Dialogue and Doxography in Indