating to the life of King Candråpœ∂a:68 Once a bråhmaña woman who was undergoing voluntary starvation (pråyopave†a) and had been questioned [on this account] by the law officers (dharmådhikårin), spoke thus to the king as he was seated in court’.69 The woman laments the violent death of her husband, and says (4.84): ‘It is indeed a great humiliation for a king of righteous conduct, if untimely death overtakes his subjects’.70 The woman then expressed her suspicion about another Brahmin, but the king, without proof, could not act, especially against a Brahmin: (4.96) ‘Not even another person can receive punishment if his guilt is not established: still less a Brahman, who is exempt from capital punishment although guilty’.71 Then the woman says, somewhat threateningly: ‘Four days have passed, O king, without my having taken food. I have not followed [my] husband [into death], because I was anxious for retaliation on the murderer. As this [man] has not received punishment, I seek death by starvation’.72

As the woman persisted, King Candråpœ∂a resorted himself to prayopave†a, at the feet of an image of Vißñu, and after three days of it he had a dream in which he was instructed on how to proceed. The temple courtyard’s floor would have to be covered in rice powder, and the culprit, at night, would have to be made to circumambulate three times around the temple. If behind his footprints were seen the footprints of Brahmahatyå, then he would be found guilty. This trial was made, and the extra footprints were discovered. Thus the suspect was revealed to be the culprit, and as he was a Brahmin, he was not sentenced to death, but punished otherwise. That Brahmin later performed abhicåra, black magic, in order to kill the king, but even then, though about to die of it, King Candråpœ∂a refrained from destroying him, on the account of his brahminhood. Here one sees the belief in the personification of the invisible brahmahatyå who follows the culprits around as a female spectre even during their lifetime. There is also, doubly stated, the strong interdiction to kill a guilty Brahmin, and even if he is guilty of high treason (see supra, Falk 68

Transl. by Stein 1989. Connecting paraphrase by Baldissera. Råj., 4.82: kadåcana sabhåsœna∫ pr¢ß™vå dharmådhikåribhiΔ Ù pråyopaviß™å råjanå∫ bråhmañœ kåcid abravœt ÙÙ. 70 Råj., 4.84: eßaiva mahatœ lajjå sadåcårasya bhüpateΔ Ù yad akålabhavo mr¢tyus tasya sa∫spr¢†ati prajåΔ ÙÙ. 71 Råj., 4.96: nånyasminn api dandasya prasaõgoni†citågasi Ù ki∫ punar bråhmaño dañ∂yo yo doße’pi vadha∫ vinå ÙÙ. 72 Råj., 4.97b-98: catasraΔ kßañadåΔ kßœñå råjann ana†anasya me ÙÙ nånv agå∫ pariñetåra∫ hantuΔ praticikœrßayå Ù tatråvihitadañ∂e’smi∫s tyajåmy ana†anair asün ÙÙ. 69

V. Violating Tradition and its Boundaries

537

observations, and the MA). A further point is the double pråyopave†a of the protagonists, as the king also undertakes it in order to discover the truth. Formally the double pråyopave†a could look like a duel of resistance between claimant and king, as in the case of creditor and debtor, but here the king does not starve against the woman, but eventually against the fault itself, and in order to propitiate Vißñu and obtain a boon. His attitude recalls more that of the preliminary purification undertaken for a particular dœkßå: the kings prepares himself to enter a delicate situation, where it is extremely important that he judges correctly. The notion of brahmahatyå (and of its relative/objective cañ∂ålahatyå) is very important here because the whole range of emotional responses to pråya, whether or not it was performed by actual Brahmins, is imbued of these cultural connotations, as emerges also from the following stories and historical records. 4. Single and collective suicide in medieval Kashmir: examples from Kathå., Råj., Narm. and De†.73 4.1. Suicide stories in Kathå74 The Kathå. presents a variety of issues, that go from the possibility of suicide of a pretended madman crazed by love to the threat of suicide, to suicide from despair, or religious suicide at the end of one’s life. Heroes go to death alone, or accompanied by their wives and counsellors; desperate couples ingest poison together, and a group of kings are ready to ritually fight each other in a fight to the death, in an episode that seems to anticipate later Rajasthan chivalry. - Story of the Brahmin hero Vi∂üßaka (Kathå., 3.4.70-406), who has fallen in love with a supernatural being, and pretending to be crazy refuses to eat and roams about wildly (249). The king (who is also his father-in-law) decides to let him free to roam, fearing brahmahatyå if he is checked (251). - Main story (Kathå., 8.2.74), some kings whose daughters (all promised in marriage to Naravåhanadatta) had been magically abducted were ready to commit suicide by killing each other in battle (considered to be the honourable mode of killing oneself for kings) but were stopped by a heavenly voice. - Main story (Kathå., 16.1.80-84), the King of Vatsa in his old age with his two wives and his counsellors throws himself into a precipice75 from 73 This material, again elusive and interrelated, has been arranged in different categories, so that successive points emerge in different subchapters. 74 Here simple suicide, usually due to despair, is distinguished from pråya of the claimant. 75 For the extreme frequency of religious suicides from precipices, see Sax 1992: 204-206.

538

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

the top of mount Kåla∞jara, later called (16.1.91) mahåpathagiri, ‘the mountain of the great voyage’, ‘the mountain of death’. - Story of the wise Buddhist King Vinœtamati and the robber, (Kathå., 12.5.190-369). The robber tried to commit suicide twice, first because his parents had died for his sake, then because he was not able to find Vinœtamati and obtain wise teachings from him. (He eventually finds him, and the king then instructs him in the Buddhist perfections.) - Story of the Brahmin woman, the donkey and the washerman: in Kathå., 12.5.205-215 a Brahmin couple commits suicide by ingesting poison (214), as a judge had given an outrageously wrong judgment of their case.76 (The king then has the judge executed for causing the death of a Brahmin.) - Story of Müladeva: in Kathå., 12.22.89 a Brahmin woman –or rather, a young Brahmin temporarily transformed into a woman– declares that she will commit suicide if the man her guardians want her to marry (even though she had been left in their care as already married to a Brahmin) touches her before having gone on a pilgrimage for six months. She says (78): “Otherwise, know that I shall be dead having cut my tongue with my teeth” (eva∫ na cet kr¢tajihvå∫ dantair jånœhi må∫ ∫r¢tåm). - Story of Jœmütavåhana (second version): Kathå., 12.23.70-80 presents a case of attempted suicide for love. The young princess Malayavatœ, overcome by love for Jœmütavåhana and incapable of waiting for the outcome of their story, is about to hang herself in front of the temple of Gaurœ when a divine voice stops her and tells her that Jœmütavåhana will soon be her husband.

In none of these cases is there any question of the suicide being a crime, or forbidden. 4.1.1. Suicide stories in Kathå.: people who sacrifice themselves to goddesses The Kathå. presents many instances both of helpless people kidnapped in order to be sacrificed to wild goddesses77 by some tribal group, as well as of people of different social standing who declare themselves willing to be human victims in sacrifices to deities or demons. Here is a brief enumeration of those who voluntarily sacrificed themselves to goddesses:78 - Story of the Brahmin and the divine garden (Kathå., 1.6.71-85). A lazy Brahmin decided that he would sacrifice himself “this stupid beast” 76 The Brahmin’s pregnant wife had chased a washerman’s donkey that was plundering her vegetable plot. The donkey fell in a ditch and split his hoof, and the angry washerman beat the woman with a stick and kicked her so that he had her miscarry. The judge to whom the Brahmin couple appealed decreed that the Brahmin had to carry the donkey until its hoof was healed, and that the washerman had to impregnate the Brahmin’s wife, as he had caused her an abortion. 77 For an assessment of human sacrifices to wild goddesses in Kathå., see Baldissera 1996, passim. 78 A particular case, treated above, is that in which people are willing to sacrifice themselves to a deity for the sake of another.

V. Violating Tradition and its Boundaries

539

(mürkham ima∫ pa†um),79 to Vindhyavåsinœ. The goddess however stopped his sacrifice and ordered him to create a divine garden. From that time on, he became known as (75): “a Brahmin who kept the vows of silence and did not eat” (maunœ niråhåro dvijaΔ). - Story of the Brahmin Jœvadatta who wants to sacrifice himself as he had not been able to revive his dead beloved (Kathå., 9.2.160). Gaurœ/Vindhyavåsinœ stops him. - Story of the kßatriya Ya†ovarman (Kathå., 9.4.151-209). After the death of his wife, and as the king had stopped showing him favour, Ya†ovarman decided to go to Vindhyavåsinœ (161-162) and do severe penance in front of her [image] without eating (niråhåras tapas), sitting on darbha grass. As he was engaged in this extreme fast, the goddess appeared to him and granted his wish. - Story of the exchanged heads (Kathå., 12.13.5-3). In a fit of bhakti a young washerman (rajaka), just married, who had gone to see the goddess, sacrificed his head to Gaurœ in her temple; his brother-in-law, on seeing him dead, became so upset, that he did the same. When the newly wed wife came upon the sorry sight, she somehow blamed the goddess for depriving her at once of both her male supporters, in spite of her constant devotion to her. And as she was about to hang herself, a heavenly voice spoke, instructing her to reconnect the heads of her relatives to their bodies. (The young woman obeyed and the two men revived, but in her flurry she connected the husband’s head with the brother’s body, and the brother’s head with the husband’s body). Here three people of the very low rajakas class sacrifice themselves to the goddess, the first in a fit of devotion, the other two because they were overcome by despair. The woman’s words to the goddess, though, carry perhaps the slightest hint of defiance, in that they sound like a veiled moral blackmail.

The next instance is that of a great prince, destined to become the emperor of all vidyådharas: - Story of Naravåhanadatta and Cañ∂ikå Kålaråtri (Kathå., 15.1.90-105). Kålaråtri, who guarded the northern entrance of the Tri†œrßå cave of the Kailåsa, had put under a stupefying spell the whole army of Naravåhanadatta. After trying, to no avail, to please her by a hymn, Naravåhanadatta offered her his head in sacrifice, but as soon as he held his sword ready to strike, Kålaråtri restrained him and wished him victory (104-105).

Even in stories of people who are imprisoned (usually by tribal groups) in order to be sacrificed to some deity against their will there are sometimes details that point to voluntary human sacrifices. In the story of prince Sundarasena (Kathå., 12.34.41-384), for instance, there is a very clear reference to previous self sacrifices of vœra devotees. The prince was taken as a sacrificial victim to a temple of Ambikå, described in these terms (301): 79 As the story is told by a Kashmirian author of the eleventh century, when esoteric ‡aivism was very popular, one could even trace an esoteric ‡aiva overtone in pa†u, which has also the meaning of ‘an uninitiated person’.

540

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

It looked like the mouth of death: the flame of the lamp looked like his80 rolling tongue, the row of bells was like a garland of teeth, gruesome with the hanging heads of the [self sacrificed] vœras. (dœpajvålåcalajjihvaμ ghañ™ålœdantamålayå Ù vyåsaktavœra†iraså ghora∫ mr¢tyumukha∫ yathå).

It appears that the sacral aura of these devotional self sacrifices of people who sometime were deified81 contributed to heighten the prestige of ritual suicide, in whatever form it occurred. 4.2. Suicide stories in Råj.82 The Råj. presents some instances of suicide induced by old age or despair, and it supports the view that suicide settles questions of honour. Though for kßatriyas it is best to die in battle, any form of suicide is acceptable, as dishonour is to be shunned at all costs. There are also a couple of instances of suicide occasioned by fear, but this seem to have been committed by Brahmins.83 - Two kings killed themselves in fire: King Mihirakula, (Råj., 1.309; 1.312-316), and according to one of the reports on his death, also King Lalitåditya (Råj., 4.368). - In Råj., 7.852 King Utkarßa, dejected, kills himself secretly with a pair of scissors. - In Råj., 7.1407, King Harßa’s ministers suggest to him suicide in order that he may not suffer indignity at the hands of his enemies. But Harßa thinks otherwise, 7.1408: “He replied to them: ‘I am incapable of killing myself. Therefore you should slay me when misfortune arrives’”.84 (The ministers however decline).

Occurrences of mass suicides are first recorded in Råj., 4.631-633: King Jayåpœ∂a oppressed the land, so that many Brahmins killed themselves. Once he said “[…] let it be reported [to me] if a hundred Brahmans less one die in a single day” (Råj., 4.633b). Later in fact, when it happened (4.638) that 99 dispossessed Brahmins sought voluntary death in the waters of the Candrabhågå river, he stopped confiscating agrahåras. The following are instances of suicides or attempted suicide occasioned by fear: 80 In Sanskrit death is personified as a male being, like the representation of death as ‘the grim reaper’ in English. 81 For comments on the deification of people who suffered a violent death, see infra, § 8.1. 82 Suicide by widows (and especially instances of satœ) shall be examined separately, see infra. 83 Their varña is not mentioned, but one is a magician and the others are counsellors, so they could be either Brahmins or kßatriyas. 84 Råj., 7.1408: sa tån uvåca sva∫ hantu∫ na †akto’ha∫ tato mayi Ù bhavadbhir eva vißame prahartavyam upasthite ÙÙ.

V. Violating Tradition and its Boundaries

541

- In Råj., 5.241. a magician managed to have King Gopålavarman (902904 CE) die of a magically induced fever. Later, afraid of punishment, the culprit committed suicide. - During the reign of Jayasiμha (1128-1149 CE) two ministers who had not managed to accomplish their mission thought of committing suicide (Råj., 8.2612).

4.2.1. A finger breadth of courage or of submission? Cutting one’s finger by Caranas and in Råj. Courage is the highest value for the martial classes, and is believed to be rewarded also by deities, so that it would not seem to generate controversial issues. At times however the same martial custom, that seems to stem from an identical tradition of honour and resistance to physical stress, seems to go in two opposed directions. Regarding this particular custom, recorded in Råj., it would seem that the cutting off of one’s finger in the presence of a stronger opponent, or of an adverse party, was a sign of submission (see Råj., 5.150; 8.1594, 1738, 2272, 2308, 3300). On the other hand, in the case of the Caranas, a low martial caste from Sindh, Rajasthan and Gujarat, who pledged their honour and their life on their ability to safely guide and guard caravans as they traversed desert tracts, the cutting of a finger (or of each successive finger) was the proud sign that they were ready to commit suicide if robbers attacked the caravan in their charge.85 In Japan, the same practice, especially in the contemporary criminal underworld, amounts to admitting one’s wrong, and to asking forgiveness, so that it would be in keeping with the symbolic meaning found in the Råj. occurrences. On the other hand from recent history we are also acquainted with the silent, highly dramatic protest against the Vietnam war of some Vietnamese Buddhist monks who burnt themselves alive, or just stood motionless while burning one of their fingers. Looking at these different examples of the same act, it would seem that they all proceed from the same source, an extreme ability to show both restraint with extreme physical pain and courage. It is again an example of the closeness of the heroic and the renouncer traditions, as the Vietnamese instance shows clearly. These people mutilate themselves (like bhakti devotees often did to please a bloodthirsty goddess) to show that they put themselves under the protection of their opponents, who should show respect for both their courage and helplessness, and treat them with equanimity. As in the ‘pråya for redress’ practice, there is an element of blackmail which here appears twofold: if the enemy, opponent, robber treats them unfavourably, the blame for their death or mutilation will attach to them, and not only tarnish their honour, but also negatively influence their future birth with bad karma. So it is both a case of worldly honour and of reward in another world. 85

See, for instance, Hardy 1995: 105-106.

542

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

4.3. Suicide stories in Narm. Narm. has a few verses that show not only the frequency of recourse to pråyastha86 as well as to other types of death through despair, and the particular reasons why it was resorted to, but also the fact that the moral blackmail often did not work its magic. Desperate people threatened to kill themselves and were left to do it by cruel village officers. Narm., 1.84a has a treasurer, ga∞jadivira, who is “proud of the ill treatment (khalœkåra) inflicted on people reduced to fasting unto death” (sa pråyasthakhalœkårån månœ). Narm., 1.120 shows a whole village harassed by an officer, niyogin, where people died in hundreds and thousands, while Narm., 1.122 shows what sort of actions prompted the suicides. In both cases the villain is the niyogin, ‘officer, bureaucrat’: In his [the niyogin’s] village there was a likeness of hell as thousands of people had died by hanging themselves from trees, and hundreds had starved through fasting to death.87 ‘Confiscate the entire property! Imprisonment! Corporal punishment! Break into their houses!’ such terrifying utterances were never far away from his mouth.88

4.4. Altruistic suicides in Kathå. and Råj. Suicides of kßatriyas and also of courageous Brahmins often occur for more noble reasons than personal honour or the furthering of their own advantage. Both works report cases of generous self sacrifice of both ministers and kings, while Kathå. has instances of heroic individuals who sacrifice themselves for others, or for the success of others: - Story of Jœmütavåhana (Kathå., 4.2.15-255), a vidhyådhara prince, who sacrifices himself in order to save a young någa that should have been eaten by Garu∂a. When the naga notices that Jœmütavåhana is being eaten in his stead, he tries to persuade Garu∂a to stop and eat him (234235). So here two people are ready to sacrifice themselves to save each other, and in the end through the favour of Gaurœ, pleased by their courage and generosity, Garu∂a stops exacting living tributes from the Throughout his satires Kßemendra uses this term rather than pråyopave†a. Narm., 1.120: vr¢kßårohasahasreßu pråyaΔklånta†ateßu ca Ù gråme tasya vipaññeßu narakapratimåbhavat ÙÙ. 88 Narm., 1.122: sarvasvaharañam bandhonigraho gr¢habha∞jana Ù iti tasya mukhåd ghora∫ na cacåla vacaΔ sadå ÙÙ. 86 87

V. Violating Tradition and its Boundaries

543

någas, and Jœmütavåhana as well as all the eaten någas are revived and their limbs restored. (This story has a second version in Kathå., 12.23.5205, which is slightly different). - Story of the ‡abara chief who is willing to be sacrificed to Cañ∂œ instead of the appointed victim (Kathå., 4.2.65-70). The goddess grants him his life and the boon of being her devotee forever. - Story of the heroic King Vikramatuõga, the Brahmin and the bilva fruit (Kathå., 7.1.55-79). King Vikramatuõga went hunting, and in the woods he saw a Brahmin attending to a sacrifice of bilva fruits. Returning from the hunt, the king saw the Brahmin still attending to his sacrifice, and asked to help him. After some hesitation, the Brahmin gave him a fruit, and the King sacrificed it for the sake of that Brahmin while silently meditating thus (67): “If you are not satisfied with this bilva fruit I am giving you, Agni, I shall offer you my own head” (hutenånena bilvena na cet tußyati tacchiraΔ Ù tvayy agne sva∫ juhomœti dhyåtva tasmi∞ juhåva tat ÙÙ). Agni immediately granted him a boon, to the delight89 and dismay of the Brahmin who had sacrificed bilva fruits for many years. So the fire god explained to him (76): “Success quickly presents itself to those who posses extreme resolution90 while it presents itself after a long time to those possessed of but a weak resolution” (tœvrasattvasya na ciråd bhavanty eva hi siddhayaΔ Ù mandasattvasya tu ciråd tu brahman yußmåddr¢†asya tåΔ ÙÙ). - Story of the young prince who wants to sacrifice himself twice for his brother (Kathå., 7.8.165-175). His sacrifice is stopped by Vindhyavåsinœ. - Story of the child Sattvavara, Vœravara’s son, who willingly agrees to be sacrificed to Cañ∂œ for the health of his king (Kathå., 9.3.141). He undergoes sacrifice, and later is revived. - Story of the Brahmin Vœravara, who willingly sacrifices himself to Cañ∂œ for the health of his king (Kathå., 9.3.165-180). The goddess restrains him, grants another hundred years to the king, and restores to life his family. - Second version (Kathå., 12.11.20-105), of the same story, where also the king (King ‡üdraka), moved by their courage, is about to sacrifice himself in order to restore the lives of the heroic family. Goddess Cañ∂œ stops him, then the vetåla asks his question: “Who has been the most heroic?” and the king replies: “King ‡üdraka”, explaining that kings usually have others sacrifice their life for them, not the opposite (131).

Råj. shows the noble examples of three royal counsellors to three different rulers who killed themselves out of loyalty to their masters. Two of them actually tried to save the life of their respective lords by a desperate endeavour, the third committed suicide when his king’s son decided to leave the throne. - In Råj., 4.277-300 the counsellor of an enemy of King Lalitåditya mutilated himself in order to cheat him, and was even prepared to die to carry out his deception. He pretended that his mutilation had been a punishment dealt to him by his master and declared that he was seeking revenge. With this pretext he led astray Lalitåditya’s army in a sandy waste. 89 90

King Vikramatuõga had wished that the Brahmin could become very rich. Or ‘courage, firmness, strength’.

544

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

- In Råj., 4.551-577 Deva†arman, counsellor of King Jayåpœ∂a, saved him from jail by killing himself in order to be used by his king as a life-buoy of inflated skin in the troubled waters of a nearby river. - Another case of extreme loyalty to the idea of royalty is that of a counsellor who killed himself in water, accompanied by his wife, when his king’s son (Kuvalayåpœ∂a) abandoned the throne for an ascetic life. Råj., 4.391 says that Minister Mitra†arman “abandoned life at the confluence of the Vitastå and Sindhu, and was followed by his wife”. - Among single examples of fasting out of loyalty, the Råj. extolls also an animal, that gave a vihåra its popular name. The Råmåyaña showed several monkeys fasting unto death, and here instead there is a she cat (Råj., 8. 2410-13), who wails and refuses food after her mistress dies (Råj., 8.2413b [utsr¢janty åhr¢ta∫ bhojya∫ så†ucå jœvita∫ jahau]).

Finally the Råj. presents the example of an heroic king who sacrifices himself for others, like the kings in Kathå.: - In Råj., 3.31-61, King Meghavåhana offers himself in sacrifice to Cañ∂ikå (also called Cåmuñ∂å in this episode) in order to both save the innocent victim and the sick little son of the Kirå™a sacrificer. The king says, (Råj., 3.40): “I make my own body an offer to Cañ∂ikå. Strike boldly, may these two persons live!”. Later Meghavåhana, who does not want any living creature to be killed, is ready to die again, this time when a Brahmin complains that his sick son will die if an animal is not sacrificed to Durgå (Råj., 3.82-93). As the king offers himself for a victim, Durgå heals the boy just before the king’s sacrifice.

5. Satœ in Kathå. and Råj. In the eleventh and twelfth centuries instances of satœ appear to be very frequent. The differences between the two sources examined are more marked here than elsewhere. The Råj. in fact gives numerous reports of actual queens, and of both female and male servants who burnt themselves at the death of either kings or queens, whereas the Kathå., though giving frequent cases of satœ, does not relate interesting episodes because such instances are taken as a matter of course, and are usually only told in passing. Many characters in fact, mainly of kßatriya or Brahmin descent, recount that at the death of their father their mother ‘followed him’, generally implying that she burnt herself, but without recounting particular details of such deaths. - The first instance, that can serve as a paradigmatic example of most instances, is that of the wife of the heroic King ‡atånœka (Kathå., 2.1.17), whose satœ is expressed in these words: “and the queen followed [in death] the king” (ca devœ tå∫ nr¢pam anvågat).

Of some interest are two episodes, where the personality of the satœ is described at some length. The first is a supernatural creature, who lures a hero into a wonderful adventure, the second an extremely courageous Brahmin woman, wife to a hero:

V. Violating Tradition and its Boundaries

545

- Story of the Brahmin hero A†okadatta (Kathå., 5.2.140-145). On one particularly dark night, the hero sees in a burning ground the beautiful wife of an impaled thief, who asks his help to give her husband water, and tells him she wishes to ascend the pyre as soon as her husband dies (Kathå., 5.2.142b): ni†citå†å sthitåsmœha citårohe sahåmunå. She is in fact a supernatural creature, who deludes A†okadatta with her magic, and later shall give him her daughter in marriage. - An extraordinary heroic example is that of the wife of the Brahmin Vœravara, who asked him to build her a pyre after she had seen both her children die (Kathå., 9.3.155-160). Her last words are a wish that her death be for the prosperity of the king, for whose sake her young male child had just been sacrificed.

This last example is a border case, because by her last wish the woman could belong also to the separate category of people who offer themselves as willing human victims of sacrifices.91 Compared with this offhand treatment of satœ (to the exception of the last two examples) the slightly later Råj., instead, concerned with the actual royal families of Kashmir, presents numerous episodes of known widows ascending the pyre. They usually are queens, concubines, sometimes mothers,92 nurses and female attendants. Interestingly, there are also cases of male attendants following their masters in fire. What is more, two of these satœs utter curses, a queen against the people who had caused an estrangement between her husband and her son, and another noble woman against the king who had decreed the execution of her husband. Both women are mothers and faithful wives, and the curse of a mother, as well as that of a faithful wife, carries great weight. By their undertaking satœ, moreover, their stance is assimilated to that of the old ascetic: the curse has therefore a very powerful triple connotation of strength: - The first case of satœ mentioned, without much ado, is that of Queen Våkpuß™å (Råj., 2.56-57), while the next one is that of the wife of Minister Mitra†arman, presumably a Brahmin woman (Råj., 4.391). - At times satœ is used as a term of comparison, from which it appears that this practice was very common (Råj., 4.501): “What the embrace of the wife who is eager to follow [her husband] unto death, is to the [dead man] raised on the funeral pyre […]”. - Sometimes with a dead king both his wives and male attendants undertook self-immolation in fire. An example is Råj., 5.226-227: at the death of ‡a∫karavarman (883-902 CE) three queens, a Velåvitta93 and two male servants followed him on the funeral pyre. - Råj., 6.107 records that only one queen burnt herself with King Ya†askara (939-948 CE). A little time later (Råj., 6.138-144) another of Ya†askara’s queens, upon being courted by the new king, the low-born 91

See below, in next section. Though the case was specifically forbidden in the dharma†åstras, and still in the Nepalese M.A. See for instance Michaels 1992: 112 (case 8): “Wenn ‘nur’ der Sohn verstorben ist”. 93 Probably a kind of retainer, found also later in Råj. 92

546

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

usurper Parvagupta, unexpectedly sacrificed herself to escape his advances (Råj. 6.143): “The pious queen [thereon] suddenly sacrificed her body in a sacrificial fire which had been nourished with ghee, together with a full offering (pürñåhuti)”.94 - Råj., 6.194-196 presents a different picture with Queen Diddå. At the death of her husband the queen first wishes to burn herself, then she becomes scared, and a friendly minister on noticing it prevents her: “But in front of the funeral pyre she felt regret, and the minister Naravåhana, moved to compassion, prevented her by persistent remonstrances from seeking death”.95 - In Råj., 7.103 Bimbå, the daughter-in-law of minister Tuñga, a ‡åhi princess, entered the fire four days after the death of her husband. - Süryamatœ,96 wife of King Ananta (1028-1063 CE), after cursing the people who had caused enmity between her husband and her son, leapt from her litter into the fire (Råj., 7.461-478). Three male and three female servants followed her in the fire (7.481). This is interesting, because the curse of a faithful wife is considered as binding as that of an ascetic, and is equally feared, as the special circumstances of her death make it particularly effective. But again, the stance of a person who goes to meet death in a formal way is always assimilated to the religious death of the renouncer. - Another queen, one of the wives of Harßa, the future king (1089-1101 CE), when he was imprisoned committed suicide by slitting her throat (Råj., 7.680 [kañ™haccheda∫ kr¢två]). This case, which does not call for particular considerations from Kalhaña, is to be compared to that of Koß™aka’s wife, whose entering the fire when faced with identical conditions is held as a supreme example of wifely behaviour (see infra, last example). Most likely Harßa’s wife’s was seen as a passionate type of response: the proper mode for a widow (or a would-be widow) to confront death is not looking for a speedy end to her sufferings, but rather to patiently await her death, following the model of the ascetic.97 - At the death of King Kala†a (1063-1089 CE), seven queens and a concubine followed him in death (Råj., 7.724). Here are probably stated some of the contemporary notions about the ideal behaviour of royal widows, in that Kalhaña says that it was a scandal, ‘the whole of womankind was disgraced’ that Kala†a’s favourite concubine, Kayyå, did not ascend the pyre. Råj., 7.726 further explains it by saying that she was of low origin (and later in fact consorted with a mere village official, gråmaniyogin, in Råj., 7.727). - In the next generation there is however a counter example, of a glorious concubine who once was a temple dancer, Sahajå, in whose apartment King Utkarßa committed suicide, (Råj., 7.859): “She made her love shine forth brilliantly, just as [if it were] gold, by entering the pyre after smearing thickly over her limbs the blood of her lover, [which resembled] liquefied red chalk”.98 94 Råj., 6.143: så yågajvalena råjalalanå pœtasarpißi Ù pürñåhutyå sama∫ sådhvœ juhåva sahaså tanum ÙÙ. 95 Råj., 6.196: nißißedhånubandhåt tu sånutåpå∫ citåntike Ù kr¢pålurmarañåd etåm amåtyo naravåhanaΔ ÙÙ. 96 The queen in whose honour the Kathåsaritsågara had been composed by Somadeva. 97 About these requisite characteristics, see also Michaels 1990: 31-32. 98 Råj., 7.859: kåntåsragaurikåsyandakr¢tasåndråõgarågayå Ù premño hemna ivaujjvalya∫ pravißyag∞i∫ tayårpitam ÙÙ.

V. Violating Tradition and its Boundaries

547

- In Råj., 7.1380 an old widow enters the fire upon the death of her only son. This, though actually forbidden in the dharma texts, as stated above, occurs more than once, as for instance in Råj., 7.1468, as well as in the previously mentioned case of Kathå., 9.3.155-160. - It would seem however that in the Råj. the self-immolation in fire of widows is considered mainly a question of honour, loyalty and courage, rather than the aspiration to an ascetic ideal, an example that should teach men what is a truly loyal behaviour, as expressed in Råj., 7.1412: “Who is meaner than he who, though being a man, forgets that attachment to the master, the recollection of which makes women enter the pyre?”.99 - In Råj., 7.1468 the wife and the mother of a young prince killed by Harßa set their house on fire and burn themselves in it. - After the murder of Malla (father of future kings), his wife, her sister, two daughters-in-law, two female cousins, and six female attendants burned themselves on the same pyre (Råj., 7.1486-1488). At that time Malla’s wife, the princely Nandå, “mother of the future king” (Råj., 7.1491 [måtå bhavißyato råj∞or]), burned herself together with her nurse, while uttering a curse against the king (Råj., 7.1490-1494). - In Råj., 7.1550-1552 the wives of King Harßa, firebrands in their hands, were ready to burn themselves in a pavilion of the royal palace but were held back by Harßa himself who still hoped to win the fight (1001 CE). Later, as servants were battering the walls of the pavilion to stop the queens from burning themselves, the queens thought they were attacked by the enemy, and set fire to the roof of the pavilion (Råj., 7.1570-1571). Finally seventeen queens burned themselves, while the rest left (Råj., 7.1579). - In Råj., 8.363-369 two queens of King Uccala (1101-1111 CE) burned themselves on his funeral pyre. It seems that one of them, Jayamatœ, would have preferred to be restrained, but the minister she signalled to did not understand her veiled allusion (Råj., 8.363). Kalhaña describes her process to the pyre, and shows that there were people who did not share his great consideration for these unfortunate heroines (Råj., 8.368): “Then as she (Jayamatœ) was ascending the pyre her limbs were hurt by the pilferers who robbed her in eager desire of her adornments”.100 Her behavior allows for a general consideration of a type often found also in Kathå. about the simultaneous coexistence of both firmness and unsteadiness in a woman’s mind (Råj., 8.365-366). In particular the latter verse says: “Though given to unfaithfulness and killing their husbands, yet they step with ease into the fire. In no manner can one be sure of women”.101 - Under King Salhaña both a wife (Råj., 8.445) and a sister of dead men enter the pyre. The latter case is quite remarkable, as she had just reviled her brother. Kalhaña decrees, in Råj., 8.448, that she: “followed the fitting course for a proud woman by entering the fire”.102 99 yoßito’pi vi†anty agni∫ ya∫ dhyåtvå vismr¢ti∫ vrajet Ù bhartr¢snehaΔ sa pu∫so’pi yasya ko’nyas tato’dhamaΔ ÙÙ. 100 Råj., 8.368: atha tasyå†cit åroha∫ kurvantyå bhüßañårthabhiΔ Ù luñ™hakair luñ™yamånåyå vyayå gåtreßu paprathe ÙÙ. 101 Råj., 8.366: dauΔ†œlyam apy åcarantyo ghåtayantyo’pi Ù helayå pravi†anty agniμ na strœßu pratyayaΔ kvacit ÙÙ. 102 Råj., 8.448: ta∫ yå ninindånißpannapurußa∫ tatsvasus tadå Ù tasyå vahniprave†ena siddha∫ månavatœvratam ÙÙ.

548

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

- A very special case (Råj., 8.1223) is that of four noble women, all of them palace attendants, who enter the fire at the death of their mistress, Queen Meghama∞jarœ, chief queen of Sussala (1112-1120 CE, restored 1121-1128 CE). - At times the queens’ burning had to be postponed to a suitable time and place, like in Råj., 8.1440, when four queens of King Sussala had to wait for less troubled times, but still had to be burnt in haste near the palace because their people were kept back by both the enemy’s party and the unbearable frost. - The last example of a woman who burns herself in the Råj., 8.23342339, is also peculiar. This ritual is performed by the wife of Koß™aka, even before she has a chance of becoming a widow. Her husband in fact was only wounded and imprisoned, and might have lived, so Kalhaña sings her praise in five verses, where he reviles the costumes of other ˘åmara and Lavanya women. With her, two male attendants also burnt themselves (Koß™aka in fact died of his wounds in jail only a few days later). The praise of Kalhåña here is very pointed, maybe because she followed the proper mode of killing herself, on the pyre, or because of her dignity, –or perhaps also because it was advisable, for a high courtier, to praise people living in a time close to his own.

As satœ is an extreme proof of loyalty, it can also be compared to the case of the minister who killed himself together his wife when his king’s son (Kuvalayåpœ∂a) abandoned the throne for an ascetic life (Råj., 4.391), which belongs also supra, to § 4.4. 6. pråyopave†a of the claimant in Kathå. The Kathå. presents only two cases of pråya undertaken in order to redress a wrong, in both of which the claimant resorts to the superior judgment of the king.103 The first one presents another instance of the fact that it is the duty of kings to protect the gold of Brahmins, poena their threatening to commit suicide (and see infra the first example of the Råj.). - Story of King Prasenajit, the foreign Brahmin and his stolen one thousand golden dinars (Kathå., 6.7.133-159). The Brahmin, once his gold was stolen, announced his desire to start for a tœrtha in order to abandon his life by fasting (Kathå., 6.7.140): gatvå tœrtham abhu∞jånaΔ pråñå∫s tyaktum iyeßa ca Ù buddhvå ca so’nnadåtåsya vañig anyaiΔ sahåyayau. Here the term for ‘fast’ is abhu∞jånaΔ. (The intelligent king with a thorough investigation discovers the culprit and restores his gold to the Brahmin).

This is another example of the well known habit of Brahmins to sit fasting if their money was stolen, as it meant that the king had not protected their right. Compare what Bharata says in Råmåyaña, 2.111.14-17 (see supra, § 2.5), and ‡atapathabråhmaña, 11.5.71.1 (quoted in Falk 2001: 134 n. 6 [see supra, § 3]). 103 Naravåhanadatta at that time was only the crown prince, but he already had his own ministers.

V. Violating Tradition and its Boundaries

549

The second instance is for a similar claim, but as one does not know the origins of this servant, it cannot be said that this is another case of a Brahmin fasting to retrieve his money, which here had been only promised to him. - Story of the pråyopave†a of the wronged servant (Kathå., 9.5.2-11). A servant of minister Marubhüti, sitting in front of Naravåhanadatta’s door, engages in pråya against Marubhüti because instead of receiving his promised salary all he got were insults and a kick (Kathå., 9.5.5b): “So I sit in pråya against him in front of your gate” (tenopraviß™aΔ pråye’ha∫ siμhadvåre’sya tåvake Ù). (Naravåhanadatta has Marubhüti pay him.)

6.1. pråyopave†a of the claimant in Råj.: the threat posed to royal authority by fasts to the death The Råj. has an impressive number of cases of pråya undertaken for redress. Already about a hundred years before, according to the satirical works of Kßemendra, the frequent situations of misgovernment from the part of the kingdom’s officials (and especially from those appointed to collect the revenue) had instigated numerous cases of collective protests by hunger strike, in which not only several Brahmins, but the population of whole villages had perished. Single cases of fasters who demanded justice were also quite frequent, and sometimes fasts were undertaken for political reasons. Another indication of the frequencies of these protest fasts is found in Råj., 6.14, that shows that there were special officers appointed to investigate and report cases of pråya, called pråyopave†ådhikr¢ta. Their duty was to control that the faster was in earnest, and to report the fact (and eventually also convey the actual faster) to the king. Here (Råj., 6.14-25) a man is fasting against a dishonest merchant who had a document counterfeited in order to appropriate a well fitted with stairs. The man is brought into the presence of the king, who after a clever investigation discovers the truth and punishes the merchant (Råj., 6.26-41). The fact that Kashmiri kings thought of appointing particular officers as a means of protecting themselves from impostors testifies to the popularity of the practice, that underwent also several abuses. This is further corroborated by a much older example of a case of pråya undertaken under false pretences, taken from the Artha†åstra, 5.89.1-9.104 This text prescribes that, in order to get rid of a minister suspected of sedition, the king’s men should induce the minister’s brother to claim his part of inheritance. He would then come at night to sit (upa†œ) at the door of the minister. A spy of the king should secretly kill the faster, and the king should arrest the minister as guilty of homicide. 104

See, also, Renou 1943-1945: 121.

550

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Artha†åstra, 4.7.16 gives also another hint of the frequency of recourse to suicide to redress a wrong, when it says of the informer, sücaka: “In the case of one who has hanged himself, he should find out [if] any improper wrong [has been done to him]”. Here it is implied that the king should punish the eventual culprit. That the problem of fasters unto death was felt as a very real menace to royal policy in the times of Harßa (1089-1101 CE) is alluded to in Råj., 7.879, where it is said that in the beginning of his reign King Harßa had bells put on to his gate so that people could sound them to announce a complaint. It is in front of this gate, called ‘the lion gate’ (si∫hadvåra), that usually single claimants sat fasting, while large crowds of fasters used to gather in the courtyard of one of the main temples of the city. 6.1.1. pråyopave†a of the claimant in Råj.: pråya of a single person Several cases of individuals fasting to redress a wrong are recorded, all for different reasons: - An example of pråya undertaken by a Brahmin in order to retrieve his stolen gold is recounted in Råj., 6.42-66. A Brahmin threatens to kill himself of starvation at the king’s gate for the sake of his golden coins. The king then settles the matter to the satisfaction of the claimant. This episode shows again the enactment of the stipulation that the king has to protect the gold of the Brahmins.105

There are also a few examples of pråya undertaken by kings and pretenders to the throne, such as those of Harßa and Bhoja. Harßa was in jail, whereas Bhoja was under a sort of house arrest by the minister Alaμkåra. - Harßa, when he was a prisoner before ascending to the throne, under-

took a political pråya in jail in order to be exiled rather than remain imprisoned (Råj., 7.745-746). - Later Bhoja, in the same way, threatened to commit suicide more than once, as told in Råj., 8.2703 and 3094, and by that device he also managed to have his way. - An isolated example of a warring lady in pråya for strategic reasons is told in Råj., 7.1172: a strong ˘åmara woman had fasted at the door of King Kala†a to offer him her fort of Dugdhaghåta, but he had refused it.

Obviously fasting in itself carried tremendous prestige and awe in the population, as from ancient times they were used to pay respect to fasting ascetics, believed to subsist on air, whereas on a more mundane scale were acquainted with the ‘fasting duels’ between creditors and debtors. The fast unto death proved an efficacious weapon 105

See supra, § 6.

V. Violating Tradition and its Boundaries

551

against the absolute power of kings, but it could also be used by kings in the same way, as already seen in the first episode recorded in Råj., 4.82-116, when in response to the fast of a Brahmin widow, King Candråpœ∂a started fasting himself. A more extreme example of this ‘countermeasure’ is the solemn promise of King Uccala (1101-1191 CE), expressed in Råj., 8.51: “As he had taken a vow that he would commit suicide if any person should die by starving himself (pråyopave†a), he caused the judges to be careful”.106 A typical illustration of the consequences of a careless judgment was shown above, in the Kathå. story of the Brahmin woman and the washerman’s donkey. By Uccala’s time, however, Kashmiri politics had already been influenced by a number of collective fasts. 6.1.2. pråyopave†a of the claimant in Råj.: collective pråyopave†as of protest Collective fasts in the Råj. seem to have been staged as very dramatic performances, with great masses of people making the loudest possible noise and carrying a vast array of religious props. - The first description of a pråyopave†a undertaken by a crowd in order to exert an incisive political pressure is found in Råj., 5.465-466. In the interregnum after King ‡üravarman II was deposed (939 CE), the Brahmins spent a few days discussing who should be king, and in the meantime a great host of fasting purohitas assembled: “While the Brahmans in this fashion passed five or six days, there assembled an immense host of Purohitas of sacred places (pårißadya), causing a mighty din by their drums, cymbals and other musical instruments, raising glittering flags, ensigns and umbrellas, and carrying seats on load-animals”.107 This is the first large gathering of religious people engaged in solemn fast recorded in the Råj., where there occurs also the first instance of a queen who approaches the crowd of fasters through her envoys in order to negotiate with them and eventually persuade them to side with her son. Råj., 5.468 narrates: “Then the wife of the parricide [king] sent officials to those engaged in voluntary starvation (pråyopave†a), to beg the throne [from them] for her supposititious son”.108 - Later, under the reign of Diddå (980/981-1003 CE), there were other collective pråyopave†as undertaken with political aims. A nephew of Diddå had induced the Brahmins entitled to the most important agrahåras to fast against her rule, and the people went into a great tumult (Råj., 6.336). But Diddå was a subtle diplomat who knew the importance of bribing (Råj., 6.339): “By presents of gold, she gained 106 Råj., 8.51: pråyopaviß™apramaye dehatyågapratij∞ayå Ù nibaddhayåpratyavekßå∫ dharmådhyakßåñ akårayat ÙÙ. 107 Råj., 5.465-466: pa∞caßañi dinåny eva yåvat tasthur dvijåtayaΔ Ù kåhalåkå∫syatålådivådyakolåhalåkulam ÙÙ utpatåkadhvajacchattra†obhi yugvårpitåsanam Ù aße†a∫ pårißadyånå∫ tåvat tatråmiladbalam ÙÙ. 108 Råj., 5.468: pitr¢ghativadhü† channaputraråjyårthinœ tataΔ Ù pråhiñod råjåpurußån pår†va∫ pråyopave†inåm ÙÙ (the ‘son’ had actually been exchanged at birth).

552

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

over Sumanomantaka and other Brahmans, and the fast ended”.109 Some time later the situation reproduced itself in the identical manner (Råj., 6.343-344): the same nephew instigated again a solemn fast, the Brahmins complied, but “as they were willing to take bribes” (utkocåditsayåΔ), the queen’s man, the able minister Tuõga, could easily subjugate them. - Sometimes the instigators of these political hunger strikes, still called pråyopave†as, were the Brahmin counsellors themselves, as in Råj., 7.1314, where they plotted with the Brahmins and purohitas of the pårißadyas in order to oust Tuõga, the strongest minister, protected by the queen. Even later, in the time of King Ananta (1063-1089 CE) it is said that he had to face trouble with Brahmins who held solemn fasts (Råj., 7.177). When the hostilities between Ananta and his son Kala†a were ruining the country, large crowds of Brahmins held solemn fasts against both, so that they managed to bring them to a reconciliation (Råj., 7.400-401). - The next pårißad fast is a local one against King Harßa (1089-1101 CE) who is spoliating temples: Råj., 7.1088 sees the collected purohitas manage to be granted, in compensation for the spoliations of their temples, exemption from the forced carriage of loads (rüdhabhåro∂hi). This system of levying forced labour of this nature, a very heavy corvée later called bëgår, was resorted to by Kashmiri kings and administration as the only means available for military transport up to rather recent times.110 The parißad’s exemption from it thus represents a great victory for their town.

The strangest extension of the collective hunger strike resorted to for practical ends, however, was the soldiers’ protest, first met with in Råj., 7.1156-1157. Usually protest fasters were either single individuals or groups of Brahmins and purohitas. The protest of the army is a further step in the politicization of the conflict which opposed wronged citizens to the government.111 Here however, differently to the subsequent case of soldierly protest fasts,112 an increase in salary is only the apparent motivation, because the strike is actually instigated by a politician of dubious faith. This first protest in fact was staged by Sunna, a prefect of police, as the enemy, besieged in a mountain fort by the forces of King Harßa, had bribed him to create a diversion. The traitor Sunna enticed Harßa’s soldiers to claim a larger marching allowance, and as the treasury was far away this created immediate disarray in the army. Though often pråya was an act of truth, where an individual (usually a Brahmin) pledged his life to obtain justice, Kalhaña is quick at pointing out the abuses, and gives his negative opinion on those whom he names the professionals of fasting, pråyopave†aku†alas, Brahmins who harm rather than help their king. In Råj., 7.1611 he blames the 109 Råj., 6.339: tayå svarñapradånena sumanomantakådayaΔ Ù bråhmañåΔ samagr¢hyanta tataΔ pråyo nyavartata ÙÙ. 11o See Lawrence 1895: 413. 111 See, also, Naudou 1963: 223. 112 The next instance of such a military protest takes place under King Sussala, Råj., 8.808 (see infra).

V. Violating Tradition and its Boundaries

553

ministers of Harßa, who had shut their doors to him in his need: “Those (Brahmans) who are clever in solemn fasts (pråyopave†a), are thoroughly useless in the end. Vain is the reliance which kings put on wretched Brahmans”.113 As the eighth book of the Råj. contains the highest number of examples of collective fasts, undertaken mainly by Brahmins and pårißads, but also by soldiers and other groups of disaffected or harassed people, Kalhaña adds other negative comments. In Råj., 8.110 he evaluates the five different causes of ruin of ‡rœnagar through the mouth of one of his characters. A wise counsellor tells his king: As the town of Bhüte†a, which had been destroyed by a conflagration, has quickly recovered its [former] splendour by the power of your order, thus, O King Uccala, may you restore to happiness and comfort this, your own city, which has been destroyed by the five fires of Kåyasthas, [royal] relatives, [obnoxious] regulations (? klr¢pti),114 ministers and solemn fasts (pråyopave†a)!115

This is repeated in different words a few hundred verses later, when Råj., 8.709 says that, together with a number of other loafers, the people who delight in the king’s misfortunes are: “the Brahmans from the Purohita corporations (pårißadya) who are expert in arranging solemn fasts (pråyopave†a)” (pråyopave†akußalåΔ pårißadyadvijåtayaΔ). The occurrences of collective fasts listed below show the amplitude of the phenomenon, as well as the different attitudes of the fasters, from the Brahmins who acted in earnest to those corrupted by bribes, to the soldiers who used brutal force to achieve their goal. On social/political grounds, the most interesting notation of Kalhaña occurs perhaps in Råj., 8.905, when he describes the citizens who go to watch the solemn fasts of the parißads in order to discuss politics with the fasters (see infra). - A widely spread outbreak of collective pråya is recorded during the reign of Sussala (1112-1120 CE), when the ˘amaras had profited of the estrangement of Tilaka, the commander-in-chief, to damage the royal stores. Råj., 8.658 recounts: “Terrible scandal arose in every town 113 Råj., 7.1611: pråyopave†aku†alåΔ †aktås tv ante na kutracit Ù mithyåsa∫bhåvanåbhümir bhüpånå∫ brahmabandhavaΔ ÙÙ. 114 This word, like the kalpana used in Narm., 1.86, probably means ‘a tax, an extra tax’, such as the one the feudal lords had to pay the king, or simply ‘taxation’. See Baldissera 2005: note 88. 115 Råj., 8.110: bhüte†asya yathå purœ hutavahanluß™å tvadåj∞åbalåd bhüyaΔ svå∫ †riyamåsasåda sahaså tadvatsamaståm imam Ù tva∫ kåyasthakutumbiklr¢pti sacivapråyå [here there is a lacuna of one akßara] pa∞cån alœlœ∂ham uccaladeva nirvr¢tisukhasthityå purœ∫ svå∫ kriyåΔ ÙÙ. It is interesting to notice that the king to whom this invocation is addressed is that same Uccala who had vowed to commit suicide if any of his subjects should resort to pråya (Råj., 8.51).

554

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

owing to Brahmans, who were exasperated by these troubles, holding solemn fasts (pråya) and immolating themselves in fires”.116 - And soon in fact there was another protest, staged by the Brahmins of Råjånava™ikå (Råj., 8.768-777). Again these fasters stop the protest on receiving bribes (Råj., 8.777): “By offering bribes of gold he (Sussala) brought over the chief intriguers among them and thus with difficulty induced some to abandon the fast”.117 - Poor Sussala has to face yet another challenge of this type, but from a different quarter. As it had happened before (in the times of Harßa [Råj., 7.1156-1157]), trouble came from the army. This time the soldiers resorted to both pråya and a blockade in order to increase their salary (Råj., 8.808): “With drawn swords they blocked the doors in the royal palace, and everywhere held solemn fasts with a view to getting marching allowances”.118 This then seems to be a case of both obstruction and pråya. - But soon also the hereditary temple priests engage in another fast to obtain more riches (Råj., 8.811): “The temple-purohitas (sthånapåla), too, performed a solemn fast, and by besetting him with violence made him pound up golden vessels and other valuables for distribution”.119 - Even the next sovereign, Bhikßåcara (1120-1121 CE), is treated to a series of general fasts (Råj., 8.898-906). There were fasts of Brahmins holding agrahåras as well as of Brahmins who lived in different parts of the Capital, but none as spectacular as that of the parßads that met in the Gokula (Råj., 8.901-902): “Such an assembly of Purohita of sacred shrines (pårißadya) had never been seen before. The courtyard [of the Gokula] was thronged everywhere with rows of sacred images, which were placed on litters and embellished with glittering parasols, dresses and Chowries, and all quarters were kept in an uproar with the din of the big drums, cymbals and other [musical instruments]”.120 - And in a subsequent verse one can finally see the emotion induced in the crowds by the spectacular attraction of such a political display, as the citizens come to watch the fasts, and hold political discussions with their protagonists. (Råj., 8.905): “Upon what plans did this host of Purohita-corporation not debate day after day with the citizens who came to watch the solemn fasts?”.121 - In Kalhaña’s own times, during the reign of King Jayasiμha (1128-1149 CE), the Brahmins in a similar way tried to oust Sujji from the position of commander-in-chief (Råj., 8.2076). Later the Brahmins rose again (Råj., 8.2224), this time against the minister Citraratha, who had increased taxation, and as he did not satisfy them, some of the fasters 116 Råj., 8.658: åtaõkodvejittair vipraiΔ kr¢tapråyaiΔ pure pure Ù vahnau hutågnibhir ghorå kukœrtir udayat ÙÙ. 117 Råj., 8.777: kå∞canotkocadånena tanmadhyedhikacakrikåm Ù kåμ†cit svœkr¢tya sa pråya∫ kathaμcid vinyavœvarat ÙÙ. 118 Råj., 8.808: te kr¢ß™a†astrå dvåråñi rundhanto nr¢pamandire Ù pravåsavitte labdhavye pråya∫ cakruΔ padepade ÙÙ. 119 Råj., 8.811: sthånapålair api pråyakr¢dbhir åkramya dåpitaΔ Ù dhana∫ suvarñabhåñ∂ådi cürñœkr¢tya vi†r¢õkhalaiΔ ÙÙ. 120 Råj., 8.901-902: yugyårpitaiΔ sitacchattravastracåmara†obhibhiΔ Ù vibudhapratimåvr¢ñ∂aiΔ sarvata† chåditåõganaΔ ÙÙ901ÙÙ kåhalåkåμsyatålådœvådyakßobhitadiõmukhaΔ Ù adr¢ß™apürvo dadr¢†e pårißadyasamågamaΔ ÙÙ902ÙÙ. 121 Råj., 8.905: pråya∫ prekßitum åyayåtaiΔ pauraiΔ saha dine dine Ù amantrayat kåμ kå∫ na vyavasthå∫ parßadåμ gañaΔ ÙÙ.

V. Violating Tradition and its Boundaries

555

burned themselves alive (Råj., 8.2225). The heavy exactions continued, though, and the example of the fasters was followed also by a cow-herd (Råj., 8.2226): “When his [Citraratha’s] servants confiscated even the grazing land (caraka) of the sacred cows, one cow-herd also, overcome by compassion, burned himself”.122

It is interesting to notice that a similar cruelty to cows is expressed also in Kßemendra’s Narm., 1.26, in the description of the divira, the governement bureaucrat: Destroyer of the gods (or: ‘Stealing the property of the gods’) taxing even salt and hay for cows, day and night the officer (divira) gobbles and guzzles.123

Another verse (Narm., 1.57a) refers to the officer’s killing of Brahmins and of cows: Massacres of Brahmins do not worry him, so why should he care about slaughtering cows?124

In Narm., 1.121 the Kåyastha is said to even punish cows: How could a person who punished cows by tying them up until they died, have any consideration for human beings, who are naturally guilty (or: who are the receptacle of various kinds of offences?) of all plundering?125

Here a particular aspect of these collective fasts calls for consideration: by their fasting the Brahmins and purohitas express their anger against those whom they blame for their political, economical or moral behaviour. That the refusal to take food (especially when one was in a situation of dependence) was usually interpreted as an expression of anger126 is very explicit in Råj., 8.2690. While prince Bhoja is kept almost as a prisoner of minister Alaμkåracakra, defined ‘the robber, the slave’ (dasyur), this telling scene occurs: “As Bhoja 122 Råj., 8.2226: carake dharmadhenünåm uttabdhe’pi tadå†ritaiΔ Ù vahni∫ gopålako’py ekaΔ karuñyapravañovi†at ÙÙ. 123 devåpahåriñå tena goghåsalavañacchidå Ù bhujyate pœyate bhüri divireña divåni†am ÙÙ. 124 Narm., 1.57: brahmahatyå na gañyante govadheßu kathaivakå Ù. 125 Narm., 1.121: gavå∫ dañ∂åya ya† cakre nidhanavadhi bandhanam Ù kå nåma gañanå tasya nr¢ßu sarvåparådhißu ÙÙ. 126 And compare with the nervous anorexia induced by frustration and anger towards the family in some of our contemporary young adolescents.

556

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

was told by someone that the robber might take his abstention from food as [an indication of] anger and might change his conduct, he took his food”.127 The same situation however had occurred before, under Bhikßåcara (1120-1121 CE), after a crowd of purohitas from sacred places engaged in solemn fasts had been dispersed by the armed soldiers of Bhikßu (Råj., 8.940): “The few that [remained behind] guarding the empty litters of the divine [images], were not troubled by Bhikßu, as they declared that they had renounced their fast”.128 Here it is worth of notice that the king would have stopped the fast even at the expense of killing or harming a few Brahmins! - Other large concentrations of Brahmins engaged in pråya in order to influence royal diplomacy were held in order to obtain that King Jayasiμha campaign at once against the rebellious ˘amaras. That such a pressure was very difficult to shake, as that could have easily created internal troubles, is shown by the fact that finally, even against Jayasiμha’s better strategic judgment, the fasters obtained the king’s agreement (Råj., 8.2733-2734). - The power of the collected Brahmins, however, kept pushing this king’s policy (Råj., 8. 2737): “The king dismissed from his presence the minister Alaμkåra upon the petition of the Brahmans who had been excited against the latter by [other] ministers, and who obstinately persisted in their perfidious course”.129 And when in spite of this they continued fasting, like in the previous case, the king was only able to stop their protest by promising to follow their advice in political matters, as it had happened to other rulers. - As fasting had become such a customary weapon, minister Alaμkåra himself, though ousted by it, in order to obtain his way resorted to the same device of sitting in pråya in front of a bridge (Råj., 8.2895-2897), followed by his retinue (Råj., 8.2897): “When the troops of the Darad chief saw him preparing to die, together with his followers, who were mostly young men, they became alarmed and felt pity for him”.130 This is the last instance of pråya, or rather of the threat of pråya, in the Råj., this time resorted to by an important minister in order that the terms of a compact be respected.

From the records of Kalhåña it would seem that in most cases the kings were very sensitive to the pråyopave†a (and eventually to other types of protest-suicide) of their subjects, and that therefore this was felt as a dangerous but very efficacious means to control and eventu127 Råj., 8.2690: tvayi dasyur viparyasyen manyuμ jånann abhojanam Ù bhojas tatreti kenåpi kathite vyadhitå†anam ÙÙ. 128 Råj., 8.940: †ünyåni surayugyåni rakßantaΔ ke’pi bhikßuña Ù pråyån nivr¢ttå vayam ity uktavanto na bådhitå ÙÙ. 129 Råj., 8.2737: amåtyadattavaimatyaiΔ sva†å™hyama™harair atha Ù dvijair nißiddho’laμkåro mantrœ råj∞ojjhito’ntikåt ÙÙ. 130 Råj., 8.2897: mr¢tyaiΔ saha yuvapråyair vœkßya ta∫ martum udyatam Ù daråtura∫ daradråjasainya∫ taddainyam åyayau ÙÙ. Probably here a better translation would be: “[…] felt pity for them”.

V. Violating Tradition and its Boundaries

557

ally redirect the excessive power of kings.131 The Råj. shows only one instance where it almost did not work, until the suicides’ number rose to such an excessive figure that the king finally had to give in. When 99 people killed themselves (by drowning) in a single day, only at that point did King Jayåpœ∂a (Råj., 4.631-639) relent somewhat in his misappropriations. He stopped confiscating agrahåras, but did not actually give back the land he had taken from individuals (Råj., 4.639).132 Another person in power not fazed by fasts is the Kåyastha Citraratha (Råj., 8.2224-2226). In spite of frequent collective pråyopave†as against him and of the self-immolation by fire of a vast number of citizens, he kept taxing people very heavily, not stopping even at confiscating land used for the grazing of the sacred cows. A similar situation, where even cows did not escape the greed of the revenue officials, is sketched in the Narm., which anticipates Kalhaña’s picture by at least a hundred years.133 7. Suicides by despair that meet a tragic end in Kßemendra’s satires In Narm. and De†., as well as in other satirical works of Kßemendra, the evils wrought by a corrupt administration are shown at their worst. To protect himself, in the incipit of Narm.134 his villains are said to be the corrupt Kåyasthas of the past, who thanks to righteous King Ananta (1028-1063 CE) had been removed from office. In Narm., 1.4 Kßemendra writes: This discriminating One, after removing the troubles of his subjects reduced all corrupt officials to nothing but a memory.

This is obviously a literary as well as a diplomatic device, as the situation apparently did not undergo any dramatic change. Kalhaña in fact reports very similar vexations as occurring at the time of Kßemendra, and continuing for at least another hundred years. In the body of the satires, though, Kßemendra insists especially on showing the evil wrought by the general corruption of the administration, while leaving at the very end of Narm. an instance of its partial redressing by the king, with the exemplary punishment of some of the main culprits. 131

See also the considerations on Kashmirian kingly authority of Naudou 1963: 217. See supra, § 4.2. 133 See supra, § 6.1.2. 134 Narm., 1.4: tena prajopasargeßu våriteßu vivekinå Ù durniyogißu [sarveßu] nœteßu smr¢ti†eßatåm ÙÙ. The manuscript presented a lacuna with three dots, emended in sarveßu already in the 1923 edition. 132

558

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Satire presupposes a public consensus on what is to be made fun of, and the Narm.’s verses on pråya and pråyastha are a case in point. They depict an actual contemporary situation in which the depraved behaviour of the all-powerful Kåyasthas, the government officials, is held responsible of the frequent protests of the citizens, voiced by the collective hunger strikes of the Brahmins and purohitas. The ridiculous and undignified end of the main character expresses what was in all probability popular consensus on the treatment people would have liked to allott to the actual living models, outside the literary fiction. By treating a serious matter in a highly sarcastic manner, four consecutive verses in Narm. allude to a collective pråya in a temple courtyard. The passage is one of the most outrageous episodes in Kßemendra’s satires, showing the baseness and hypocrisy of the corrupt officers at their worst. Here the Kåyastha in charge goes to the temple in order to make a loud public display of his piety by reciting a famous ‡aiva hymn, the Stavacintåmañi of Naråyañabha™™a. At the same time, though, the important bureaucrat intersperses each half verse of the beautiful prayer with impious commands to his henchmen regarding the treatment to be reserved to the pråyasthas. Even the formal aspect of these interspersed interpolations is grotesque, because each half verse of the Kåyastha’s own commands to his assistants is conceived in the shape of the known literary technique of †lokapürña, “completing a given hemistich by a new half of one’ own invention”, and completes the initial part of the verses in the most unlikely and incompatible manner. As Narm. had not yet been translated in a western language,135 it seemed best to quote the entire passage (Narm., 1.39-44), in order to show the shocking contrast between the high concepts of the hymn (shown in italics), and the cruelty of the Kåyastha’s (possibly whispered) injunctions. In the first hemistichs of the last two verses, however, the Kåyastha recites part of another hymn (Narm., 1.43a, 44a), which I have not yet been able to trace, intermingling it with general considerations on how better to extract profit from his harassed subjects. Oμ! Thanks to Pa†yantœ of beautiful words, who captivates the mind as soon as she is seen –how many fasters unto death did I offer in the temple of Vijaye†vara yesterday? His infinite greatness136 shines forth. Glory to Parame†vara! 135

There is now a first translation in Baldissera 2005. Here ‘His infinite greatness’ translates the reading found in the manuscript (ullasitånantamahimå), whereas the customary reading is ullasitånandamahimå, ‘His greatness made of bliss’. Here it would seem that Kßemendra paid a compliment to King Ananta by this variant reading. 136

V. Violating Tradition and its Boundaries

559

–let these seventy three people be added to those who are already there. To Him whose unfolding is prosperity, compassion, understanding, supreme bliss, efficiency, –let those learned Brahmins (bha™™as) who died in the pråyasthåna, be dragged out by the chains on their ankles. And to Him whose majesty is resplendent with knowledge, may there be glory to him, to Aparåjita! –let those officers (niyogins) go to the villages, and cause dark smoke! To him whose nature is absolute bliss, who is the cause of all auspicious things –those who defy punishment should be killed, and all their wealth be confiscated! To Him who removes all suffering, to that Brahman who is pure consciousness, salutation! –Only after they are squeezed (or ‘after they are tortured’) do the subjects exude fat (or ‘exude wealth’) like the seed of guggulu.137

In the next chapter the main protagonist of Narm., the niyogin, receives a letter from his assistant, who describes the vexations done in his name in his country seat. Here are mentioned other people engaged in pråya, and the way they are dealt with (Narm., 2.96-98): On account of a pipe for ghee138 a man has been arrested and taken into custody; so this insolent Brahmin went without food and died. I have laid before your feet the document [listing] his possessions, have taken his wife into custody and have had his whole house sealed. Those fasting unto death whom the paripålaka presented to me for the sake of the monthly skandhaka139 137 Narm., 1.39-43: oμ sugirå cittahåriñyåpa†yantyå dr¢†yamånayå Ù –hyaΔ kiyanto mayå dattåΔ pråyasthå vijaye†vare ÙÙ39ÙÙ jayaty ullåsitånantamahimå parame†varaΔ Ù –ådau sthitånåm upari pråyån tv ete trisaptatiΔ ÙÙ40ÙÙ –yaΔ sphœtaΔ †rœdayåbodhapramånandasampadå Ù –pråyaståne mr¢tå bha™™åΔ kr¢ßyantåμ gulphadåmabhiΔ ÙÙ41ÙÙ vidyoddyotitamåhåtmyaΔ sa jayaty aparåjitaΔ Ù […] [corrupt reading ‘nirdhåm’] adhümakartåro gråmån yåntu niyoginaΔ ÙÙ42ÙÙ sarvånandarüpåya sarvamaõgalyahetave Ù –sarvasvaharañaμ kr¢två vadhyå dañ∂anißedhinaΔ ÙÙ43ÙÙ sarvakle†åpahartre ca cidrüpabrahmañe namaΔ Ù –pœ∂itåΔ prasravanty eva prajå guggulubœjavat ÙÙ44ÙÙ. 138 Probably gone missing, or lent and not returned. ghr¢tanå∂œ seems to be a pipe for feeding ghee to cows. 139 This seems to be the monthly installment of a tax.

560

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

have been placed in the house of that [dead Brahmin].140

These are only some of the ways in which the government officials become rich at the expenses of the vexed citizens and especially Brahmins. Exposing the Kåyasthas as they insult and harass their source of misappropriated riches is a perfect way to illustrate the corruption and degradation of these officers. 7.1. Kßemendra’s covert criticism of the professionals of fasting In another set of Narm.’s verses, as well as in a couplet from the other satire mentioned here, the De†opade†a, there seems to be a further development. Kßemendra was a sophisticated writer, alive to contemporary events and customs, and it seems that also the abuses of some of the professionals of pråya did not escape his satirical notation. So far the blame was on the side of the bureaucrats, but in some of the these verses there seems to be a double intent. These episodes are not only used to show the inhuman behaviour of the Kåyasthas, who, not content of having reduced good people to such extremes of despair, mistreat even their corpses, but by making light of the pråyasthas themselves they might allude to the abuses of recourse to pråya. When Kßemendra compares the medical doctor flurried and exhausted, full of the insignia of his learning, who rushes from a patient to another, to people in pråya arriving in front of Citragupta, we can picture these sweaty and exhausted pråyasthas as equally full of their religious insignia. This is exactly how the Råj. shall describe the professionals of great collective fasting about a hundred years later. Here Narm. 2. 69-70 says: Rushing in a hurry to hundreds of houses, sighing as if he were carrying a great weight, frequently wiping with his hand the drops of sweat from his forehead, And carrying a strip of cloth on which were collected his appointments and prescriptions, [the vaidya] arrived, like someone who is fasting to death appears before the attendant of Kr¢tånta.141 140 Narm., 2.96-98: ghr¢tanå∂œnimittena ya† cåsau bandhane dhr¢taΔ Ù so’pi vipro niråhåras tikßñaΔ pa∞catvam ågataΔ ÙÙ96ÙÙ pådåntike ca prahita∫ tasya pradhanapatrakam Ù baddha∫ mayå tatkalatra∫ mudrita∫ sakalaμ gr¢ham ÙÙ97ÙÙ yån skandhakanimittena pråyasthån paripålakaΔÙ pradadau måsavr¢ttyaiva te mayå tadgr¢he dhr¢tåΔ ÙÙ98ÙÙ. 141 Narm., 2.69-70: bhrånto gr¢ha†ata∫ türña∫ bhåråkrånta ivocchvasan Ù lalå™asvedasalila∫ påñinå vikßipan muhuΔ ÙÙ69ÙÙ vahann außadhasaõketanåmasa∫yogacœrikåm Ù kr¢tåntådhikr¢tasyågrådyaΔ pråyastha ivågataΔ ÙÙ70ÙÙ.

V. Violating Tradition and its Boundaries

561

In a similar vein is also a verse of De†., (in the sixth Upade†a, dedicated to the teachers and students inmates of a ma™ha) that describes a chattra142 from Gau∂a who lives in a ma™ha, a religious college. Although the aspect and behaviour of the student is the same as that of the pråya professionals, the reason for his perpetual fasts is his addiction to gambling. The chattra, among other evil pursues, is “constantly engaged in fasting unto death” (pråyasadå), but only because he keeps loosing at dice. The figure of such a ‘religious’ student is then further ridiculed in the subsequent verses (De†., 6.31-32): Attached to courtesans, addicted to dice, a cåkrika,143 always on hunger strike, with a pained belly, the student is an ascetic with five fires in that forest which is his monastery. This is not a religious student, nor a householder, a forest-dwelling sage nor a renouncer: this is a fifth stage of life, that of the students, called ‘endowed with the five auspicious things’.144

These satires, written at least a hundred years before the Råjataraõgiñœ, are another proof that in contemporary Kashmir, because of the bad administration, resort to pråya and pråyopave†a was very common. There is also, albeit expressed in jocular terms, the notion that the person engaged in pråya is comparable to an ascetic doing tapas with five fires (though here the ‘fires’ are prostitutes, gambling, […]). Narm. already testifies to the existence of collective groups of pråyasthas sitting in temple courtyards. The mention of the pråyasthas in the letter of the Kåyastha’s attendant implies also the presence of a group of fasters in the country. What is remarkable is that on one occasion Kßemendra gives what appears to be a real figure, when to an indefinite, presumably large number of people sitting in pråya in a temple’s courtyard is added another group of 73 people (Narm., 1.40). This is in fact rather precise, compared to the vague accounts of the masses of purohitas from the parßads mentioned in several passages of the Råj., and gives the impression of coming from actual observation. 142 From the manuscript it is difficult to establish if the actual word is chåttra (‘sheltered, pupil’), or chattra (‘the one who gives shelter, teacher’), because the two spellings seem to have been used at random, and obviously metrical considerations here do not help. Kaul took it to mean the ‘student’, both in Narm. and in De†., whereas I prefer to read it as ‘teacher’ in Narm., and ‘student’ in De†. 143 This problematic term in a Kashmirian context seems to mean ‘an intriguer’ (the other meanings of the term do not seem to apply, here), but probably because the cåkrika belonged to a particular circle of people, cakra. This could be intended as a group which met for religious purposes, like doing cakrapüjå, or a political clique. In both cases, the members would be under an oath of secrecy. See also supra (§ 6.1.2), Råj., 8.777, for the ‘intriguers’ among the fasters. 144 De†., 6.31-32: ve†yåsakto dyütakara† cåkrikaΔ pråyakr¢t sadå Ù kuk†ibhedœ ma™havane chåttraΔ paõcatapå muniΔ ÙÙ31ÙÙ na brahmacårœ na gr¢hœ na vanastho na vå yatiΔÙ pa∞camaΔ pa∞cabhadråkhya† chåttråñåm ayam å†ramaΔ ÙÙ32ÙÙ.

562

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

8. Gandhi’s experience: pråyopave†a interpreted as the Satyagraha campaigns in the non cooperation program Gandhi started campaigning against legislative discrimination while working as a lawyer in South Africa, with a passive resistance program which he named satyagraha, a Sanskrit term meaning ‘adherence to truth’. On his return to India, one of his first campaigns against English domination was known as ‘The Rowlatt Satyagraha’, from the name of the judge who proclaimed the emergency legislation. In 1930 Gandhi launched the civil disobedience campaign against the British monopoly on salt: Gandhi and a group of satyagrahins, braving the law, marched for several days in order to go and collect salt from the sea, and their example was soon followed by thousands of people throughout India. In 1931, again on his return to India after attending the second conference of the Round Table in London as the official representative of the Congress movement, Gandhi was imprisoned. As a form of protest against the institution of separate electoral colleges for the untouchables Gandhi announced that he would undertake a fast to the death and gained enormous popular support. Released, and again brought to jail, in 1942 he undertook anew a severe fast against British accusations, and again he survived, until in 1944 he was freed on health grounds. Gandhi used the cultural (religious) prestige of fasting and of the fasting ascetics and heroes, as well as the scope and efficaciousness of pråya practices in manipulating the political decisions of people in power. His informed use of the media allowed his personal satyagraha action to become a huge public statement. In particular, the political value of his fast unto the death in jail had illustrious precedents, found in Råmåyaña and in Råj., for instance with the mythical pråyas of Aõgada (see supra, § 2.3) and the historical ones of Harßa and Bhoja (§ 6.2 a). This practice has been followed even recently in several countries by minority groups and political prisoners.145 The resort to pråya to redress a wrong would seem to be precisely one of those cases in which the individual uses of the greatest freedom towards the institutions. Formally, it only depends on him or her whether they want to enter upon pråya. These individuals give, or abandon, their most precious possession, their life, so their formal declaration of intents (sa∫kalpa), is –or should be– at the same time a declaration of truth, satyagraha. This is the only way they deem adequate to oppose 145 An interesting case is that of the hunger strikes of I.R.A. detainees in English jails. These, however, need not be copied on the model of Gandhi, as they could stem from an ancient Irish custom. According to Maine (1876: 280, 297), the old Irish code of law had already a similar provision, that states that a creditor may distrain a debtor, but in the case of a debtor of the chieftain grade, the creditor should also fast.

V. Violating Tradition and its Boundaries

563

the violence of injustice. It is debatable, in the light of what we know of the social background where these ideas took shape, whether it was also a truly non violent form of protest. The prestige of the pråyastha’s stand comes from its having been first used by ascetics and Brahmins, whom it was forbidden and very inauspicious to kill. The culprit of such a crime would have been held responsible for any evil that might have befallen the community, so fear of a negative reward in case of death of the faster would apply also to instances of fasts undertaken by people who did not necessarily belong to the ascetic or Brahmanical class. By its ritual aspect and demeanour –sitting in silent concentration on darbha, without touching food nor water, after a formal declaration146 in which one touches water (as a witness)– it amounts to a sacred action, a sort of sacred oath. It certainly is a kind of moral suasion, and often it has entailed a form of blackmail. This aspect is particularly obvious when it is simultaneously engaged in by large groups of people, as found in verses of the Narm. or in the Råj., but it stands out quite clearly even in the first account found so far in the Kaußœtakyopañißad, which is older than law or epic, and concerns a single faster. Obviously people were in awe of the fasters, and at times felt protected by them, especially when they upheld justice. What is remarkable is that even when the protest was prompted by shady political motives, or was performed merely to obtain one’s own personal advantage, the manner of proceeding in the popular imagination was still assimilated to the holy death in restraint of the pious ascetic. Kalhaña could blame the abuses of the professionals of pråya, but common people would still have believed in the power of the formal display of courage in front of death, and would have both admired the fasters, and feared their potential danger. It seems probable that this belief in the magical power of pråya has always been one of the first causes of the popular prestige it enjoyed. 8.1. Frightening ascetic heat, popular uprising and deification of violent deaths From the notion of religious prestige in these terms one easily passes to the notion of danger, threat, as it is well known how dangerous ascetics’ curses can be. But given the popular support and admi146 Such declaration occurs already in Mahåbhårata, 1.2.304, when it is said of Draupadœ: “what time she, resolved upon not eating, besieged her husbands” (kr¢tåna†anasaμkalpå yatra bhart≤n upåvi†at [transl. by Hopkins 1900: 154). This ‘resolve’ (saμkalpa), in later times becomes both an oral formal declaration of intent, important for any type of ritual, as well as a written one, according to Naråyañabha™™a’s Tristhalœsethu (sixteenth century). In Kale’s translation (1930-1962, vol. 4: 610): “A man desirous of committing religious suicide at Prayåga should first perform a pråya†citta, should perform his own †råddha up to the offering of piñ∂a (if he has no relatives), one day of fast, one day with a saμkalpa preceded by writing, and enter the holy waters contemplating on Vißñu”.

564

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

ration for ‘solemn fasters’, people in power who oppose these ‘heroes’ may face more than an automatic reward for their evil actions in a future life. There is an even more threatening impending danger, for popular uprising and rebellions are in store for the impious king or functionary who allows the pråyopave†in to die. From the religious point of view, self sacrifice, in spite of its prohibition in most legal texts, results in a great accretion of prestige. According to Michaels (1990: 31): In Indian folk religion, on the other hand, it is the concept of violent death, including religiously motivated suicide, that can lead to deification, while in the Brahmanical traditions any unnatural death is more connected with ghost worship (bhüta, preta, etc.).147

At times pråyopave†ins became local gods. Crooke reports a case of dharña from 1427 CE which transformed the Brahmin who had fasted to death into a god still worshipped in that author’s times in a Bengali village. Harßu Pånre, also called Harßu Baba, was the Brahmin family priest of Råja Sålivåhana of Chayanpur, near Sahsaram in Bengal. Due to the envy of one of the queens Harshu Baba had his beautiful house demolished by the king and his estate withdrawn. The angry Brahmin did dharña at the door of the king’s palace and fasted until he died. The Råja’s family was destroyed except for one daughter who had been kind to the Bråhman in his misfortune, and through her the family continues to this day. Harßu is now worshipped with the fire sacrifice and offerings of brahmanical cords and sweetmeats.148

Stories of women who become devœ through violent death (and not necessarily suicide) are also frequent throughout India. In medieval Central Orissa, in tribal areas, many legends for instance recall that new reigns were founded when a hero, the future king, would ask a person met by chance –who often happened to be a pregnant ‡abara woman– if she would consent to be beheaded by him. The hero promises her that from then on he would have her worshipped as his iß™adevatå.149 In later centuries it is well known that satœs became deified, down to the last known episode of Roop Kanwar, who after ascending the pyre in 1987 has now a shrine dedicated to her in her village, Deorala near Jaipur, in Rajasthan, which has become a place of pilgrimage.150 147

Here Michaels provides also the following reference: Blackburn 1985: 255-274. See Crooke 1894: 121-122 (who quotes three sources: Cunningham 1879: 160 ff; Buchanan 1833: vol. 1, 488; North Indian Notes and Queries, vol. 2: 38). 149 See Kulke 1984: 12-24. See also Baldissera forthcoming. 150 For evaluation of the satœ of Roop Kanwar, see, for instance, Michaels 1990: notes 4 and 5; Leslie 1992: note 5; Leslie 1991: 8-11. 148

V. Violating Tradition and its Boundaries

565

8.2. pråya, sallekhana, satyagraha: playing with words Perhaps it is worth mentioning that if one now looks at Internet under pråya and pråyopave†a one finds the eulogies of contemporary Jain sallekhana, the final fast unto death of Jain monks, nuns or simple devotees. It is somehow strange that only this acception of the word has become prevalent in cyberspace, because pråyopave†a already from ancient times has often been interpreted as, prevalently over other meanings, a demonstration of anger against injustice, rather than of the ascetic calm required by the Jain renouncer who seeks a ‘death in restraint’. A pious Jain –usually a monk, but recently also a growing number of lay people– wants to die in restraint, sa∫yama. In their prayer, pious Jains sometimes wish for death in meditation, samåhimarañam. The holy death, sallekhanå, is a final ritual, a gradual fasting under the supervision of ascetic teachers. The meaning of sallekhanå is ‘properly thinning out [both passions and one’s body]’. Traditionally there are four occasions for it. The first is unavoidable calamity (so that one cannot keep one’s vows), upasarga; the second is a great famine, durbhikßa; the third is old age, jarå; the fourth is a terminal illness, niΔpratikårå rujå. These prerequisites are in fact similar to the reasons that make voluntary death possible for ascetics in the dharma†åstras (see supra, § 2.1.). A Jain layman has to make a formal declaration to a teacher, asking to be instructed: ‘I have come to seek sallekhanå’. The act is public, and there is renunciation to all possessions and associations. The fact that in contemporary Internet language the term pråya has come to mean sallekhanå shows perhaps how in history words are constantly shifting their meaning, but at times experience a return to their initial source, which in this case really seems to be the progressive ceasing from activity and life processes of the old ascetic. Terminology is always historically determined, and a certain shade of meaning may have become prevalent in a particular period. Gandhi’s struggle provides an interesting example of this shift. Gandhi made use, even in his outlook and demeanour, of the ascetic/heroic ritual prestige pråyopave†a had represented for centuries, in a deliberate attempt to capture popular favour. He however acted upon the popular religious feelings inspired by this practice and attitude and turned them into an active, heroic movement of resistance to foreign occupation. Maintaining that particular ritual stance, which he renamed satyagraha, he fought also in opposition to discrimination on the base of birth right, while at the same time enjoying the relative impunity allotted to those who had first invented both the privileges of birth as the practice of pråya. He used pråya in fact not only to blackmail the foreign invaders in order to achieve his political ends, but also against ideas and behaviour of the Brahmanically oriented social traditions of India which he deemed unjust. Gandhi’s behaviour in his late Indian years seems to have

566

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

entwined together several different strands, all of which originate in the double prestige of the ascetic and of the hero. There was courage in his challenges to an extraordinarily powerful authority, restraint and chastity similar to that of the brahmacårin and of the satœ in his daily behaviour in his ashram, and generosity in his political stand towards weaker groups. His pråyopave†a in a wider community that comprised ideally the whole of India was immediately perceived as an act intended to help others. By his ritual silence, fast and restraint he was easily compared to the ascetics of the past, whose rightful behaviour was actively appreciated by a very resonant public of supporters.

BIBLIOGRAPHY Sources and Translations Arrien (2002), Le voyage en Inde d’Alexandre le Grand, Traductions et commentaires de Pascal Charvet et Fabrizia Baldissera, NiL Éditions, Paris. Bha™™anaråyaña (1908), Stavacintåmañi, in The Stavacintåmañi of Bha™™anaråyaña with Commentary by Kßemaråja, KSS n. 10, Srinagar. Bhler, G., (1886), The Laws of Manu, The Sacred Books of the East, n. 25, Dover Publications, New York, (repr. [1969]). Jhå, G.N. (1921), The Laws of Manu with the Båßya of Medhåtithi, University of Calcutta, Calcutta. Kangle, R.P. (1960-1965), Kau™ilya Artha†åstra, Bombay (repr. Delhi [1986]). Kßemendra (1923), De†opade†a, in Id., De†opade†a and Narmamålå, ed. by Kaul, M., KSS n. 40, Poona. Kßemendra (1923), Narmamålå, in Id., De†opade†a and Narmamålå, ed. by Kaul, M., KSS n. 40, Poona. Månavadharma†åstra (1983), ed. by Shastri, J.L., Manusm®ti, with the Sanskrit Commentary Manvårtha-Muktåvali of Kullüka Bha™™a, Delhi. Månavadharma†åstra (2005), ed. by Olivelle, P., Manu’s Code of Law. A Critical Edition and Translation of the Månava-Dharma†åstra, Oxford University Press, New York. Somadeva (1993), L’oceano dei fiumi dei racconti. Kathåsaritsågara, (transl. by F. Baldissera, V. Mazzarino, M.P. Vivanti), Einaudi, Torino. Somadevabha™™a (1977), Kathåsaritsågara, ed. by Pañ∂it Durgåprasåd, Kå†inåth Påñ∂uraõg Parab, Bombay [1889] (revised by Shåstri Pañsikar V.L., Motilal Banarsidass, Delhi [1915]).

V. Violating Tradition and its Boundaries

567

Stein, M.A. (ed.) (1989), Kalhana’s Rajatarangini or Chronicle of the Kings of Kashmir, Motilal Banarsidass, Delhi, 2 voll. (repr.). Studies Baldissera, F. (1996), Cañ∂ikå/Cañ∂œ, Vindhyavåsinœ and Other Terrific Goddesses in the Kathåsaritsågara, in Michaels A., Vogelsanger C., Wilke A., Wild Goddesses in India and Nepal, Studia Religiosa Helvetica Jahrbuch, Peter Lang, Bern-BerlinFrankfurt a. M.-New York-Paris-Wien. ______ (2005), The Narmamåla of Kßemendra (Critical Edition, Study and Translation), South Asian Studies, Ergon Verlag, Heidelberg. ______ (forthcoming), Le armi della Devœ, in Dolcini D., (ed.), Armi e battaglie da Rudra a Gandhi, University of Milan, Milano. Biardeau, M., Malamoud Ch., (1976), Le sacrifice dans l’Inde ancienne, Bibliothèque de l’École des Hautes Études Sciences Religieuses, vol. 79, Presses Universitaires de France, Paris. Blackburn, St. (1985), Death and Deification: Folk Cults in Hinduism, in «History of Religions», 24. Buchanan, F.H. (1833), Eastern India, London, 3 voll. Crooke, W. (1894), The worship of the sainted dead, in Id., An Introduction to the Popular Religion and Folklore of Northern India, Government Press, North Western Provinces and Oudh, Allahabad. Cunningham, A. (1878), «Archeological Reports», XVII. Falk, H. (2001), Suicidal Self-Scorching in Ancient India, in Karttunen, K., Koskikallio, P. (eds.), Vidyårñavavandanam, Essays in Honour of Asko Parpola, ‘Studia Orientalia’, n. 94, Helsinki, pp. 131-146. Féer, L. (1887), Etudes Bouddiques, Le Suttam d’Upåli, in «Journal Asiatique», tome IX. Féer, L. (1888), Etudes Bouddiques, Le commentaire de l’Upåli-suttam, in «Journal Asiatique», tome XII. Fezas, J. (2000), Du bon usage des textes normatifs: l’abolition de la peine de mort au Népal en 1931, in Brazer-Billoret M-L., Fezas J. (eds.), La norme et son application dans le monde indien, EFEO, Paris. Garzilli, E. (1997), First Greek and Latin Documents on Sahagamana and Some Connected Problems, Parts 1 and 2, in «Indo-Iranian Journal», 40. Hardy, F. (1995), The Religious Culture of India, Cambridge University Press, Cambridge-New Delhi. Heestermann, J.C. (1985), The Inner Conflict of Tradition: Essais in Indian Ritual, Kingship, and Society, Chicago University Press, Chicago. Hopkins, E.W. (1894), The Religions of India, Ginn and Company, Boston.

568

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

______ (1900), On the Hindu Custom of Dying to Redress a Grievance, in «Journal of the American Oriental Society». Karttunen, K. (1989), India in Early Greek Literature, ‘Studia Orientalia’, n. 65, Helsinki. ______ (1997), India and the Hellenistic World, ‘Studia Orientalia’, n. 83, Helsinki. Kulke, H. (1984), Tribal Deities at Princely Courts: the Feudatory Råjas of Central Orissa and their Tutelary Deities (iß™adevatås), in Mahapatra S. (ed.), Folk Ways in Religion: Gods, Spirits and Men, Cuttack. Lawrence, W.R., (1895), The Valley of Kashmir, Oxford University Press, London. Leslie, J. (1991), Suttee or Satœ: Victim or Victor, in Id. (ed.), Roles and Rituals for Hindu Women, London. ______ (1992), A Problem of Choice: The Heroic Satœ or the Widow-Ascetic, in Id. (ed.), Rules and Remedies in Classical Indian Law, Brill, Leiden. Lingat, R. (1967), Le sources du droit dans le sistême traditionnel de l’Inde, Mouton & Co, Paris. Maine, H.J.S. (1876), Lectures on the Early History of Institutions, J. Murray, London. Michaels, A. (1990), Widow Burning in Nepal, in Toffin, G. (ed.), Nepal, Past and Present, CNRS Éditions, Paris. ______ (1992) Recht auf Leben, Tötung und Selbsttötung in Indien, in Mensen B. (Hrsg.), Recht auf Leben - Recht auf Töten, ein Kulturvergleich, Steyler Verlag, Nettetal (Vortragsreihe der Akademie Völker und Kulturen St. Augustin, 15), pp. 95-124. ______ (forthcoming), The Price of Purity. Edition and Translation of the 89th Chapter of the Nepalese Mulukœ Ain of 1854. Naudou, J. (1963), L’autorité royale au Ka†mir médiéval, in «Journal Asiatique», 2, pp. 217-227. Renou, L. (1943-1945), Le jeune du créancier dans l’Inde ancienne, in «Journal Asiatique». Sax, W. (1992), Pilgrimage unto Death, in Veitch, J. (ed.), To Strive and not to Yield, Essays in Honour of Colin Brown, Victoria University of Wellington, Wellington. Thakur, U. (1963), The History of Suicide in India. An Introduction, Munshiram Manoharlal, Nai Sarak-Delhi.

VI. Thinking about Traditions in South Asia Today

CHRISTOPH EMMRICH When Two Strong Men Stand Face to Face. The Indologist, the Pandit and the Re-Making of the Jaina Scholarly Tradition 1 ‘But there is neither East nor West, Border, nor Breed, nor Birth, When two strong men stand face to face, though they come from the ends of the earth!’ (Kipling 1990: 187)

Jainism is paradigmatic of one variant in the Orientalist approach to Indian traditions, whereby that tradition is neglected both as an object of research and in its reception. In addressing the question of why this should be the case with regard to Jainism it might be necessary to extensively analyse the emergence of certain prejudices which mark the lines along which the Jaina tradition has been represented in western Indological discourse, i.e. its ‘dependence on Buddhism’, the ‘rejection of the body’, its ‘hybridity’ and its ‘minority status’ to name a few. One of the most striking examples of the Orientalist perception of Jaina tradition is Louis Renou’s famous reference to it as “Buddhism’s darker reflection” (Renou 1953: 111). In this conception of 1953 of Renou’s, 70 years after Albrecht Weber’s account of the literature of the Jainas –epochal in being the first such work of its kind–, we can identify three elements pertaining to the early conceptions of Jainism: its dependence on Buddhism, its extreme asceticism which was perceived negatively and, finally, the view that it was both baffling and meaningless.2 More modestly, Arthur Llewlleyn Basham in his book about the Åjœvikas, evaluated the significance of research on Jainism, inappropriately as it happens, in terms of the critical reception of Buddhism, saying “The history of Jainism, though it has 1 The precursor of this article is an earlier version published in German in Schalk 2004: 357-375. 2 See Weber 1883; Weber 1885.

572

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

much of interest to the specialist, is less spectacular than that of Buddhism” (Basham 1951: 261). Margaret Stevenson’s book, The Heart of Jainism –influenced by Christian missionary views–, which came out in 1915, was the first western publication on Jainism which reached a wider public both in Europe and in India.3 This work speaks, long before the formulation of the above-mentioned conclusions, of […] a highly resilient prejudice according to which Jainism is seen as grey and unappealing, as austere as its followers who are themselves negligible in number and therefore in interest, its tenets less profound than those of Buddhism, its mythology less spectacular than that of Hinduism. (Stevenson 1915: 26)

Despite these relatively sensitive observations, Stevenson’s book has come to be regarded in India as the most virulent attack on Jainism till today. Particular attention was paid to the fact that the book seemed to convey a contradictory message. On the one hand, Jainism was positively represented as an unknown, ritually diverse, morally consistent and philosophically challenging tradition while, on the other, conclusions were drawn from all this that the heart of Jainism was ‘empty’ in that it did not contain anything comparable to the Christian gospel or an understanding of the grace of God (Stevenson 1915: 289). This contradictory attitude can be categorized as a typical Orientalist gesture by which, positive and negatively loaded traditions as examples of arrested development are constructed and presented to both baffle and instruct (Said 1978: 206). A vehement reaction on the part of the Jainas, who by the time Stevenson’s spectacular book was published had already put behind themselves the first phase of their European-Indian acculturation, was articulated in the polemic of the theosophically influenced Digambara lawyer Jagmanderlål Jaini. Jaini rejected the book as unscientific and damaging (Jaini 1925).4 The extreme stance of Jainism towards asceticism and towards non-violence and the relationship, in turn, between these stances, on the one hand, and society and ritual, on the other, requires theoretical approaches to the body and action which have emerged in the European context only in the last decades. The situation is similar with regard to Jaina philosophy whose meaning is less to be sought in the content of individual doctrines and more in meta-theoretical and compilatory achievements. Finally, it is only recently that a strong interest has emerged in the composite character of the Jaina tradition, 3 Weber 1988: 203 regards Stevenson’s book as “really worthwhile” and cites it repeatedly (213f). 4 On the Christian overtones of the work it was said, “Such an attitude cannot but be disastrous to unbiased scholarship and is an absolute bar to the proper study of an alien religion”, and, above all, “It is not Jainism, but a gross travesty of Jainism” (Jaini 1925: 1f). Jaini’s own Outlines of Jainism, which appeared in the same year as Stevenson’s book, published by F.W. Thomas (Jaini 1915), was considered the most important modern defence of Jainism.

VI. Thinking about Traditions in South Asia Today

573

as evolving out of both ascetic and brahmanical origins. Hence, attention has now been directed towards the practices of dissemination, of the alternative forms of conservatism and transformation, of the closeness and openness of smaller traditions within Jainism. The increasing interest in Jainism as a living tradition of India with an unbroken cultural and social history has finally enabled western Jaina scholarship to emerge out of the towering shadow of research on Buddhism. Nevertheless, even as Orientalist views towards Jainism have become rare in contemporary research the positions which presuppose such views have decisively influenced the history of research on it. They have led to deficits in the collation of text materials, to a self-imposed, entrenched narrowness of inquiry and, finally, they have resulted in the widening of an existent divergence between western and Indian scholarship on Jainism and, within India itself, between scientific publication activity and the practical context of the text. In the following sections these developments are examined in the context of two epochal events in the history of Jaina research and in the invention and reinvention of a scholarly tradition in both the West and India itself: firstly, the constitution of the object of research called the ‘Text’ and secondly, the usage of such a constituted ‘Text’ for political purposes. On both sides and in both enterprises, there have been powerful forces at work, the philological Orientalist enterprise, represented by the German Indologist on the one hand and South Asian modernist enterprise represented by the Jaina pandit on the other, at the same time constituting and reshaping their respective scholarly traditions as well as the other’s. It will become apparent that by paying attention to the practice of constructing an ‘Eastern’ and a ‘Western’ tradition the boundaries and differences start to dissolve revealing instead individual agents, their goals and strategies, just as in the above quoted verses of Kipling’s ballad. 1. The Indologist and His Canon. Most of the problems of contemporary Jaina research are not new but stem from a process which not only perpetuates the deficits referred to but actually enforces them. Padmanabh Jaini’s critical analysis of the history of research on Jainism in the West (Jaini P.S. 2000a: 24-36), which remains till today a primarily European tradition, refers to some of the important issues.5 European enquiry into Jainism is not primarily about to this tradition as such but instrumentalises it in order to raise other questions. Hence, the enquiry is really about the Jaina doxography on other traditions such as the Åjœvikas, the Vai†eßika or the Lokåyata (Chattopadyaya 1959) or attention is directed to Jaina historiography (Jaini P.S. 2000a: 26), or the focus is on 5 “[…] The few and isolated cases of American scholarship can be safely viewed as extensions of European scholarship” (Folkert 1993: 25).

574

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Buddhism. The minimal contact between European Indologists and Jains has been confined largely to the acquisition of manuscripts or to philological cooperation, directed towards the classical texts. Generally, there was no knowledge of the literature in modern Indian languages. The English writings of Hiralål Jain, Jagmanderlål Jain and Adhinåtha Neminåtha Upådhye had been reviewed (Jaini P.S. 2000a: 24).6 In contrast, it is authors who wrote in Hindœ, Gujaråtœ, Kaññadå and Mårå™hœ7 who have had and continue to have the most important influence on the Jaina community (Sangave 1980: 291f). An important critique, which Jaini first pointed out, was the neglect of the Digambaras in favour of the ‡vetåmbaras. Western Jaina research remains primarily ‡vetåmbara research, inasmuch as the claim of the ‡vetåmbaras, particularly that of the Mürtipüjaks, was accepted, to the effect that with their siddhånta they possessed the oldest and the most comprehensive canonical corpus of texts (Jaini 2000a: 28; Folkert 1993a). The Digambaras question the authority of the ‡vetåmbara canon and hold the view that the original texts were, both in form and in the Pråkrit used in them, diverse. They believe that the Jaina canon was lost as a whole and only still available to us in a fragmented form.8 A reason for the research focus on the ‡vetåmbaras which one must not underestimate, which goes back to their history of reforms, is the more or less greater openness on the part of the ‡vetåmbara community with regard to their texts.9 One consequence of this was an early European editorial and translation effort which hardly concerned itself with the non-canonical ‡vetåmbara materials.10 Rarely has an influential Digambara author such as Kundakunda, Samantabhadra, Püjyapåda, Jinasena, Akalaõka or Somadeva been systematically studied in western Jaina research to date.11 Altogether, the philosophical literature of the Jainas, considering the importance of its role in the pan-Indian scholastic world, has been disproportionately neglected. Significant for Jaini is the fact that, for example, prevalent dogmatic discussions, particularly among the Digambaras, has found no resonance whatsoever in western research.12 6

A rare if somewhat less substantial exception to this is the short report of Leumann

1924. 7

“Although occasionally tinged with sectarian spirit” (Jaini P.S. 2000a: 24). The oldest Digambara texts, the two Pråkrit works ¯a™khañ∂ågama and the Kaßåyapråbhr¢ta, can be dated to a period before the earliest ‡vetåmbara texts. 9 In the context of the reform movements within Jainism, starting from the 16th century, a new canonization took place. The Sthånakavåsis constitute one canon which limits itself to 32 out of 45 texts of the ‡vetåmbara canon. 10 Exceptions are, e.g., Hemacandra’s Trisastisalakapurusacarita or Anyayogavyavachedika. 11 P.S. Jaini (2000a: 29) rightly admonishes the scant references in Schubring and Frauwallner. 12 Jacobi (1884: 18) is still cited in Moritz Winternitz’s Die buddhistische Litteratur und die heiligen Texte der Jainas: “The philosophy of the Jainas is like a dead language: it is learned and used but it cannot further develop” (Winternitz 1920: 355). Padhmanabh 8

VI. Thinking about Traditions in South Asia Today

575

Jaini’s critique recalls an important point: the one-sided canonization of Jaina texts on the part of the European research on Jainism. Hermann Jacobi’s imprint of the Åcåråõga is the sole Pråkrit work to be included in the Series of the Pali Text Society (PTS). The pattern of research on Buddhism, in which with the PTS imprints an early orientation towards the canon of one particular school, i.e. that of the Theravådins dominated, which, further, was also seen as ‘early’ and ‘original’,13 also methodically further influenced research on Jainism. On the other hand Jacobi, through his findings regarding the relationship between Jainism and Buddhism and the proof that Jainism takes recourse to a line of tradition which is older than Buddhism, perhaps made as significant a contribution in the Jaina context as he did to western Indology.14 Jacobi’s position, though, did not imply in the least that Jainism research, within the framework of western inquiry, could free itself from the model of the Buddhism research. Rather it remained not just subordinated to the latter but was modelled upon it.15 In fact, this result is in line with Jacobi’s responsibility for the acknowledgement of the ‡vetåmbara claims to authenticity, both within western Indology as well as with regard to ‡vetåmbara self-perception itself.16 This position relied primarily on Jacobi’s study of the Kesigautamasamvåda from the Uttarådhyayanasütra, a dialogue between Kesi, a disciple of the predecessor of Mahavœra Pår†va and Gautama, the chief disciple of the Jina Mahavœra. In this text it is stated that the tradition of the robes goes back to Pår†va and, hence, that this tradition which the ‡vetåmbaras take recourse to is the original one (Jacobi 1884a: xxii). The focus in the early research on one canon, in particular in Jacobi’s works, lead to the neglect of the extensive literature of the Digambaras.17 The stress placed on this one remark about the Digambara canon and the undue importance given to it in the context of evaluating Digambara literature led to a partiJaini (Jaini P.S. 2000a: 30) refers to a Hindi publication Tattvacårca, Jaipur 1967 in which a widely publicised dispute, held in Jaipur, is documented which, based upon the medieval Digambara Dogmatics of Kundakunda and his work Samayasåra, concerned itself with the non-modifiable sequential order (kramabaddhaparyaya) of the modifications (paryaya) of substance (dravya). 13 For instance, publications such as Sakya or Buddhist Origins (Rhys Davids 1931). The critique of this approach has been summarized by Steve Collins (Collins 1990: 89). 14 Jacobi 1884b: xviii-xxxv. Nevertheless, Jacobi alleged simultaneously throughout “[…] that both Jainism and Buddhism owed to the Brahmans, especially the Sa∫nyåsins, the groundwork of their philosophy, ethics, and cosmogony” (p. xxxv). 15 “In establishing the Jains’ independence from Buddhism, Jacobi had in effect treated the Jains as a miniature Buddhism, i.e. as a parallel but distinctive ascetic movement whose history should be understood on the same model as Buddhist history” (Folkert 1993c: 98f). 16 The article of Bhler (1887: 165-180), which was elaborated upon in subsequent years, attempts to look anew at the Jaina canon and argues in favour of its antiquity and against its later compilation in imitation of a Buddhist exemplar. 17 A response which challenges Jacobi’s interpretation is to be found in Prafulla Kumar Modi’s Hindœ introduction to his edition (1965) of the Påsanåhacariu. For further reactions to Jacobi’s position see Nahar 1929: 167f, as also Nahar 1930.

576

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

san approach on the part of western researchers vis-à-vis a religious group which thematizes the rupture of a canonical tradition, but would refuse to maintain that their collections do not contain texts or doctrines which bear witness to word of the Jina or that these collections would not in themselves serve the purpose of canonical texts.18 The problem of the western tendency to favour canons has been addressed by Kendall Wayne Folkert.19 An example mentioned by Folkert is the inclusion of canonical ‡vetåmbara texts into the series The Sacred Books of the East following a Christian-Protestant tradition. John Edward Cort elaborates on the influence of Protestant hermeneutics for the reception of Jaina and ancient South Asian texts in general and how Jaina texts and history are viewed today (Cort 1989: 28). In Jaina accounts on the history of their religious lineage texts usually play a fairly minor role. Folkert remarks that Jainology has so far failed to study how Jains themselves view the role the texts play in their tradition (Folkert 1993b: 89). Typical for 19th century Jainology is its one-sided interest in doctrinal questions which reduced comparisons with Buddhism to the goal of defining Jaina religious tenets and to writing a parallel history of ideas which remained widely de-contextualized. Cult and practice did hardly figure. Folkert’s achievement in his research on ‡vetåmbara Jains in Western India is to have directed the focus on the utilization of the canon in both ritual and exegesis, especially the role of the Pratikramañasütras, anthologies of canonical passages with expositions in Gujaråtœ. This particular approach to texts and contexts within which tradition is handed down enables the western researcher to refrain from joining in the activity of canonizing, and instead to open up the space in which the community handles the text and thereby performs tradition.20 Cort’s critique of Jainology up to the present date (Cort 1989: 31-41), which is more radical than Jaini’s, analyzes meticulously this process of adapting new data to old models and the excision of those data which would demand a comprehensive remodelling.21 He describes this as a process whose various stages are marked by 18 Klaus Bruhn calls this the ‘canonical factor’ (kanonischer Faktor) in the history of Jainology (Bruhn 1987). 19 “Western scholars have treated the Jain scriptures as a closed canon, on the model of the Bible, dating from the 5th century C. E. Scholars have therefore tended to treat deviations from this closed canon as indications of ‘heresy’ or of sectarian tendencies. However, my own research indicates, that the standard treatment may be in error” (Folkert 1993b: 88). 20 “[…] The interplay between the historical and written text and its use as ritual recitation opens up the arena of oral/aural significance for scripture in a tradition heretofore studied almost solely in terms of written texts and their propositional content” (Folkert 1993b: 93). 21 Cort draws on the Geertzian opposition of ‘model of’ and ‘model for’, which refers to the alternatives of a model which is derived from and one which is applied to data (Geertz 1973: 93).

VI. Thinking about Traditions in South Asia Today

577

the reconfirmation of a model which is useful to both the researcher and the apologetic preserver of a tradition because it protects, conserves and regulates the access to canonical knowledge.22 Jainologists and representatives of Jaina orthodoxy meet in affirming the centrality of mokßamårga, the path to liberation, i.e. the ascetic ideal of renunciation.23 Authors like Carlo della Casa and Padmanabh Jaini consider the relation of practice and orthodoxy as one in which practice is gradually assimilated to and argues in favour of a position which regards the transient as secondary and peripheral (Della Casa 1971: 362; Jaini P.S. 2000b: 267). Instead, it can be shown that presumed accretions have existed for a long time or have been identified as accretions only on the basis of a model which presupposes stasis. Cort distances himself from a critique of Jainology which, like Jaini’s, remains within the orthodox framework of a mutually supporting Orientalism and restricts itself to demanding (Digambaras in addition to ‡vetåmbara, philosophical in addition to narrative literature, modern in addition to medieval commentaries) merely new fields of research.24 As Folkert does and following Stanley Tambiah (Tambiah 1970: 367-370; Tambiah 1984: 7f) he proposes a model which represents textual transmission as running parallel to the tradition of religious practice and argues in favour of studying the cross currents between these two strands.25 This model is thought to resist the Orientalism inherent in text-oriented approaches which enhances the status of the past at the expense of the present. Cort argues against Edward Said that in fact “Jain dogma and Orientalist scholarship coincide in their valuation of the past over the present” (Cort 1989: 27), and that western Indology selected Jaina normative judgements and translated them into historical description (Cort 1989: 39f). The 24 tœrthåõkaras all lived in a phase of cosmic decline (avasarpiñœ) in which salvation was still possible. However, Jainism presents a soteriological cosmology while Orientalism insists on the inherent value of history and thrives on a critique of tradition. This however does not mean that a different derivation and a different goal may not share a common practice. These differences appear only in all their distinctness if one considers the colonial context and regional developments within the Jaina community. As a matter of fact, as far as the Jains were concerned the forced ‘opening’ of the Jaina libraries by western scholars led to the formation of a distinct strategy to instrumentalize their own tradition (Flgel 1999: 8). 22

For the term ‘protective strategy’ see Proudfoot 1985: 198. For the conflicts regarding this issue within the Jaina tradition see Dundas 1993:. 24 In his contribution to ‘sectional studies’ Klaus Bruhn has shown the urgent need for a ‘scheme of research’ (Bearbeitungsschema) in Jainology and has made fundamental methodical and formal proposals (Bruhn 1990: 36-54). 25 For a comparison of Jainism and Buddhism from this perspective see Carrithers 1990. 23

578

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

2. The Pandit and His Cause. The so-called ‘opening’ of the Jaina libraries is a frequently narrated myth in the history of Indology. In Jaina historical awareness this event hardly figures at all.26 Books, whose production originally was regarded an act of violence, developed into objects of worship and sectarian conflict (Granoff 1993). Since the 11th cent. they have been hidden in subterranean bañ∂års (‘treasure houses’) out of fear of destruction by Hindus and Muslims.27 In the history of the ‘opening’ of the bañ∂ars three motifs figure predominantly. It is interpreted as a protest of reformist laypeople against the privileges of ‘landowning monks’ (yatis). Secondly, it is said to be the result of the endeavours of reformist scholars such as Åtmåråm (1837-1896) to make Jaina texts accessible to a larger public.28 Thirdly, this presumably historically extended process is reduced to the well-documented event of their ‘discovery’ and utilization by western Indologists. Georg Bhler and Hermann Jacobi visited Jaisalmer in 1873-1874 to inspect the local library and “[…] to make its content accessible to scientific research” (Bhler 1875: 82).29 The success of this visit was mainly due to Bhler’s friendship with ‡rœpüj Jinamuktisüri, whom Bhler held in high esteem30 and who was the head of most of the Khartara Gaccha, an orthodox ascetic branch dating back to the 11th cent. Bhler himself was aware of the difficulties of his undertaking and knew of Col. James Tod’s description of his visit to the Hemacandra Bhañ∂år in På™an (Tod 1839). In his report Eine Reise durch die indische Wste, written for the Österreichische Rundschau, Bhler relates his negotiations, the resistance encountered and his apparent success. As if he had foreseen the outcome he writes: “Much of it requires conflicts in which the Europeans do not always triumph” (Bhler 1883: 518). Bhler assumed the fear that he may demand the manuscripts as presents to be the reason for the reluctance to answer to his questions regarding them (Winternitz 1898: 344). Later he assumed a conflict between his friend and supporter and the Jaisalmer authorities, especially after the latter presented a fake report on the number of manuscripts contained in the library. Bhler was so 26 Flgel rightly calls this “[…] merely a footnote in the history of the Jaina renaissance in the late 19th and 20th centuries” (Flgel 1999: 10). 27 The key article on this topic is Cort 1995. 28 For a comparison of Jaina reform movements see Dundas 1985, as well as Dundas 1987. 29 For the stage of the survey at that period see Dalal 1923. 30 With Alsdorf too we find an appreciation similar to that of his preceptor, though more balanced: “Munimahåråj Puñyavijaya, a true scholarly person, sponsor of Western scholarship of deep understanding […] has, through dedicating his life to old manuscripts, acquired merit which can hardly be underestimated both regarding his own religious community and Indology” (Alsdorf 1974: 160 [my translation]).

VI. Thinking about Traditions in South Asia Today

579

enraged about the report that he reported to Thakur Juvansingh that the Jaina committee was trying to deceive him. As a matter of fact, Bhler, by threatening to file a report to the colonial administration which could cost the committee’s members their posts, used his position to pressurize his adversaries and finally make them give him what he wanted (Bhler 1883: 530-532). However, Bhler did not succeed in cataloguing the entire library as Puñyavijayas catalogue dating 1972 shows.31 In this case the authority over the manuscripts seems to have lain with the local council, the pa∞cåyat, and not with the yati. It is possible that in Bhler’s case pa∞cåyat and yati were actually not following conflicting agendas, as the German Indologist assumed, but that they decided to sacrifice a part of the manuscripts in order to safeguard the remaining corpus (Flgel 1999: 6). Hirarlål Jain and Adhinåtha Neminåtha Upådhye acted in a similar problematic way when they, in their quest for the in their view last manuscript of the Digambara ¯a™khañ∂ågama, managed to secretly remove the texts from the Mudbiri monastery (Dundas 1992: 56). Although today the contents of the bhañ∂års of Jaisalmer are wellknown, the access to old material, as Paul Dundas remarks,32 remains difficult.33 Dundas also points to the basic conflict originating in the Indologists’ demands and the Jains’ fear the books might be shipped away and become inaccessible: “For the European, the value of the manuscripts lay in their content by means of which European history could be reconstructed, while for the Jain their true worth lay in their role as sacred objects” (Dundas 1992: 57). While in Bhler’s case the parties seem to have been united against the colonial authorities, there is also evidence of conflicts between conservatives and reformers within the Jaina community, as between Vijayavallabhasüri (1896-1954) and the members of the ‡vetåmbara Jain Conference consisting of both monks and laypeople. Frequently, that led to lawsuits over the possession of manuscripts, setting communities of monks and nuns, as well as laymen and women, saõghs and yatis against each other, conflicts which were usually won by the yatis (Cort 1995: 81). Matters relating to ownership and the continuing trade in manuscripts has been investigated by John Cort.34 Donald Clay Johnson has pointed out that the rating of and dealing both with and in Jaina manuscripts at the time of Bhler’s activities 31 Here the number of 2697 Mss is given (Muni Puñyavijaya 1972: xi) as compared to the 460, which Bhler counted on the basis of the catalogue of 1780 (Bhler 1883: 532). 32 “Usually requiring a presence of all trustees, a rare event” (Dundas 1992: 72). 33 Alsdorf writes: “Using [the texts] on the spot is impossible; the trustee, Mr. Sevantilal Shah, however, kindly allowed me to take the required photos at his house. He told me that his father had bequeathed him before his death to carefully preserve the Mss and not to show them to anybody. The second advice being fortunately not heeded any longer, […]” (Alsdorf 1974: 165 [my translation]). 34 Cort 1995: 82-86. Bhler’s acquisitions of manuscripts as the result of negotiations with several yatis are an early example of this practice. See Johnson 1992: 203; 205.

580

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

generally imparted a different thrust and meaning to the study of Jaina texts and has actually given important impulses to traditional scholarship (Johnson 1992: 201f). Nowadays, the need for secrecy is being outweighed by a growing activity in printing and distribution of Jaina texts which is not only taken for granted but is actually the central part of the activity of networking and missionarizing. The strategy has reversed from restricting to spreading, from book worship to the production of a large number of copies, which itself of course draws heavily from the charisma which is carried by the book. One of the most important historical contributions of Indological research is undoubtedly to have supplied arguments to Jaina lawyers fighting for the recognition of Jainism as a religion separate from Hinduism. In 1867 the Calcutta High Court ruled that Jainism was no religion in its own right. Three Jaina lawyers, all Digambaras, Padmaråj Pañ∂it, Jagmandirlål Jaini and Campatrai Jain, were reportedly successful in the appeal proceedings mainly by quoting edited canonical texts and specifically so-called ‘law texts’ in order to prove the existence of a separate Jaina juridical corpus. Additionally, they had to face the resistance of the local Digambara convention (Jaina Mahåsabhå) where, according to Campatrai Jain, the fear was articulated that the sharpening of the religious profile and the conflation with a political agenda could lead to an escalation and possible repression.35 This victory was the first in a series of successful fights for the complete recognition of the Jains as a separate confessional group fought with Indological weapons forged by western Jaina studies and utilized as such by the scholarly informed laymen.36 In the light of the successful repeal of the Calcutta High Court’s judgement the Bhårat Jain Mahåmañ∂al, founded in 1882, decided to explicitly honour Hermann Jacobi whose work, owing to Bhler’s compilatory activity, had contributed to shed light on the antiquity of the Jaina tradition and thereby “[…] to force the British Indian courts to recognize the faith of the Jainas as a separate religion, and not as a mere sect of the Hindus”.37 On occasion of his visit on the ‘All-India Jaina Literary Conference’ in Jodhpur on December 27, 1913 (von Glasenapp 1925: 77), Jacobi was awarded the title Jaina Dar†ana Divåkara (‘Enlighter of the Jaina Teaching’). This was followed by a strengthening of contacts between European Jainologists and Jaina reformers such as Vijayaråjendra (1826-1906), Vijayånandasüri (a.k.a. Muni Åtmåråmjœ, 1837-1897) and Vijayadharmasüri (1868-1922), who pursued similar research interests. Moritz Winternitz (Winternitz 1926: 349-377) writes about the help he received from Vijaya35 “They repeatedly passed resolutions against printing. The effect of this has been that the world has not yet known what Jainism is like” (Jain C.R. 1926: 8). 36 In his book Jaina Law, the public prosecutor and judge Jaini describes the case brought forward at the High Court of Judicature, Indore, as to whether the Jains should be classified as a group of “Hindu dissenters” (Jaini J.L. 1916: ix-xi, 120ff). 37 The Times, 3.7.1882, p. 5.

VI. Thinking about Traditions in South Asia Today

581

dharmasüri for the compilation of the passages on Jaina literature in his pathbreaking literary history (Winternitz 1920: 289-35), as well as about the manuscripts received through that contact.38 Alsdorf’s report ‘Neues von alten Jaina-Bibliotheken’ is carried by a positivistic intention of describing the preconditions for the best possible exploitation of the Indian resources.39 He also points out that the results of this European Indological enterprise would be of great interest for the Jaina scholars themselves.40 The interest of Jains in western academic training became, at least for a certain period, part of a changing Jaina scholarship. Well-known is Muni Jinavijaya’s (1888-1976) visit to Hamburg and Bonn where he met Hermann Jacobi.41 The most visible result was a strong interest in editing and textual criticism of Pråkrit, mainly ‡vetåmbara texts in Ardhamågadhi, as well as the planning and establishment of a series (Banerjee 1987-88: 87-97). The editorial activities concerning Jaina works, especially since the 1960s, have been immense. Important work has been done by the illustrious LD Insitute in Ahmedabad which in its publications has shifted its interest increasingly towards questions of comparative religion. A further milestone was the founding of the Pråkrit Text Society in 1953 and the beginning of publications within the Pråkrit Text Series.42 These activities are fed by the Jaina conviction, shared for instance by Buddhism, that the reproduction of texts produces religious merit. However, by stressing and strengthening the close link between Jaina ‡vetåmbara identity, text and the scriptural language Ardhamågadhi,43 at the same time they reinforce textoriented scholarship. This hinders the development of Jaina studies, be it western or Indian, towards a multidisciplinary research effort, 38 Arrangements had already been made for a meeting to take place in October 1922, when Winternitz learnt about the Jaina scholar’s death, so that he could only come to Shivpuri for the consecration of Vijayadharmasüri’s cenotaph (samådhimandir) on January 22, 1923, following an invitation issued by Upådhyåy Indravijay. See Winternitz 1926: 352. 39 “The foreseeable implementation of this plan regarding the basic written textual foundations will provide the starting point for a European critical edition of the ‡vetåmbara canon, as envisaged by Schubring” (Alsdorf 1974: 166 [my translation]). 40 “Also, such a complete edition has become attainable due to the demand made upon German Jainology by Dr. Raghu Vœra to contribute to his edition of the complete Sanskrit works by publishing a standard edition of the canonical works including the commentaries” (Alsdorf 1974: 166 [my translation]). 41 “When I went to Germany in 1928 with a view of acquiring first-hand information knowledge of the methods of research and with a view to establishing close contact with German scholars working on Indological subjects and especially on Jain literature, the great scholar, Dr. Hermann Jacobi immediately came from Bonn specially to meet me in Hamburg and invited me with great affection to come there and stay with him for some months” (Jina Vijaya Muni 1946: ii). 42 A survey of the Jaina Ågama Series, which was initiated in 1968, is given by Muni Jambuvijaya (1993: 1-12). A synopsis of the multiplicity of series is still a desideratum of Jainology. 43 Cort (1989: 51) points at the pairing “Pråkrit and Jainism” which is to be found as the designation of many academic institutes and departments as well as of the relative section on the All-India Orientalist Conference.

582

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

which would include methods and aims shared both by the social sciences and philology as well as connect different regional Jaina traditions. The focus on canon and language produce borders which become difficult to cross. This applies to an assessment of the ‘grey literature’ of the Jains, dealing mainly with ritual in modern Indian languages, tapping into epigraphical material (Ekambaranathan, Sivaprakashan 1987; Någaråhaiah 1999; Joseph 1997) as well as into the medieval hybrid-language Jaina texts in Tamil and Kannada which played a fundamental role in the creation of an epic literature and the formation of these two regional languages as a whole .44 The research on these two vast fields is still in its infancy. Only the quasi-missionary impetus dating back to the times and works of Jagenderlål Jain, which recalls the early 20th century attitudes of Neo-Hinduism seems to uphold the claim to transcend boundaries and tends to present neither the texts nor the practice, but rather a handful of Jaina doctrinal contents in a western garb. As Padmanabh Jaini writes, the universalist claim, which was a decisive part of the Indological re- or co-foundation of Jaina scholarship in the late 19th and throughout the 20th centuries, still remains to be met.45 This will necessitate dogmatic and institutional transformations (Flgel 1999: 10). Importantly, modern Jaina authors have developed an assertiveness as a result of which they present Jaina classical philosophical tenets as a contribution to philosophical world culture, especially regarding ethics and truth (see Jain, Pandey 1998). Two terms have to be mentioned in this context, i.e. ahi∫så, denotating the doctrine of non-violence and its purported influence on M.K. Gandhi (Tatia 1967-68: 53-57), as well as anekåntavåda (also known as, though the term is less prominent in popular discourse, syådvåda), denotating the doctrine of the many-sidedness of truth, which is praised as a remedy against dogmatism and intolerance.46 In promulgating messages which contain the promises of inclusivism and pragmatism directed both at western as well as Indian middle class hearers in certain aspects contemporary Jainism resembles Neo-Hinduism and Tibetan Buddhism in as much as it has embarked on the enterprise of developing a popular and last but not least Orientalist version of its own dogmatism which is both polemical and benign. 44 See the immense influence of Jaina authors on Tamil literature such as on I¬aõkova™ika¬ with his epic Cilappatikåran (ca. 450 A.D.) or on Koõkuvëlir, whose Peruõkatai (ca. 800-900 A. D.) utilizes Sanskritic and Pai†åcœ material from the Br¢hatkathå. For Kannada one has to mention the early Jaina champu-kåvyas, mixed forms in prose and metre, such as the Ådipuråña and the Virårjunavijaya (ca. 950 A.D.) ascribed to Pampu I. See, for this, Zvelebil 1974: 131-140; Narasimhacharya 1988: 12-18. 45 “Having not confined to the original homeland, India, having made new homes in all parts of the world, now there is an opportunity, indeed a duty, to make this benevolent religion accessible to the whole world” (Jaini P.S. 1990: 9). 46 For a specific occidentalist approach see Jain and Pandey 1999; Bharucha 1984.

VI. Thinking about Traditions in South Asia Today

583

3. Concluding remarks. Whether one takes the historically legitimate intentions of an Indology engaging in the only seemingly counterintuitive enterprise of trying to lay foundations, i.e. Grundlagen, by opting for the top down approach and thereby creating a discipline where the informant in exchange for the know-how he supplies can learn to rediscover or reinvent his own tradition, or whether one takes the need on part of sections of a modernizing Jaina community, both clerical and lay, to have their texts undergo a process of text-critical or popularizing reformulation and re-contextualization which makes them compatible with colonial and postcolonial knowledge systems they themselves are ready to establish and perpetuate, both agents engage, more or less knowingly, in the same historical process and the ways they collaborate, resist or compete, as I have tried to show, are complex. Building traditions is so crucial, because –as Gadamer has shown– their strength and their dynamism lies in that they are valid without requiring a foundation, that they themselves are our unfounded but constantly reassembled foundation (Gadamer 1960: 285). It is the strength of Kipling’s two men, the strength they draw from their respective traditions of learning, that makes them meet and face each other as individuals. The powers which make East and West are the ones which dissolve it.

BIBLIOGRAPHY Åcåråõga Sütra, edited by H Jacobi, Oxford University Press for the Pali Text Society Series, London 1882. Alsdorf, L. (1974), Neues von alten Jaina-Bibliotheken, in Wezler, A. (ed.), Kleine Schriften, Franz Steiner, Wiesbaden, pp. 160-166. Anyayogavyavachedika, with Mallisena’s commentary Syadvadamanjari, translated and commented on by F.W Thomas, The FlowerSpray of the Quidammodo Doctrine, Sri Mallisenasuri, Syadvadamanjari, Akademie Verlag, Berlin 1960. Banerjee, S. (1987-88), Pråkrit Textual Criticism, in «Jain Journal», n. 22, pp. 87-97. Basham, A.L. (1951), History and Doctrines of the Åjœvikas, a Vanished Indian Religion, With a Foreword by L.D. Barnett, Luzac, London. Bharucha, F. (1984), The Role of Space-Time in Jaina Syådvåda and Quantum Theory, Sri Satguru, Delhi. Bruhn, K. (1987), Das Kanonproblem bei den Jainas, in Assmann, A., Assmann, J. (eds.), Kanon und Zensur, Beck, Mnchen, pp. 100-112.

584

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Bruhn, K. (1990), Sectional Studies in Jainology, in Oberhammer, G. (ed.), Proceedings of the VIIIth Sanskrit World Conference, Verlag der Österreichischen Akademie der Wissenschaften, Wien, pp. 36-54. Bhler, J.G. (1875), Dr. Bhler on the Celebrated Bhañ∂år of Sanskrit MSS. of Jessalmir, in «Indian Antiquary», n. 4, pp. 81-83. ______ (1883), Eine Reise durch die indische Wste, in «Österreichische Rundschau», n. 1, pp. 517-535. ______ (1887), On the Authenticity of the Jaina Tradition, in «Wiener Zeitschrift fr die Kunde des Morgenlandes», n. 1, pp. 165180. Carrithers, M. (1990), Jainism and Buddhism as Enduring Historical Streams, in «Journal of the Anthropological Society of Oxford», n. 21, pp. 141-163. Chattopadyaya, D. (1959), Lokåyata. A Study of Ancient Indian Materialism. Munishiram Manoharlal, Delhi. Della Casa, C. (1971), Jainism, in Bleeker, C.J., Widengren, G. (eds.), Historia Religionum, II: Religions of the Present, Brill, Leiden, pp. 346-371. Collins, S. (1990), On the Very Idea of the Pali Canon, in «Journal of the Pali Text Society», n. 15, pp. 89-126. Cort, J.E. (1989), Liberation and Wellbeing. A Study of the Svetambar Murtipujak Jains of North Gujarat, Dissertation, Harvard University, Cambridge, MA. ______ (1993), Introduction, in Cort, J.E. (ed.), Scripture and Community. Collected Essays on the Jains, Scholars Press, Atlanta. ______ (1995), The Jain Knowledge Warehouses: Traditional Libraries in India, in «Journal of the American Oriental Society», n. 115, pp. 11-87. Dalal, C.D. (1923), A Catalogue of Mss. in the Jain Bhandars at Jesalmer, Gaekwad Oriental Institute, Baroda. Dundas, P. (1985), Food and Freedom: The Jaina Sectarian Debate on the Nature of the Kevalin, in «Religion», n. 15, pp. 161-198. ______ (1987), The Tenth Wonder: Domestication and Reform in Medieval ‡vetåmbara Jainism, in «Indologica Taurinensia», n. 10, pp. 181-194. ______ (1992), The Jains, Routledge and Kegan Paul, London. ______ (1993), The Marginal Monk and the True Tœrtha, in Smet, R., Watanabe, K. (eds.), Jain Studies in Honour of Jozef Deleu, Honno-Tomasha, Tokyo, pp. 237-259. Ekambaranathan, A., Sivaprakashan, C.K. (1987), Jaina Inscriptions in Tamilnadu: A Topographical List, Research Foundation on Jainism, Madras. Flgel, P. (1999), Jainism and the Western World. Jinmuktisüri and Georg Bhler and other Early Encounters, in «Jain Journal», vol. 24, n. 1, pp. 1-11. Folkert, K.W. (1993a), Introduction to Jainism, in Cort, J.E. (ed.), Scrip-

VI. Thinking about Traditions in South Asia Today

585

ture and Community. Collected Essays on the Jains, Scholars Press, Atlanta, pp. 1-19. ______ (1993b), The Jain Scriptures and the History of Jainism, in Cort, J.E. (ed.) Scripture and Community. Collected Essays on the Jains, Scholars Press, Atlanta, pp. 85-94. ______ (1993c), Jain Religious Life in Ancient Mathurå, in Cort, J.E. (ed.), Scripture and Community. Collected Essays on the Jains, Scholars Press, Atlanta, pp. 95-112. Gadamer, H.-G. (1960), Wahrheit und Methode, Mohr, Tbingen. Geertz, C. (1973), The Interpretation of Cultures, Basic Books, New York. von Glasenapp, H. (1925), Der Jainismus. Eine Indische Erlösungsreligion, A. Häger, Berlin. Granoff, P. (1993), Going by the Book. The Role of Written Texts in Medieaval Jain Sectarian Conflicts, in Smet, R., Watanabe, K. (eds.), Jain Studies in Honour of Jozef Deleu, Honno-Tomasha, Tokyo, pp. 315-338. Hemacandra, Trisastisalakapurusacarita, edited and translated by H.M. Johnson, The Lives of Sixty-Three Illustrious Persons, Oriental Institute Gaekwad Series, Baroda 1931-1962, 6 voll. Jacobi, H. (1884a), Jaina Sütras I-II, Oxford University Press, Oxford. ______ (1884b), Über die Entstehung der Çvetåmbara und Digambara Sekten, in «Zeitschrift der Deutschen Morgenländischen Gesellschaft», n. 38, pp. 1-42. Jain, C.R. (1926), The Jaina Law, The Development of Print and Publ. Co., Madras. Jain, S., Pandey, S. (eds.) (1998), Jainism in a Global Perpective, Parshvanatha Vidyapitha, Ahmedabad ______ (eds.) .(1999), Multidimensional Application of Anekåntavåda, Parshvanatha Vidyapitha, Ahmedabad. Jaini, J. (1915), Outlines of Jainism, Jain Literature Society, London. ______ (1925), The Heart of Jainism of Mrs. Sinclair Stevenson, M.A., Sc.D. (Dublin). A Review, Tract No. 78, Shri Atmanand Jain Tract Society, Ambala City. Jaini, J.L. (ed.) (1916), Bhadrabahu sa∫hitå. Text with Translation and Appendix Containing Full Text of an Important Judgment in a Jaina Case by the Original Side of the High Court of Judicature, Indore, Kumar Devendra Prasad, Arrah. Jaini, P.S. (1990), Jainism as a World Religion, in «Jinamanjari», vol. 1, n. 1, pp. 1-9. ______ (2000a), The Jains and the Western Scholar, in Id. (ed.), Collected Papers on Jaina Studies, Motilal Banarsidass, Delhi, pp. 24-36. ______ (2000b), Is There Popular Jainism?, in Id. (ed.), Collected Papers on Jaina Studies, Motilal Banarsidass, Delhi, pp. 267-279. Jambuvijaya, Muni (1993), The Jaina Ågama Series, in Smet, R., Watanabe, K. (eds.), Jain Studies in Honour of Jozef Deleu, HonnoTomasha, Tokyo, pp. 1-12.

586

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Jina Vijaya Muni (1946), Introduction, in Id. (ed.), H. Jacobi, Studies in Jainism, Jaina Sahitya Sambodhara Karyalay, Ahmedabad. Johnson, D.C. (1992), Georg Bhler and the Western Discovery of the Jain Temple Libraries, in «Jain Journal», vol. 26, n. 4, pp. 197-209. Joseph, P.M. (1997), Jainism in South India, International School of Dravidian Linguistics, Thiruvananthapuram. Kipling, R. (1990), The Complete Verse, With a Foreword by M.M. Kaye, Kyle Cathie Ltd., London. Kumarapalapratibodha, edited and translated by L. Alsdorf, Der Kumarapalapratibodha. Ein Beitrag zur Kenntnis des Apabhramsa und der Erzählunsliteratur der Jainas, Friedrichsen, de Gruyter and Co., Hamburg 1928. Leumann, E. (1924), Einiges von der neueren Tätigkeit der JainaGenossenschaften in Indien, in «Zeitschrift fr Buddhismus», n. 5 (N.F. 2), pp. 127-129. Modi, K. (ed.) (1965), Pasåñåhacariu, Prakrit Text Society, Ahmedabad. Någaråhaiah, H. (1999), Jaina Corpus of Koppala Inscriptions: X-Rayed, Ankita Pustaka, Bangalore. Nahar, P.C. (1929), A Further Note on the ‡vetåmbara and Digambara Sects, in «Indian Antiquary», n. 58, pp. 167-173. ______ (1930), A Further Note on the ‡vetåmbara and Digambara Sects, in «Indian Antiquary», n. 59, pp. 151-154. Narasimhacharya, R. (1988), History of Kannada Literature, Asian Educational Services, New Delhi. Proudfoot, W. (1985), Religious Experience, University of California Press, Berkeley. Puñyavijaya, Muni (1972), New Catalogue of Sanskrit and Prakrit Manuscripts: Jelsamer Collection, L.D. Institute, Ahmedabad. Renou, L. (1953), Religions of Ancient India. Jordan Lectures 1951, Althone Press, London. Rhys Davids, C.A.F. (1931), Sakya or Buddhist Origins, Kegan Paul, London. Said, E.W. (1978), Orientalism, Routledge and Kegan Paul, London. Sangave, V.A. (1980), Jaina Community: A Social Survey, Popular Prakashan, Bombay. Schalk, P. (ed.) (2004), Religion im Spiegelkabinett. Asiatische Religionsgeschichte im Spannungsfeld zwischen Orientalismus und Okzidentalismus, Co-edited by M. Deeg, O. Freiberger and C. Kleine, Uppsala University Press, Uppsala. Stevenson, M. (Mrs. Sinclair) (1915), The Heart of Jainism, Oxford University Press, London. Tambiah, S. (1970), Buddhism and the Spirit Cults in North East Thailand. Cambridge University Press, Cambridge. ______ (1984) The Buddhist Saints of the Forests and the Cult of Amulets, Cambridge University Press, Cambridge. Tatia, N. (1967-68), Ahi∫så, in «Jain Journal», n. 2, pp. 53-57.

VI. Thinking about Traditions in South Asia Today

587

Tod, J. (1839), Travels in Western India: Embracing a Visit to the Sacred Mounts of the Jains and the Most Celebrated Shrines of the Hindu Faith between Rajputana and the Indus, with an Account of the Ancient City of Nehruwalla, W.H. Allen, London. Weber, A. (1883), Über die heiligen Schriften der Jaina I, in «Indische Studien», n. 16, pp. 211-479, ______ (1885), Über die heiligen Schriften der Jaina II, in «Indische Studien», n. 17, pp. 1-90. Weber, M. (1988), Gesammelte Aufsätze zur Religionssoziologie, J.C.B. Mohr, Tbingen, vol. 2. Winternitz, W. (1898), Georg Bhler. In Memoriam, in «Indian Antiquary», n. 27, pp. 337-344. Winternitz, M.(1920), Geschichte der indischen Litteratur. Die buddhistische Litteratur und die heiligen Texte der Jainas, C.F. Amelangs, Leipzig, vol. 2. ______ (1926), Erinnerungen an Indien. Eine Woche bei den JainaMönchenin Shivpuri (Gwalior), in «Zeitschrift fr Buddhismus», n. 7 (N.F. 4), pp. 349-377. Zvelebil, K.V. (1974), Tamil Literature , Harrassowitz, Wiesbaden.

BRUNO LO TURCO Evaluation or Dialogue? A Brief Reflection on the Understanding of the Indian Tradition of Debate* 1. Introduction Those who have been investigating Indian traditions in recent years will be perfectly familiar with the fact that Wilhelm Halbfass in his brief yet famous text laden with implications, entitled Preliminary Postscript and published in the volume India and Europe (Halbfass 1988: 160-170), suggested that Indological praxis should acquire a theoretical framework constituted by hermeneutic philosophy, understood especially as philosophy that developed along the Heidegger-Gadamer axis.1 Here, in re-examining Halbfass’s suggestion and comparing it with other possible approaches, we intend to advance the possibility of reinstating the study of the Indian tradition of debate programmatically –and when referring to the past, where possible, descriptively– within a framework that comprises, as well as the above-mentioned philosophers, also Richard Rorty, who follows along the Heidegger-Gadamer axis, reviving its obsolete aspects,2 * I especially wish to thank Federico Squarcini for making available his unpublished text, La stagione ermeneutica aperta da ‘India and Europe’, on Wilhelm Halbfass, which “woke me from my dogmatic sleep”. I am also grateful to Alessandro Graheli, for a conversation that aroused my interest in the subjects dealt with in this article. 1 I owe this definition of ‘hermeneutic philosophy’ to Vattimo (1994: 5). 2 It was precisely the fact that themes in Heidegger’s philosophy, and particularly the theme of being, were deemed to be obsolete that induced Halbfass apparently to partially rethink his hermeneutic attitude with a view to adhering to analytic methods (see Halbfass 1992: 14). It must be remembered, however, that in actual fact there is also a Heideggerism that emerges with Gadamer and culminates in Rorty, in which the hermeneutic influence is stronger than the ontological. This Heideggerism “departs considerably from the notion of being” (Marconi, Vattimo 1986: xxiv). Hence, the rejection of the theme of being does not in itself imply the abandonment of hermeneutics. It does not seem that Halbfass in his ‘rethinking’ (on this whole question see Dallmayr 1997: 55-58 and the response of Halbfass 1997: 148-152) has taken this into account, despite his earlier support of Gadamer’s

590

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

though he is at the same time the heir to the developments of that opposite current of neopositivistic and analytic speculation (see Marconi, Vattimo 1986: viii-ix). One of the most salient features of Indian traditions is precisely the presence among them of a transversal tradition of systematic reflection and theoretical debate.3 Within the ambit of this tradition “in principle there were no distinctions made between ‘human’ sciences and ‘natural’ sciences” (Staal 2001: 611), hence it consists of a vast and variegated accumulation of scientific and philosophical knowledge and methodologies. One of the preliminary issues to examine closely when seeking to understand the Indian world is how this prominent tradition is to be understood.4 Namely, what are the conceptual tools to be employed in this work of understanding? This work of understanding is all the more important since a knowledge of the specific features of one’s own culture cannot fail to go hand in hand with a knowledge of other cultures. In fact, how is it possible to understand the specific features of a culture except by seehermeneutic view. He does not seem fully aware of the fact that if for Heidegger metaphysics is the history and destiny of being, it is precisely for this reason that Heidegger tried to reflect on being in terms that are no longer foundationalist (see Marconi, Vattimo 1986: xxix-xxx). In other words the question of being is the focus of Heidegger’s concerns –and his insistence on this point becomes in the words of Halbfass (1992: 9) “stubborn, almost obsessive, and highly idiosyncratic”– precisely because Heidegger attempts to avoid thinking of being in foundationalist terms. The epistemologists are far more obsessed by being, understood by them as objectivity, than Heidegger. The fact that Heidegger’s concerns were not vain seems to be confirmed moreover by the fact that Halbfass decided –but perhaps, at least in part, knowingly and ironically, as he claimed later (see Halbfass 1997: 150151)– to throw himself into the trap of being as objectivity: he resolved to apply ‘objective and universal’ methods of analysis to Indian thought (“clarification implies a commitment to objective standards of precision and analysis”, Halbfass 1992: 13). He claims, in other words, that we must ask ourselves to what extent Indian thought grasps being, since he adheres to what might be defined as “a metaphysics rewritten in terms of the linguistic turn that has occurred in our century” (Gargani 2001: xi). Halbfass (1992: 15) moreover, states that he does not want, however, to abandon the hermeneutic approach completely, thus actually risking to run into the danger of what Dallmayr (1997: 58) describes as ‘ambivalence regarding method’. In replying to Dallmayr, Halbfass (1997: 151) states that a “limited and cautious application of modern analytic tools as tools of clarification” does not imply abandoning the hermeneutic approach. All things considered, I think this position can be shared insofar as it does not claim to understand the Indian traditions exhaustively by applying these tools. Rather, this application leads to a clarification of if and how the Indian traditions have historically contributed or can still contribute to contemporary science, which I hope will become evident in the course of this article. 3 Bronkhorst speaks of ‘rational debate’ instead of ‘theoretical debate’(Bronkhorst 2001a); however, in agreement with what Rorty (1989: 44) maintains, I consider the distinction between rational and irrational obsolete. Our discourse, which takes as its starting point the above-mentioned text by Bronkhorst, will therefore replace the term ‘rational’ with ‘theoretical’ (taking ‘theoretical’ to mean that which offers a systematic model of understanding), including both scientific theoretical elaboration and philosophical theoretical elaboration. 4 As Bronkhorst (2001b: 196) states: “[…] on peut de façon raisonnable se demander comment il faut comprendre la philosophie indienne, en supposant qu’une certaine sort de compréhension, quelle qu’elle soit, est possible”.

VI. Thinking about Traditions in South Asia Today

591

ing how it differs from our own? A comparison with another culture, which has also produced a widespread theoretical debate, is very rewarding since it allows us, at least partially, to trace the specific features of Western theoretical debate.5 2. The hermeneutic approach It is immediately evident that approaching the Indian tradition of theoretical debate and systematic reflection from a Western standpoint means first and foremost dealing with a vast mass of texts that belongs to a cultural context that is very different from that of the contemporary West or more generally of the contemporary world as a whole.6 It would therefore seem opportune from the outset to resort to the tools provided by philology and hermeneutics7. Let us consider hermeneutics. It is to be understood in the first place, beginning approximately with Schleiermacher, as a doctrine of the ways of understanding a text whose meaning is not immediately evident, because it is historically, linguistically, psychologically etc. far removed from us.8 Since this starting point, the understanding of hermeneutics (both in the subjective and objective sense) has undergone considerable developments, especially thanks to Heidegger and, even more so, Gadamer. Halbfass’s many merits include having clarified, especially in Preliminary Postscript in India and Europe, the inevitable importance of these developments as regards understanding Indian traditions. It would seem difficult, though legitimate, to doubt the fact that Halbfass has inaugurated a new ‘hermeneutic season’ –according to Squarcini’s (2001) well-chosen phrase– by effectively opening up infinite possibilities of understanding. The central thesis of Preliminary Postscript is that the hermeneutic prospects revealed by Gadamer’s work are valid not only within the context of our Western tradition, to which Gadamer restricts his discourse, but also within the trans-cultural context of the encounter between Indian traditions and Western traditions. 5 If it is true as Bronkhorst (2001a: 11-15) maintains, that only two cultures, the Indian and the Western culture, in the history of humanity have produced a ‘rational’ debate (but we should rather say an ‘extensive theoretical debate’), this comparison would be the only possibility of grasping the specific nature of Western ‘rational’ debate. 6 We wish here to examine especially the problem of how to deal with this textual mass. Hence we will place the question of neo-Hinduism ‘between brackets’, even though in Halbfass’s view it is quite the opposite of negligible, as Franco and Preisendanz (1997: xiii-xvii) have clearly summarized. Moreover it seems legitimately possible to maintain, as does Mohanty (1997: 164), that the question of ‘neo-Hinduism’ does not exist. 7 These two disciplines are logically inseparable, since the former provides us with the empirical tools and the latter provides us with the speculative tools. And yet it seems that the relationship between philology and hermeneutics has changed in recent years to such a point that they astonishingly reached complete reciprocal ignorance (Segre 2001). 8 I am paraphrasing here the clear definition of ‘interpretation’ in Vattimo (1981: 451).

592

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

With Heidegger and Gadamer hermeneutics, originally the “doctrine of the ways of understanding a text”, extends almost to infinity, thanks to the recognition of the fundamental ‘linguisticity’ of understanding, or rather the fact that every entity has always been inextricably tied to language. Thus hermeneutics becomes the doctrine of the interpretation of the world, since the world, consisting of entities that are in essence linguistic, is text. Hence hermeneutics is the doctrine of human understanding itself. Since the world is language, or a combination of languages, language is the world. And hence the vast mass of texts of the Indian tradition of theoretical debate brings with it a world (at least one world, but perhaps several worlds) of exceptional complexity, which is not our contemporary Western world. In Halbfass’s view, therefore, the undeniable historical ‘linguisticity’ constituting existence, which is Gadamer’s starting point, is not only the positive basis for the possibility of a fruitful dialogue with our own Western tradition, but also for a fruitful dialogue with Indian traditions. A dialogue is fruitful when it affects the interpreter’s understanding of the world, when the experience of the dialogue, arising against the background of former experiences, reinterprets those former experiences, when, in other words, it extends knowledge. 3. The ‘objective’ approach However, as Halbfass points out: An increasing number of Western scholars try to do justice to Indian philosophy by treating it simply ‘as philosophy’, regardless of its cultural and historical origin and context, and dealing with it in terms of truth and validity. In most cases, this amounts to an application of methods and criteria of modern logic and epistemology or, more specifically, of current Anglo-Saxon analytic philosophy. This perspective focuses on the technical, systematic achievements of Indian philosophy, and it tries to measure and clarify them by using the ‘most advanced’ standards of modern West thought. Almost inevitably, Indian thought appears as a more or less successful approximation to these standards. (Halbfass 1988: 163)

Hence, in actual fact Indian theoretical debate has also been the object of epistemological and analytic investigation, an investigation whose presuppositions and tools are completely different from those of hermeneutics. “Treating Indian philosophy simply ‘as philosophy’” means not acknowledging the existence of a gap between the text and the reader: namely, the text is considered immediately available, its meaning evident and the need for interpretation is excluded. This attitude means that the relationship between Indian theoretical debate and contemporary Western thought is not understood as an encounter between two historical traditions, the bearers of (at least)

VI. Thinking about Traditions in South Asia Today

593

two distinct languages, but as the encounter between a historical tradition, that of India, and the West understood as the bearer of objective methods of enquiry.9 But what are the assumptions of epistemological enquiry applied to this context? These assumptions (re-)emerge quite clearly in some recent Indological publications. In one of these J. Bronkhorst (2001b: 204), in reply to the question of how Indian philosophy is to be understood, proposes the following interpretive criterion: that the history of Indian philosophy should be considered in an analogous way to that of Western natural sciences. This criterion is suggested on the basis of a precise similarity: both these histories can be described as selection processes. However, Bronkhorst himself underlines the differences between the history of Indian philosophy and the history of Western natural sciences: Il est certain qu’il y a des différences importantes. Dans l’histoire de la philosophie indienne, tout comme dans la science moderne, le processus sélectif est en marche. Ce processus est relativement restreint comparé à la science moderne, parce que la philosophie indienne ne permet que rarement une confrontation systématique avec la réalité. Mais il y avait la critique impitoyable émanant des penseurs des autres écoles, et par conséquent, l’attitude critique des penseurs envers leurs propres idées. Á quoi cela a-t-il mené? Á l’acceptation de positions qui pouvaient être considérées comme des solutions à des problèmes perçus. (Bronkhorst 2001b: 204)

As Bronkhorst (2001b: 196) notes therefore: “Contrairement à la science moderne, les opinions philosophiques des penseurs indiens étaient souvent déterminées dans une grande mesure par des facteurs sociaux et religieux évidents […]”. What is especially interesting in Bronkhorst’s discourse are those premises that limit the validity of the analogy between modern science and Indian philosophy: the latter, unlike modern science, does not confront reality and is by contrast determined for the most part by social ad religious factors.10 Staal too has recently dealt briefly with the problem of how to understand the ‘scientific side’ of Indian rational debate. Staal begins with a general reflection on the nature of science: 9 It should be added that the English-speaking culture, in which analytic philosophy is dominant, is extremely self-referential: the English language publishing market has around two percent of translated books, while, for example, the Arabic market has around ninety per cent. And naturally being rarely in the habit of translation does not make for a predisposition towards the hermeneutic approach that is always rooted in a process of translation. 10 It is quite obvious that at the basis of Bronkhorst’s general approach, and similar approaches (see below), lies Popper’s epistemology, especially his famous, but not universally accepted, theories of an objective truth as the ideal aim of the path of science and of objective knowledge (theory of ‘World 3’; see Popper, Eccles 1977: 36 ff.). Then, in particular, in proposing his interpretive criterion, Bronkhorst refers back to Popper’s thesis that Western science can be thought of as a contest between hypotheses, or rather as a process that implies

594

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

[…] despite the claims of the relativistic theories of science, which invariably refer back to Thomas Kuhn’s paradigm shifts, it cannot be denied that we –or, more precisely, those of us who operate as specialists– know and understand a far larger number of things than the best informed of our ancestors. Naturally, not all sciences have steadily evolved, according to progressive lines of development; their paths have culs-desac and much knowledge has undoubtedly been lost, but in them great openings are also evident that represent not only an extension, but also a widening of our knowledge of facts: particularly about the universe, life, language, society and the brain. (Staal 2001: 615)

What is interesting in this premise, with a view to understanding Indian theoretical debate, is that science is identified with an ‘extension’ and ‘widening of our knowledge of facts’, namely with progress, and that therefore science is not to be considered fundamentally a historically determined tradition. According to Staal: Scientific knowledge emerges from intuitions rooted in facts and refined by logic, and constantly verified in the light of both these factors. In the study of the past we cannot avoid using a modern yardstick: for example, we talk of primitive wheels from the viewpoint of presentday wheels with spokes and the means of transport we are familiar with. (Staal 2001: 617)

Therefore in the view of these authors there exists really “a great progressive movement of the natural science of all humanity” (Staal 2001: 618, 629). The Indian scientific tradition must be ‘evaluated’ in the light of this great progressive movement. This evaluation leads to the acknowledgement that “enormous progress can be discerned in the Indian science of language and above all in its early stage of development” (Staal 2001: 623) and that further enormous progress can be perceived in the discovery of the zero (Staal 2001: 629). From this standpoint other cultures simply represent the preparatory stages of the only true human civilization. But what happens to what has not contributed in Indian rational debate to the ‘great progressive movement’? What has not forwarded the path of progress has a ‘negative connotation’. Staal writes: Newton had remained attached to Latin, just like the mathematicians of Kerala had remained attached to Sanskrit, two classical languages that were not formal enough to trigger a scientific revolution. But unlike Madhava and his successors, Newton had begun to develop the transition towards the use of an artificial language and the work he left unfinished was completed immediately afterwards by Euler and other scholars. The Indian mathematicians not only failed to take this road, the elimination of certain hypotheses (the result of them having proved to be false) and the survival of others. The world of language, ideas and science is, just like the behavioural world, an evolutionary world, in which from a certain viewpoint, theories live lives of their own (see Popper, Eccles 1977: 132-134). This application of evolutionary theory to human knowledge becomes concrete in the evolutionary epistemology of D. Hull and H.C. Plotkin.

VI. Thinking about Traditions in South Asia Today

595

but they made things worse: in fact, they used strange notation systems instead of resorting to that great Indian discovery of decimal notation with the zero and to the simple methods of calculation based on this. That astonishing development that concluded in an impasse is one of the greatest paradoxes of Indian science and obviously has a negative connotation. (Staal 2001: 637)

Staal also maintains: [Une] science est évaluée en fonction de ses résultats et de ses conséquences, non pas en fonction de son arrière-plan ou de ses origines. Si l’on acceptais l’hypothèse que la science dépend du contexte e de circonstances telles que les croyances des gens qui ont contribué à sa création, alors le théorème de Pythagore serait rejeté parce qu’il croyait en la réincarnation, l’astronomie de Kepler à cause de son astrologie […]. Au contraire, et en accord avec les principes de sélection qui gouvernent l’évolution biologique, les gens ne se rappellent que de la science, et, à moins d’être historiens, oublient tout le reste. (cit. in Bronkhorst 2001b: 200)

4. One step forward: Rorty’s distinction between epistemological thought and hermeneutic thought The root of Halbfass’s above-mentioned hermeneutic proposal lies in Heidegger’s famous consideration on the hermeneutic circle, which is the crux of Gadamer’s theory of the hermeneutic experience: [The circle] is not to be reduced to the level of a vicious circle, or even of a circle which is merely tolerated. In the circle is hidden a positive possibility of the most primordial kind of knowing. To be sure, we genuinely take hold of this possibility only when, in our interpretation, we have understood that our first, last and constant task is never to allow our fore-having, fore-sight, and fore-conception to be presented to us by fancies and popular conceptions, but rather to make the scientific theme secure by working out these fore-structures in terms of the things themselves. (cit. in Gadamer)

Is it not possible that, by maintaining that only Western science deals with reality, whereas ‘Indian philosophy’ is determined mostly by social and religious factors, we are allowing popular opinion to impose fore-having, fore-sight and fore-conception on us? We begin to suspect that we are still faced with that famous messianic view of the West, deeply rooted in the West itself, a view that culminated with Husserl’s Krisis and according to which the West is considered the source of rational enquiry (the touchstone of every possible form of thought) and also the place where it culminated historically in modern science. This view of the West is corroborated, among other things, by our unwillingness to grasp thought paradigms that are not established in our own society and hence by our reluctance to interpret.

596

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

The best antidote to this view is obviously to remember how Heidegger “questioned substantialistic logic itself and the objectifying thought-language that it governs, their history and their practice” (Saviani 1998: 29). In this regard, let us consider what Heidegger himself stated in a lecture: Thought – this is our Western thought, determined by the logos and in tune with it. This does not in the least mean that the ancient world of India, China and Japan has remained devoid of thought. But rather, the indication of the nature of logos, typical of Western thought, implies for us the injunction, should we dare to touch those distant worlds, to ask ourselves first and foremost if we have ears to listen to what is thought there. This question is becoming all the more pressing, since European thought is threatening to become global: now those very Indians, Chinese and Japanese very often only express their own experiences in our European way of thinking. (Heidegger 1994: 145-146 [cit. in Saviani 1998: 18])

Wittgenstein also calls into question objectifying thought-language in his last writings, where he claims that a sentence does not describe a fact, but has an operative usefulness, it is a move in a language game. Along the same lines Rorty, supporting the work of the linguist D. Davidson, maintains that intellectual progress should be understood as a succession of metaphors and re-descriptions that have proved to be useful in the course of time and not as an increased understanding of how things are in reality; truth is not the objective datum, it is constructed together with languages (Rorty 1989: 7 ff). A possible further clarification of the question of how Indian theoretical debate is to be understood lies, in my opinion, in Rorty’s (1979) distinction between epistemological thought and hermeneutic thought that follows in the wake of the questioning of objectifying thought-language. For Rorty epistemological thought seeks to solve problems within an established paradigm of ways of organizing experience, whereas hermeneutic thought has the task of engaging in a dialogue with the proposals of other paradigms, with different systems of metaphors (and by ‘paradigm’ we must essentially understand ‘language’). Rorty’s distinction is to be considered valid not only within the context of Western cultural history, but also within the trans-cultural context. We as Westerners, may well wish to approach Indian theoretical debate by observing it through the lens of epistemological thought, namely by evaluating it on the basis of the cognitive model adopted for the natural sciences, and we must admit that this way of viewing things is an inseparable part of our cultural legacy. However, is epistemological thought an absolute or a historically determined part of our identity? If we believe like Staal and Bronkhorst that “[une] science est évaluée en fonction de ses résultats et de ses conséquences, non pas

VI. Thinking about Traditions in South Asia Today

597

en fonction de son arrière-plan ou de ses origines”, and we examine Indian thought on the basis of this, then we set our understanding of that thought outside history, claiming to produce an objective and rational evaluation. The hegemony of the cognitive model of the natural sciences deprives all knowledge that does not derive from scientific method of credibility. Thus Indian theoretical debate as a whole is excluded from the number of possible experiences of truth and becomes finally relegated to the archives as a form of fideism, restricted to the sphere of erudite curiosity, rather than challenging us with its otherness. Does not approaching traditional Indian thought by considering it from the standpoint of epistemology imply extending the trend towards complete ‘Europeanization’11 also to the past? Does Indian theoretical debate only merit being studied and remembered insofar as it anticipates the scientific outlook? Can what does not anticipate this scientific outlook be consigned to oblivion as being simply irrational and therefore useless? And yet we seek to understand Indian theoretical debate. As far as that is concerned Gadamer has made it sufficiently clear that in understanding the historical context must be revealed, since understanding is in essence a process that is part of history. In other words, what we, using a contradiction in terms, might describe as objective interpretations do not exist, there is no understanding outside the hermeneutic circle and science cannot permit itself to understand without questioning that very understanding. The subjectivity of the understanding subject is constituted, in fact, solely by “historical finite language that renders possible and conditions his access to himself and the world” (Vattimo 1994: 12). Therefore, when Halbfass (1988: 166) asks: “Is the relationship between India and the West indeed an encounter between two traditions? Is the modern West still a tradition?”, the answer cannot fail to be affirmative. It has been sufficiently proved that the language of science is also historical and finite by the fact that today the theoretical terms used by the science of some decades ago have no reference point; from this it is difficult not to conclude that within a few decades the terms used by current science will have no reference point (Rorty 1979: 257 ff). 11 Here the word ‘Europeanization’ is used within the sphere of a philosophical tradition. But it is clear that it is a question of ‘Westernization’ rather than ‘Europeanization’, since this Europeanization now has a prevalent component of Americanization. J. Houben (2002: 165-168) contests the use of ‘Europeanization’ for other reasons. However, what we are talking about remains clear, namely, the tendency towards the globalization of culture and technology. Halbfass (1997: 148) states: “Nothing in my view expresses the fundamental ambiguity of our pluralistic, globalized, seemingly cosmopolitan world so well as the expression ‘Europeanization of the Earth’. It may be a somewhat hyperbolic phrase, and its precise historical implications may be questionable. Nonetheless, it is a legitimate and effective reminder of the fact that much of what we consider to be universal and global today has very parochial roots, and it suggests the possibility that the openness and universality of our modern world remains caught in a globalized parochialism.”

598

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

The epistemological approach therefore raises the following obvious yet pressing objection: is this the only possible approach, an absolute necessity, or is it the result of a particular historical process that implies a specific, historically conditioned view of reality? The answer we decide to give to this question is pivotal when confronting different cultures, including the cultures of the past. If we decide that this is the only approach possible, the encounter with a different culture will necessarily imply evaluating to what extent this culture is ‘scientific’ or ‘rational’, and how and why this is so. Nonetheless Gadamer’s work clearly shows that scientific objectivity itself, a concept that implies the idea of truth understood as conformity to things “occurs only within the (we could say with Kuhn, paradigmatic) framework of an inherited horizon” (Vattimo 2001: 66), namely only within the most original truth of the historical event of the opening up of a world. Only if we decide that the epistemological approach is historically determined, that is to say only if we openly assume the responsibility for our own tradition, if, in other words, we are aware of our own ‘prejudices’ in Gadamer’s sense of the word (Halbfass 1988: 164-165), will it be possible to establish a real dialogue, where neither of the two interlocutors is the bearer of truth, understood as the faithful description of facts, par excellence. For Gadamer understanding is a fusion of two horizons that initially understand each other as being independent. Therefore, a dialogue with the Indian tradition of debate cannot take place by approaching it through the perspective of evaluating what is rational and scientific in it, based on ‘factical reality’, and what is not. This judgement can be pronounced only if the person judging possesses an adequate description of ‘factical reality’. But, yet again, is such a description possible? Rorty elucidates and clarifies Gadamer’s critique of the absolutist claim of epistemological thought. A good part of his work may be seen as a commentary on Gadamer’s famous statement “Being that can be understood is language”. Rorty notes how for epistemological thought the paradigm of the broadening of understanding is that steadily increasing knowledge of the physical world acquired by science, namely of a world that has nothing of the linguistic. Hermeneutic thought insists that, insofar as we understand something, we do so through a description. And there are no preferential descriptions, since there is no way of getting around our language consisting of descriptions and of arriving directly at the ‘object itself’ (Rorty 2001: 49-50). We can also say that words cannot be examined on the basis of non-words with the aim of understanding which words are suited to the world. Passing off the referential relation as reality will not lead to anything; referential theories, including true-functional semantics, are still only theories, part of more general theories of the world; and we cannot justify a theory by resorting to a part of it (Rorty 1979: 284 ff). As Wittgenstein has

VI. Thinking about Traditions in South Asia Today

599

already made quite clear several times, if explaining a system or semantics means referring to some aspect of reality to which a causal role is attributed in relation to a language or a system of rules, then this explanation will also have to be grounded. But this explanation will be linguistic and will inevitably already assume that which it is intended to explain. Languages and systems of rules cannot be grounded.12 The conflict between positions that we will call epistemological, like Staal’s, and those that we shall call hermeneutic, like Halbfass’s, seems to reflect the age-old conflict between those who claim the intrinsic hierarchical superiority of the natural sciences and those who reject this claim. This claim of superiority is based on the observation that though scientific knowledge, like the whole of human knowledge, consists of descriptions expressed in language, it is also evident that this particular language is more useful for resolving those material problems that torment man independently of the language in which they are described. In other words, the objectivity of Western science is guaranteed by its results: if scientific language functions better for solving problems, then it understands its objects better. Yet, the fact that it is more useful and affords greater insight does not mean that it can get closer to the thing itself, or that it gives a better explanation or understanding of it (Rorty 2001: 55). If, as Heidegger has shown first and foremost, the concept of truth as conformity with the object is false, then the concept of the progress of science understood as the increasingly accurate description of the object is equally false. In fact, how can science guarantee the link between language and the world, given that ‘the world’ is merely that which the science of a certain period understands by world, namely a theory of the world, a constantly changing description (Rorty 1979: 287)? The epistemological approach that within Asian studies does not accept its responsibility, that sets itself up as absolute, excludes the hermeneutic approach proposed by Halbfass in India and Europe. In fact, it is one thing to seek the first signs of contemporary science in Indian theoretical debate, in order to evaluate that tradition, and it is quite another thing to seek a dialogue which brings into play the actual possibility of a Horizontverschmelzung with that tradition. Thus it seems that within the sphere of the study of Asiatic traditions there are two views of otherness: the other as a hermeneutic challenge and the other as the object of an epistemological examination by those who ask how much ‘rationality’ it possesses, namely, to what extent it has anticipated contemporary science or contributed to the ‘great progressive movement’: the more rational it is, the more it will be accepted and even well liked. 12 According to Wittgenstein (1969: 16e, §105): “All testing, all confirmation and disconfirmation of a hypothesis takes place already within a system”. See also Wittgenstein 1967: 57e, §310 ff.

600

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

If Westerners considered traditional Indian thought, namely thought that is ‘other’, merely epistemologically, and not by adopting the hermeneutic approach revitalized by Halbfass, it would mean denying that there exists any experience of ‘truth’ beyond ‘scientific method’. The purely epistemological approach cannot fail to result in the obliteration of Indian theoretical debate. The latter does not make provision for experiments, namely it lacks the very pillar of modern science, nor has it developed a scientifically acceptable formal logic; therefore for an epistemologist it has no value whatsoever. At the most Indian theoretical debate will be the object of erudite curiosity, it will not represent a challenge to the dialogue of existential importance, or a challenge that makes us question our own understanding of the world.13 The epistemological conceptual net is not capable of catching anything in the textual sea of Indian thought, except perhaps the invention of the zero and advancements in linguistics. For epistemologists everything else can be consigned to oblivion. Hence the consequences of this epistemological approach go against the human interest in conserving ideodiversity (Houben 2002). The conservation of this diversity, which is the opposite of the globalization of technology, can only be achieved by adopting a hermeneutic perspective, through dialogue.14 But without the hermeneutic perspective, if we accept the hierarchical superiority of scientific knowledge over the other forms of human knowledge, there is no dialogue. 5. How to understand Indian theoretical debate: the dialogue process If we decide to examine the question in depth, there is no choice: if “preferential descriptions do not exist”, dialogue is not merely a possibility; on the contrary it becomes a necessity. Human knowledge consists in nothing but and cannot be anything but descriptions of objects. As Heidegger says, ‘Language is the house of Being’, whether we are aware of this or not. Scientific knowledge is an accumulation of descriptions, just like Indian theoretical debate. Western science is first and foremost language, or rather, a series of languages, which includes, for example, the language of Newtonian physics and the language of quantum physics with more or less extended areas of contact or even of coordination between them. Analogously, the tradition of Indian rational debate consists of languages, like for 13 Another danger in only adopting an epistemological approach to Indian philosophy is banalization: “In their application within the modern Western world, Indian methods and teachings become part and manifestations of this world, and the constellation of science and technology: And for the foreseeable future, we may have to live with this tantalizing paradox: The globalization of the world coincides with its parochialization; the meeting and ‘dialogue’ of the cultures and religions of this world coincide with their trivialization”. (Halbfass 1988: 441) 14 I obviously support Halbfass in his preference for ‘dialogue’ as opposed to ‘comparison’, on which see Clooney (1997: 34).

VI. Thinking about Traditions in South Asia Today

601

instance the language of Såõkhya or of the Buddhist logical epistemological school. Therefore, the encounter between Western traditions and Indian traditions must consist in the coordination and not the subordination of languages. If we wish to understand Indian traditional knowledge we must connect Western description with Indian description, in other words, we must begin to fuse our own horizon with that of the other and call this process ‘advancement of knowledge’, by acknowledging that neither of the two horizons comprehends (in the dual sense of contains and understands) the thing itself more than the other.15 Understanding occurs not through a comparison between propositions, as in metaphysics, in epistemology and in analytic philosophy, but through the coordination of distinct vocabularies. To paraphrase Gadamer, we must say that in our case comprehension is to be understood as participating actively in an inevitable dialogue that is already underway, in which the East and the West constantly synthesize each other. When encountering Indian theoretical debate, what is pivotal is not the comparison of propositions, which is accompanied by inferential and logico-deductive reasoning, but setting different vocabularies in relation with one another. Here the propositions are not judged by the yardstick of truth, which is claimed to be objective, atemporal and extra-linguistic, but they are related within the sphere of a linguistic and conceptual negotiation. Hence the use of categories such as ‘true’ and ‘false’, ‘right’ and ‘wrong’, ‘rational’ and ‘irrational’ is to be considered inadequate. Such categories make sense only within the self-referential circle of one particular language; ‘false’ indicates only the violation of the inferential rules of that particular grammar and not the transgression of the cognitive relation between language and reality. Thus the answers to certain questions posed by Heidegger, which Halbfass mentions at the end of Preliminary Postscript, are to be found close at hand. There Halbfass (1988: 169) repeated the following questions: How is it possible for the dialogue between East and West to take place outside the framework of objectifying scientific thought? How can one transcend what is European? How can one think at a more fundamental level than that of Western philosophy or that of metaphysics? At present we are in a position to be able to begin to answer these questions. What is outside objectifying scientific thought, what transcends that which is European, what leads us to adopt the perspective of thinking at a more fundamental level than that of metaphysics is the coordination of languages that constitutes a manifestation of Gadamer’s hermeneutics. This coordination, insofar as it is hermeneutic, is an infinite task, and therefore not an answer that is given once and for all. Nonetheless, if we wish to reject 15 Here I am basing myself directly on Rorty’s approach (2001: 50) to the problem of the conflict between scientific theories.

602

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

establishing a rapport with the Indian tradition of debate by evaluating it, namely by operating on an epistemological basis, it is the only possible answer. At the same time it must be acknowledged that expressions such as ‘place oneself outside’, ‘transcend’ and ‘place oneself on a more fundamental level’ are potentially misleading. In fact, here ‘transcending’ is rather erosion from within. Dialogue, the coordination of languages, begins, in fact, for each and every one of us from within our own culture, unless we wish to claim that some people can engage in dialogue without using a specific language. Nonetheless, what is on the one hand a process of erosion,16 is on the other a process of construction of a new house.17 In other words, this process never leads to ‘an outside’, or to ‘a more fundamental level’, except in the sense that the process itself, since it is never-ending, is the ‘transcending’, the ‘outside’, the ‘more fundamental level’. Also the metaphor of the construction of a new house of language must however be rendered more precise: in actual fact, at every moment throughout the building process the house is already complete, given that the langue, every langue, is by nature a coordinated whole, a system, as Saussure was the first to make clear. The possibilities offered by the coordination of languages partially contradict Halbfass (1988: 441-442), when he states, on the basis of Heidegger’s Unterwegs zur Sprache: “Yet for the time being there is no escape from the global network of ‘Europeanization,’ and no way to avoid the conceptual and technological ways and means of communication and interaction which the European tradition has produced”. Overcoming this position is, in my opinion, already implicit in Halfbass’s adoption of Gadamer’s hermeneutics, since this occurs in relation to the approach to Indian theoretical debate. The latter, precisely because it is systematic and ‘rational’, though of a rationality that is other, and in the sheer mass of its texts, appears difficult to assimilate. As Clooney (1997: 34) underlines: “The very fact of its resistance to easy appropriation –its status as ‘out of place’ today– provides a space over against the hegemony of the West”. At the same time the very possibilities offered by the coordination of languages are the rea16 As an alternative to ‘erosion’, we could use the term ‘deconstruction’, also because, just as in Derrida’s ‘deconstruction’, this erosion rather than destroying complicates, or reactivates the complications doused by metaphysics. Halbfass (1997: 143) points out how in the process of understanding and dialogue construction and deconstruction imply each other: “And if understanding involves construction, it is also, and in an equally significant sense deconstruction. Understanding and dialogue are inseparable.” But, as was only to be expected, Derrida ‘deconstructs’ dialogue itself and Gadamer’s hermeneutics (for a summary of the debate between Gadamer and Derrida, see Vergani 2000: 169-174). 17 Obviously I am referring to Heidegger’s metaphors of ‘house’ and ‘dwelling’. As Vattimo (1994: 123-124) says: “The guiding metaphor for the conception of truth, here [within the hermeneutic context], is no longer grasping, ‘taking’ (understanding; Begriff, con-cipere etc.), but dwelling: telling the truth means expressing –manifesting, articulating– belonging to an opening into which we have already always been thrown”.

VI. Thinking about Traditions in South Asia Today

603

son why “the West is turning towards the East for new inspiration or even for therapy” (Halbfass 1988: 440). While “the Westernization of the world has already been completed” (Guidieri, cit. in Vattimo 1985: 160),18 the hermeneutic challenge represented by the mass of texts of the Indian tradition of debate remains inexhaustible. It is only its evaluation in an epistemological key that can eliminate this challenge and certify that Westernization has taken place. To resort once more to Heidegger’s terminology, it must be acknowledged that the dialogue process, just like the creation of a work of art –and the establishment of a Kuhnian paradigm (Vattimo 1994: 23-24)– is an excellent example of the ‘occurrence of truth’. In fact, the work of art is ‘the occurrence of truth’, according to Heidegger, since it opens up a world, namely insofar as it is paradigmatic, as we would say today. Dialogue for its part, through a fusion of horizons –albeit very partial, which never achieves a perfect Hegelian mediation– changes each interlocutor’s horizon; and insofar as the dialogue establishes a new shared linguistic dominion, and the language/thought implies a new order of all entities, effectively opens up a new world. We do not intend here to state the mere illicitness of adopting an epistemological approach to Indian theoretical debate; this kind of enquiry permits us to understand the importance of India for contemporary science: we owe the invention of the zero and the birth of modern linguistics to India. What we mean is that this kind of enquiry is very far from understanding its object, precisely because it merely reduces Indian theoretical debate to what is already known, namely to the invention of the zero and the advancement of linguistics, and little else. How does the hermeneutic approach avoid the accusation of relativism? The hermeneutic approach is, in effect, relativistic, if by relativism we mean the plurality of languages advanced in Wittgenstein’s later writings. But it maybe the case to repeat that opening up to linguistic and axiological systems that differ from the one in which we find ourselves is not relativism in the sense that one system is just as good as another: we can never leave out of consideration the system in which we find ourselves, even if we should wish to do so. In fact, ways in which we initially become familiar with other systems and find our bearings in them will inevitably be based on the features of our system. We cannot fail to be ethnocentric unless we claim to have available a universal language, which ‘the facts’ determine causally, that is capable of rendering other languages commensurable (Gargani 2001: xxiii). But then, as Habermas (1967: 226) made very clear in his controversy with Wittgenstein, the different languages or ‘language games’ (the famous Wittgensteinian expres18 However this is already an over-simplified view, which should be re-examined. (See Hannerz 1992).

604

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

sion also appreciated by Gadamer), are not monads. Languages are subject to constant change, they interpenetrate, and their systematic rules are subject to interpretation and reformulation. In other words, the lines of Gadamer’s horizons change, both due to internal processes and to the fusion with external horizons. But, yet again, in agreement with the concept of the hermeneutic circle, for each subject each move starts from within a certain historical and finite horizon, to draw anew the confines of that horizon.

BIBLIOGRAPHY Bronkhorst, J. (2001a), Pourquoi la philosophie existe-t-elle en Inde?, in Bronkhorst, J. (ed.), La Rationalité en Asie / Rationality in Asia. Actes du colloque de l’Institut International pour les Études Asiatiques (IIAS), tenu à Leiden les 4 et 5 juin 1999, in “Études de Lettres”, 3, pp. 7-37. ______ (2001b), Pour comprendre la philosophie indienne, in Bronkhorst, J. (ed.), La Rationalité en Asie / Rationality in Asia. Actes du colloque de l’Institut International pour les Études Asiatiques (IIAS), tenu à Leiden les 4 et 5 juin 1999, in “Études de Lettres”, 3, pp. 195-221. Clooney, F.X. (1997), Halbfass and the Comparative Project, in Franco, E., Preisendanz, K. (eds.), Beyond Orientalism. The Work of Wilhelm Halbfass and its Impact on Indian and Cross-Cultural Studies (Poznan Studies in the Philosophy of the Sciences and the Humanities 59), Rodopi, Amsterdam, pp. 29-45. Dallmayr, F. (1997), Exit from Orientalism? Comments on Wilhelm Halbfass, in Franco, E., Preisendanz, K. (eds.), Beyond Orientalism. The Work of Wilhelm Halbfass and its Impact on Indian and Cross-Cultural Studies (Poznan Studies in the Philosophy of the Sciences and the Humanities 59), Rodopi, Amsterdam, pp. 49-69. Franco, E., Preisendanz, K. (1997), Introduction and Editorial Essay on Wilhelm Halbfass, in Franco, E., Preisendanz, K. (eds.), Beyond Orientalism. The Work of Wilhelm Halbfass and its Impact on Indian and Cross-Cultural Studies (Poznan Studies in the Philosophy of the Sciences and the Humanities 59), Rodopi, Amsterdam, pp. i-xxiv. Gadamer, H.-G. (1960), Warheit und Methode, J.C.B. Mohr, Tbingen. Gargani, A.G. (2001), La vita contingente. Prefazione, in Rorty, R., La filosofia dopo la filosofia (Biblioteca Universale Laterza 532), Laterza, Roma, pp. ix-xxxi. Habermas, J. (1967), Zur Logik der Sozialwissenschaften, J.C.B. Mohr, Tbingen.

VI. Thinking about Traditions in South Asia Today

605

Heidegger, M. (1994) Gesamtausgabe. LXXIX: Bremer und freiburger Vorträge, Klostermann, Frankfurt a.M. Halbfass, W. (1988), India and Europe. An Essay in Philosophical Understanding, State University of New York Press, Albany. ______ (1992), On Being and What There Is. Classical Vai†eßika and the History of Indian Ontology, State University of New York Press, Albany. ______ (1997), Research and Reflection: Responses to My Respondents. II. Cross-Cultural Encounter and Dialogue, in Franco, E., Preisendanz, K. (eds.), Beyond Orientalism. The Work of Wilhelm Halbfass and its Impact on Indian and Cross-Cultural Studies (Poznan Studies in the Philosophy of the Sciences and the Humanities 59), Rodopi, Amsterdam, pp. 141-159. Hannerz, U. (1992), Cultural complexity. Studies in the social organization of meaning , Columbia University Press, New York. Houben J. (2002), Filosofia e filologia tra Oriente e Occidente, in Squarcini, F. (ed.), Verso l’India Oltre l’India. Scritti e ricerche sulle tradizioni intellettuali sudasiatiche, Mimesis, Milano, pp. 153-171. Marconi, D., Vattimo, G. (1986), Nota introduttiva, in Rorty, R., La filosofia e lo specchio della natura (Studi Bompiani), Bompiani, Milano, 1986, pp. vii-xxxii. Mohanty, J.N. (1997), Between Indology and Indian Philosophy, in Franco, E., Preisendanz, K. (eds.), Beyond Orientalism. The Work of Wilhelm Halbfass and its Impact on Indian and CrossCultural Studies (Poznan Studies in the Philosophy of the Sciences and the Humanities 59), Rodopi, Amsterdam, pp. 163-170. Popper, K.R., Eccles, J.C. (1997), The Self and its Brain, Springer international, Berlin. Rorty, R. (1979), Philosophy and the Mirror of Nature, Princeton University Press, Princeton (N.J.). ______ (1989), Contingency, Irony, and Solidarity, Cambridge University Press. ______ (2001), ‘L’essere, che può essere compreso è linguaggio’. Per HansGeorg Gadamer in occasione del suo centesimo compleanno, in Di Cesare, D. (ed.), “L’essere, che può essere compreso è linguaggio”. Omaggio a Hans-Georg Gadamer (Opuscola 113), Il melangolo, Genova, pp. 45-59. Saviani, C. (1998), L’Oriente di Heidegger (Opuscola 88), Il melangolo, Genova. Segre, C. (2001), Ritorno alla critica (Biblioteca Einaudi 110), Einaudi, Torino. Squarcini, F. (2001) La stagione ermeneutica aperta da ‘India and Europe’, unpublished. Staal, F. (2001), La scienza nella cultura indiana, in Storia della scienza. Cina, India Americhe, Istituto della Enciclopedia Italiana, Roma, vol. 2, pp. 610-638.

606

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Vattimo, G. (1981), Interpretazione, in Enciclopedia Garzanti di filosofia […], Garzanti, Milano, pp. 450-452. ______ (1985), La fine della modernità, Garzanti, Milano. ______ (1983), Introduzione, in H.-G. Gadamer, Verità e metodo (Studi Bompiani), Bompiani, Milano, pp. xiv-xv. ______ (1994), Oltre l’interpretazione. Il significato dell’ermeneutica per la filosofia (Economica Laterza 270), Laterza, Roma. ______ (2001), Interpretare il mondo è cambiare il mondo, in Di Cesare, D. (ed.), “L’essere, che può essere compreso è linguaggio”. Omaggio a Hans-Georg Gadamer (Opuscola 113), Il melangolo, Genova, pp. 60-67. Vergani, M. (2000), Jacques Derrida (Testi e pretesti), Bruno Mondadori, Milano. Wittgenstein, L. (1967) Zettel, edited by G.E.M. Anscombe and G.H. Von Wright, B. Blackwell, Oxford. ______ (1969) Über Gewissheit / On Certainty, edited by G.E.M. Anscombe and G.H. Von Wright, B. Blackwell, Oxford.

FERNANDO TOLA, CARMEN DRAGONETTI Unity in Diversity: Indian and Western Philosophical Traditions *

Since many years we have been interested in a problem that worries many scholars in Indology: whether philosophy existed in ancient India (Tola, Dragonetti 1983). It is our idea that philosophical thinking has been an important part of Indian tradition all along its history and that its origin can be traced back to the Vedic epoch. Of course we admit that in Vedic texts philosophical thinking still appears in a rudimentary form, as ‘preformations’ –we could say– that were to be developed in later centuries. We know that it is not easy to give a definition of the word ‘philosophy’, acceptable to all. The long series of articles concerning the notion of ‘Philosophie’, included in the Historisches Wörterbuch der Philosophie (v. 7, col. 572-926), shows the great number of opinions that have existed concerning this notion. We assume that philosophy basically is what in Indian technical terminology is called a dar†ana, a peculiar way of looking at the reality in which we live. This word is used to designate what Indians and indologists consider to be the Indian ‘systems of philosophy’. We shall examine later on the validity of the opinion that Western philosophy is characterized by rationality, free thinking and search of truth for its own sake, conceived as being the essential features of philosophical thinking. The existence of philosophy in Indian culture is in a general way explicitly denied by many professors of Western philosophy, Western philosophers and cultivated people in general, and implicitly by the manuals or treatises of History of Philosophy or Ancient Philosophy, which starting from Greek Philosophy, do not include in them Indian * We thank the National Agency for the Promotion of Science of Argentina for its support for the accomplishment of a Project on Indian Philosophy, a part of which is constituted by this contribution.

608

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Philosophy. They think, unconsciously or consciously following Hegel’s idea, that philosophy implies rational thinking, that rational thinking was absent from India and that consequently philosophy did not come forth in India but only in Greece. The debate between acceptance and denial of the existence of philosophy in Indian culture is not a minor question, since it determines the idea one has to adopt about Indian cultural tradition, and the position one has to take concerning many issues related to that tradition. For instance, if one adheres to the negative opinion about the existence of an Indian philosophical thought, one has to adhere also to the generally held, simple and untenable opinion that India possesed only religious irrationalistic explanations of reality (dar†ana) in face of a Western tradition that had constructed rationalistic explanations of reality (philosophical systems). This is one of the factors that created the myth of an ‘irrational India’ opposing a ‘rational West’. Together with the above indicated consequence, another consequence is originated if one accepts the non-existence of Indian philosophy: given the importance always attributed in the West to philosophy, considered as one of the most effective factors to build up the identity of a culture, and one of the most adequate intellectual activities to promote sane rational thinking, one has to conclude, if one affirms the absence of an Indian philosophy, that Indian culture, just because of the lack of that positive element, has always been an inferior culture in relation to Western culture, which is believed to be the unique and privileged possessor of that most valuable element. Thus, we think that in order to have a right notion of Indian cultural tradition, in itself and in its relation to Western cultural tradition, it is necessary to give due attention to the problem of the existence or non-existence of an Indian philosophy and to find a valid solution to it. It has always been our idea that the only way to demonstrate the existence of an Indian philosophy is to point out as many as possible Indian intellectual productions (ideas, theories, doctrines) that present similarities with Western intellectual productions of the same nature, traditionally considered as philosophical productions. The Indian philosophical productions are to be found for the most part in the Indian dar†anas. However, in the first stages of India culture such intelectual productions may also be found in texts that generally are not considered dar†anas of philosophical nature. The similarities presented by Indian and Western intellectual productions may refer to the subjects they develop, to the attitudes of their authors, to the nature of the postulates from which they start their reasonings or on which these reasonings are founded, to the methods used by the authors to reach their conclusions, and other similar factors of any intellectual labour. Two facts have helped us in our intent to carry out the demonstration of the existence of an Indian philosophy in the way just indicated (by similarities found in Indian and Western intellectual productions).

VI. Thinking about Traditions in South Asia Today

609

Firstly, during a good number of years we have been doing research work on Indian dar†anas, and secondly, during many years also we have been reading and studying the works of Western philosophers. This double activity, performed with the proposed aim, allowed us to find out many similarities of the indicated kind in the Indian dar†anas and in the Western systems of philosophy, and consequently to assume that the traditionally accepted opposition between Indian thought, as contained for instance in the dar†anas and many times labelled by Westerners as ‘irrational’, and Western philosophy, characterized by them as ‘rational’, is only a myth based on ignorance and eurocentric prejudice. As we shall see, the establishment of these similarities, have important consequences that go beyond the problem of the existence of philosophy in Indian cultural tradition. We shall develop in what follows our general thesis of the existence of an Indian Philosophy, under the form of four subsidiary theses. 1. First thesis Up to the 16th century at least, in India on the one hand and in Greece and Europe on the other, there was frequent reflection on the same philosophical subjects, and it was carried out in the same way. In his Prolegomena, Kant refers to some scholars (Es gibt Gelehrte […]) “[…]who think that nothing can be said, which has not been said before” ([…] kann nichts gesagt werden, was ihrer Meinung nach nicht schon sonst gesagt worden ist [p. 113, in Werke, Suhrkamp ed. = A 3, 4]) and immediately afterwards he expresses the idea that, as human understanding has reflected on innumerable subjects in many different ways for many centuries now, it is difficult not to find new ideas to which some old and similar idea does not correspond ([…] da der menschliche Verstand ber unzählige Gegenstände viele Jahrhunderte hindurch auf mancherlei Weise geschwärmt hat, so kann es nicht leicht fehlen dass nicht zu jedem Neuen etwas altes gefunden werden sollte was damit einige Ähnlichkeit hätte). In his treatise Über eine Entdeckung nach der alle neue Critik der reinen Vernunft durch eine ältere entbehrlich gemacht werden soll (vol. 5, p. 364 [Suhrkamp ed. = BA 110, 111]), Kant refers to the same subject. Kant rightly opposes the idea that nothing new can be said: it is possible for new ideas to arise which do not have as predecessors similar ideas. This is the condition sine qua non for the advancement of thought. We would like to add to Kant’s opinion that, just as it is not difficult to find new ideas to which some older, similar idea does not correspond, it is also not difficult to find many ideas which are supposed to be new but to which older, similar ideas do correspond, especially if we broaden the geographical and chronological field of investigation. This is the assumption that makes our research on the

610

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

parallelisms between Indian and Western thought possible, and that gives a much wider range to Indian and European thought: they cease to be ‘Oriental’ or ‘Western’ thought and instead become ‘universal’ thought. The confrontation of Indian and Western ideas, theories and doctrines implies a comparative activity. It is not necessary to carry out a detailed comparative study of Indian doctrines and Greek or Western doctrines which are evidently similar. Without going into details, it is sufficient to merely point out some of the Greek or Western doctrines which present similarities to Indian doctrines without taking into account the systems to which they belong. Moreover, we think that it is not possible to compare an Indian philosophical system as a whole with a Western system as a whole. In general, Indian systems and Western systems start from different postulates (such as re-existences and anåditva or beginninglessness on the Indian side as opposed to a unique existence and a first beginning in the Western systems), and this fact makes a comparison of both an impossible task. However, we think that it is possible to point out Greek or European doctrines which constitute elements of Greek or European systems and present similarities to Indian doctrines which are in turn elements of certain Indian systems. Helmuth von Glasenapp used to say that systems of thought are like large mosaic paintings each one of which represents a different scene: all of them have many small pieces of material (glass, tiles, etc.) of similar or identical colour and form in common. It is impossible to compare a mosaic painting as a whole with another as a whole, but it is possible to discover identical or similar pieces in each of them. The same thing happens with philosophical systems: as wholes they may be utterly diverse, but it is possible to find in each of them doctrines that can rightly be compared. This is the sort of comparison which we think is necessary when we examine whether there was such a thing as philosophy in India or not. A comparison, no matter how superficial, between Indian and Western doctrines, besides being necessary with regard to the problem of the existence of Indian philosophy, can help us to understand, accept and evaluate Indian thought under better conditions. This comparative procedure allows us to discover that Indian thought is not as remote from Western thought as is generally believed, since many things that were thought in India and may appear strange, exotic or even absurd to us were also thought in the West and enjoyed a profound acceptance for a long time. In the last years we have been dedicated to a research project on this special topic of the similarities between Indian and Western philosophical thought. In 2003, in a booklet entitled Sobre el mito de la oposición entre filosofía occidental y pensamiento de la India, El sistema filosófico Så∫khya: Dualismo Espíritu/Materia. Materialismo sui generis.

VI. Thinking about Traditions in South Asia Today

611

Evolucionismo. Ateísmo, and published by our institutions, Fundación Instituto de Estudios Budistas (FIEB), we made known the first results of this research work; in it we offered examples of similarities between subjects, ideas, concepts, doctrines of the orthodox Hinduist philosophical system Så∫khya and of the Western, Greek and European, philosophy. Other results have just appeared in March 2004, in a book of 293 pages, in English, published in Germany by Olms Verlag and entitled On the Myth of the Opposition between Indian Thougt and Western Philosophy. In this book we present a great number of similarities between ideas, theories and doctrines expressed, on one side, in the Vedas, the Upanißads and the Så∫khya system and, on the other side, by Western (Greek or European) authors. In this book we offer the original Sanskrit texts containing the Indian ideas, theories, and doctrines with our English translation, and then the corresponding Western texts, Greek, Latin, German, English, French, Italian, with our own English translation, in order that readers may have by themselves a direct access, and in a way that leaves not place to any doubt, to the amazing and irrefutable similarities we adduce. We give now an example of the similarities between Indian and Western thought that are revealed by a study of this kind: The Så∫khya theory of causality, the so called satkåryavåda which maintains that the effect pre-exists in its cause before its manifestation has a perfect correspondence with the ideas exposed and accepted by Hegel (see Wissenschaft der Logik, part 1, book 2, section 3, third chapter, pp. 223-228 [in Werke, Suhrkamp ed.]), by St. Thomas (Summa Theologiae, I, 2.19.5), and finally by Leibniz (see Catena mirabilium demonstrationum de summa rerum, in Philosophische Schriften, vol. I, p. 6 [Insel Verlag ed.]). Other similarities of the same nature –shown in detail by us in our mentioned English book– are as follow: ‘doctrines’ expressed in the Vedas and, in the West, by Orphics and Neoplatonists (i.e. in myths of creation); Plotinus (i.e. the One as the origin of everything); Goethe, Saint Augustine and Leibniz (i.e. the exaltation of the effects of action in the human destiny); the Stoics, the poet-philosopher Manilius, Guillaume de Conches, Honorius d’Autun, and Arnaud de Bonneval (i.e. cosmic order); or ‘doctrines’ expressed in the Upanißad and, in the West, by the Stoics and Descartes (i.e. the soul as a blow of air); the Pre-socratics, Aristotle, Marcus Aurelius, Scotus Erigena (i.e. the notion of arché or the principle, origin, foundation and end of everything); Descartes and Spinoza (i.e. the notion of substance and the relation of the spirit with the mental functions); Plato, Aristotle, Manilius, Petrus Abelardus, Adelard of Bath, Saint Augustine, Saint Basile, Cornelius Agrippa, Giordano Bruno, Leibniz, and Schelling (the anima mundo or ‘soul of the world’); Spinoza (the abstract-impersonal notion of the Supreme Principle); Saint Augustine, Raimundus Lullius and Hegel (the triadic structures); the Sto-

612

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

ics, Plotinus, Synesius of Cyrene and Proclus (the correspondences between the macrocosm and the microcosm); or expressed in the Så∫khya system and, in the West, by Kant, Wolf, the Manichaeans (i.e. dualism); Melissus, Empedocles and Lucretius (i.e. the principle ex nihilo nihil); Hegel, Saint Thomas and Leibniz (i.e. the conception of the identity of the cause and the effect); d’Holbach, Schelling, Leibniz and Bonnet (i.e. the conception of matter and of its constituent elementos); Anaxagoras and Schelling (i.e. the constitution and nature of matter); Heraclitus, the Stoics, Aristotle, Origenes, Leibniz, J. W. Petersen (i.e. the theory of the eternal return); Aristotle and Saint Thomas (i.e. the proofs of the existence of a transcendent entity); Descartes and Leibniz (i.e. the relation between the spirit, soul, and matter, body); Orphics, Pythagoras, Empedocles, Plato, and many other, ancient as well as modern authors (i.e. the belief in transmigration); Leibniz (i.e. the notion of subtle body, the problem of the existence of evil in the world); d’Holbach (atheism). The above example refers to similarities in relation to subjects of the philosophical reflection. Now we shall point out a similarity in relation to the method frequently adopted by Indian and Western thinkers, which strongly reduces in both cases the credibility of their conclusions. In Indian Brahmanical or Hinduist tradition, for instance, there exist beliefs that come from the past, it would be better to say: that are imposed by the past, such as the belief in reincarnations, in the existence of an œ†vara (Lord, God), in the infallibility of the †ruti (‘revelation’). These beliefs are, unconsciously taken for granted by a great number of thinkers; they are for them indisputable assumptions that do not need to be demonstrated. They can be called ‘cultural dogmas’ (similar to the other dogmas that are found in other Indian traditions). These beliefs are based on faith rather than on the observation and establishment of facts or on valid rational arguments. The force they possess is incomparably stronger than the force of the arguments that some times are adduced to support them. It could be said that the fact of being based on faith gives them more strength than if they were based on logical argumentation. We do not discuss or call in question the value of faith as a foundation for a belief, but we think that what cannot be denied is that an act of faith cannot be considered as a rationalistic mental process. Many Brahmanical or Hindu thinkers used these ‘cultural dogmas’ as starting point, basis or postulate to construct upon them their doctrinaire systems. It is necessary to remark here that, even if Indian thinkers used some of their ‘cultural dogmas’, to which authority is attributed a priori (a procedure which could be considered as an example of nonrational behaviour), as a starting point, basis or postulate to construct their doctrinaire (philosophical) systems, nevertheless these same thinkers in their argumentations submit themselves to the strictest

VI. Thinking about Traditions in South Asia Today

613

requirements of reason and logic. We can question the validity of their ‘cultural dogmas’, but we cannot find logical defects in what they construct on the basis of these dogmas: their systems of thought. These are true constructions of granite and steel upon clay foundations. We must acknowledge the fact that the same opinion can be expressed with regard to Western thinkers (we could even say with regard to thinkers belonging to any culture): although following the most rational and logical requirements in their reasonings, they construct their theories on frail bases or postulates, which are nothing other than their own ‘cultural dogmas’. Among these Western ‘cultural dogmas’ are the belief in God, the belief in an immortal soul, the authority of the Christian texts, the infallibility of the Bible, etc. Descartes, for instance, referring to himself in the preface to his Discours de la méthode, clearly says: “the existence of God and of the human soul are […] the foundations of his Metaphysics” (l’ existence de Dieu et de l’ame humaine, […] sont les fondaments de sa Métaphysique [see Oeuvres de Descartes, Ch. Adam and Paul Tannery, vol. 6, p. 1]). Also, in the Sexta Responsio to the objections made against his Meditationes de prima philosophia, he affirms: “[…] certainly it is necessary to begin with the knowledge of God and thereafter the knowledges of all the other things have to be subordinated to that sole [knowledge]” ([…] nempe incipiendum est a Dei cognitione ac deinde aliarum omnium rerum cognitiones huic uni sunt subordinandae [see Oeuvres de Descartes, vol. 7, pp. 429-430]). And in his Epistola, which precedes his Meditationes de prima philosophia, in quibus Dei existentia, et animae humanae à corpore distinctio, in a more forceful way he says: “[...] Then although for us who are believers (fideles) it is sufficient to believe by faith that the human soul does not perish with the body and that God exists [...] And although it is absolutely true that it is necessary to believe in the existence of God, because this has been taught in the Holy Scriptures, and vice versa it is necessary to believe in the Holy Scriptures, because they come from God […]” ([...] nam quamvis nobis fidelibus animam humanam cum corpore non interire, Deumque existere, fide credere sufficiat [...] Et quamvis omnino verum sit, Dei existentiam credendam esse, quoniam in sacris scripturis docetur, et vice versa credendas sacras scripturas, quoniam habentur a Deo […] [see Oeuvres de Descartes, vol. 7, p. 1]).1 From these texts of Descartes emerges the important function, which corresponds to faith and the Holy Scriptures, i.e. Revelation, in relation to the subjects of the existence of God and the immortality of the soul that are for Descartes –and for many Western thinkers– the fundamental subjects of philosophy. 1 In his French version: “[...] Car bien qu’il nous suffise à nous autres qui sommes fideles, de croire par la Foy qu’il y a vn Dieu, et que l’ame humaine ne meurt point auec le corps, [...] Et quoy qu’il soit absolument vray, qu’il faut croire qu’il y a vn Dieu, parce qu’il est ainsi enseigné dans les Saintes Escritures, et d’autre part qu’il faut croire les Saintes Escritures, parce qu’elles viennent de Dieu […]” (see Oeuvres de Descartes, vol. 9, p. 4).

614

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

2. Second thesis In the history of Greek and European philosophies, we can find what is usually (correctly or incorrectly) called ‘manifestations of irrationality’ in many forms and these are as numerous as they are in the history of Indian thought. As a consequence of the comparative study of Indian dar†anas and Western systems of philosophy, it is possible to reach the conclusion expressed in the above indicated thesis, not only concerning the centuries corresponding to the ancient (Greek) and mediaeval (European) periods, but also the centuries that came after. Greece and Europe have never been characterized by rationality only, and India has never wholly submitted to irrationality. In both regions of the world, we can find the same mixture of a dominant irrationality and a limited rationality, which manifested itself only timidly and was not a predominant or excluding factor at all. The birth of Pallas Athena from the head of Zeus has never been an easy or rapid process. Many ideas, theories and doctrines both in Indian dar†anas and Western systems are based on mere beliefs or are conclusions, logically deduced from mere beliefs. As we have said, without discussing or calling in question the value possessed by these beliefs, what must be admitted is that such a kind of ideas, theories and doctrines are not of a rationalistic nature, on account of the nature of the foundation that supports them. In the Upanißads, for instance, are found many texts that affirm that the world is a triad and that everything in it is constituted by three elements or components. In the R¢gveda and Bråhmañas –texts which precede the Upanißads chronologically– many examples of triadic groups are also found. The triadic concept of reality which manifests itself in the just mentioned texts was to be developed a few centuries later by the Så∫khya philosophical system in its theory of the three guñas, which are conceived of as the components of Matter. Matter, by virtue of its evolution, gives rise to the entire material and psychic empirical reality. The guñas as constituents of Matter are present in everything and the diversity of beings and things depends on the diversity of the proportions in which they are mixed together in beings and things. This triadic interpretation of reality has a very old archaic origin, derived from the peculiar and previleged nature attributed to certain numbers. In the West we find many instances of a similar triadic conception of reality. The Trinity doctrine that affirms that God is One and Three, is one of the most important Christian beliefs. It is difficult to understand, to explain, and of course also to demonstrate it: it is a ‘dogma of faith’. Notwithstanding the belief nature of the Trinity dogma, it has served as an order principle in European culture. St. Augustine wrote his treatise De Trinitate trying to establish that the

VI. Thinking about Traditions in South Asia Today

615

triadic structure of the Trinity is found in the mental processes that take place in man. Raimundus Lullus expounds in his Nova Logica a completely Trinitarian concept of the world endeavoring to discover in creation a reflex and an image of the Christian Trinity, giving universality to the triadic model, and finding triadic structures in all the aspects of reality. And Hegel and many other Western philosophers were fond of triadic structures in their philosophical constructions (Piclin 1980). We must add, as another kind of non-rationalistic ideas, theories or doctrines, those that do not have the support of observed real facts, that are, when they are examined, nothing else than mere fanciful mental creations. In India and in the West are found many intellectual products of this kind that are seriously taken as philosophical doctrines. The union of matter (human body) and spirit (soul) was a problem both for Indian and Western thinkers, and both imagined fanciful solutions for it –even extravagant solutions. The Så∫khya system solved the problem having resource to a comparison: the union of matter and spirit consists in that the spirit reflects itself in matter –let us add: as the light of a lamp on any object. It is nothing else than a comparison, which does not establish anything. In the West the union of matter and spirit received several solutions. Descartes thought that the soul moves the pineal gland and this in turn propels the animal spirits towards that part of the body that the soul wants to move by means of them. Descartes’ disciples were not satisfied with this explanation and proposed the system of the ‘occasional causes’, adopted also by Malebranche, according to which it is God himself who moves that part of the body that the soul wishes to move. Leibniz, who does not accept that system, thinks that it would force God to accomplish ‘moving’ miracles all the time or to count on the assistance of angels to help Him to move all the limbs that are moved in the world. The same Leibniz proposed a new system under the beautiful name of ‘pre-established harmony’. According to this system God creates two automates (today we would say two robots): one is the soul, the other one is the body. These two automates (or robots) are programmed in their smallest details in such a way that, when a certain movement is produced in one, another movement occurs in the other which exactly corresponds with the previous one. In the case of the two robots and in that of the two clocks carefully coordinated (another example used by Leibniz), there are physical movements only; in the case of the soul and the body, there is a psychic ‘movement’ in the soul, and a physical ‘movement’ in the body. Really human fancy has no limits even within the realm of the sternest philosophy. Besides rationality, freedom of thought and the search of truth for its own sake are considered as essential attributes of philosophy, at least of the Western one.

616

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

Our idea is that philosophical thought has generally been subjected to many limitations, both in India and the West. Only in very few instances have Indian and Western thinkers been able to rid themselves of ‘beliefs’ imposed on them by the past. Thinking must arise in agreement with them and be subordinated to them, and consequently it cannot be considered ‘free’. It is why a great part of Western philosophy, as being founded on Christian dogmas, and as not daring to abandon the limits that these dogmas imposed on it, is nothing else than ‘Christian philosophy’, in the same way as the majority of Indian dar†anas are tightly bound to any of the great Indian religions, specially Hinduism. Nevertheless, we also think that, within the limits imposed by cultural dogmas, intellectual freedom existed to a much greater degree in India than it did in the modern European world. Tolerance was not a specific characteristic of Christian Europe. We base this assumption in relation to Indian intellectual freedom on facts such as the following ones: the acceptance by Hinduism, as orthodox systems of thought, of systems which present large and obvious differences amongst themselves; the acceptance of different Vedånta schools with great oppositions amongst themselves; the fact that, although power in India has generally been held by people adhering to the Hinduist dar†ana (philosophical and religious point of view), violence has never been used against adherents of other dar†anas; and the fact that the appearance of the Buddhist dar†ana, which negated the very foundations of Hinduist society, religion and philosophy, did not lead to bloody religious wars or ideological mass persecutions of any kind. As for the idea that philosophical thinking is a search for ‘truth for its own sake’, let us say here that this has been of very limited application in the history of Western philosophy. If the criterion of ‘truth for its own sake’ were applied strictly, many schools of philosophy would be eliminated from the history of philosophy. Examples from Greece would be Stoicism and Epicureism, whose efforts were directed towards finding the correct way of living in order to enjoy peace and happiness; an example from mediaeval Europe would be Christian philosophy with its mainly religious preoccupations. We would also reach the same conclusion if we were to analyze the works of many of the great modern European philosophers, whose real aim was often only to demonstrate a certain religious thesis via their philosophical reasoning. Leibniz composed his Théodicée with the aim of demonstrating God’s justice and goodness. At the beginning of his treatise Causa Dei, Leibniz clearly indicates his purpose when writing his Théodicée: “The apologetic treatise of the Cause of God concerns not only the divine glory, but also our own advantage, in order that we worship His greatness, i.e. His power and wisdom, and also in order that we love His goodness and whatever derives from it, His justice and sanctity, and that

VI. Thinking about Traditions in South Asia Today

617

we imitate them as much as it is possible to us” (Apologetica Causae Dei tractatio non tantum ad divinam gloriam, sed etiam ad nostram utilitatem pertinet, ut tum magnitudinem ejus, id est potentiam sapientiamque colamus, tum etiam bonitatem et quae ex ea derivantur, justitiam ac sanctitatem amemus, quantumque in nobis est imitemur). Western philosophy never forgot that it had been during many centuries the ancilla theologiae and that as such it had to follow and obey its domina and to have always in mind her interests and preoccupations; the sa∫skåras (impressions, habits or predispositions) left by this condition survived in it for a time longer than it is usually admitted. As for Indian philosophers, they were clearly aware that their activity as such had an aim external to that activity: the attainment of the supreme good, i.e. liberation from reincarnations. They were more conscious of the links of their philosophical activity with religion. 3. Third thesis Such a thing as philosophy did exist in India. If, in the history of Indian thought and of Greek and European philosophies, we can find similar subjects, approaches and solutions as well as a similar coexistence of irrationality, submission to authority and subordination of philosophical thinking to other aims in conjunction with the opposed attitudes, we are authorized to affirm that philosophy did exist in India. This thesis is a consequence derived from the similarities between many ideas, theories, concepts and doctrines found in the Indian dar†anas and in the Western systems of philosophy. Another consequence from these similarities is the elimination of the myth of the opposition between Indian thought and Western philosophy. Finally, these similarities provide us with a firm basis for new approaches and perspectives in the study, understanding and evaluation of philosophical traditions in general. 4. Fourth thesis The comparison between Indian and Western thought must limit itself to confronting both as they manifested themselves before the 16th century, or even in the centuries that followed, but in this case only when and if they maintained certain forms of the philosophies found previous to that date, prolonging them. In what precedes we have referred several times to the comparison between Indian and Western philosophical ideas, theories and doctrines. This comparison must have a terminus ante quem: the 16th century A.D. This date has not been arbitrarily chosen. From the 16th century onwards, a series of factors appeared in the West, as the coming forth of modern science and the modern scientific mind, the discovery of the New World, the increase of European economic and military power, the weakening of ecclesiastical author-

618

BOUNDARIES, DYNAMICS AND CONSTRUCTION OF TRADITIONS IN SOUTH ASIA

ity and the limitations it imposed on thinking, and the consciousness of the equality of individual rights and liberties. These factors gave a new course to universal history and led to modern culture. From the 16th century onwards, Western culture in all its expressions began to adopt a wholly novel form, which was different from all previously known forms, many times extraordinarily valuable, and which succeeded in imposing itself worldwide to differing degrees. India has only taken part in this transformation in a profound way since the middle of the 20th century. To compare Indian thought before the 16th century with Western thought after that date would be to compare two things, which belong to two completely incommensurable epochs as a result of the intrusion of the factors indicated. We can compare Indian and Western thought under these chronological conditions. In this way we can inquire as to what they have in common, what they have which differs, and in which respect one of them stands out in relation to the other; we can inquire whether India anticipated certain philosophical theories (for instance in the field of epistemology and of idealism) and was able to adopt attitudes (for instance with regard to freedom of thought and tolerance), which did not appear in the West until much later. This way of going about things will allow us to give a fairer and wiser answer to the question as to whether there is such a thing as Indian philosophy. The great emphasis we put in the similarities between Indian and Western philosophy points at the ‘unity’ of both cultures, unity that must not surprise, since both have a common past: the mysterious Indo-european people in whose culture the West as well as India are rooted, but this unity also must not make us forget the ‘diversity’ with which both cultures manifested themselves in history, developing each one of them their proper characteristics, their peculiar individuality, their identity. Coexistence of ‘unity’ in ‘diversity’ –as when the same blow of air entering into the different pipes of an organ makes them resound in a variety of musical effects.

BIBLIOGRAPHY Piclin, M. (1980), Les philosophes de la triade ou l’historie de la structure ternaire, Vrin, Paris. Tola, F., Dragonetti, C. (1983), Filosofía de la India, in Id., Filosofía y Literatura de la India, Editorial Kier, Buenos Aires, pp. 65-71. Tola, F., Dragonetti, C. (2004), On the Myth of the Opposition between Indian Thought and Western Philosophy, Olms Verlag, Hildesheim.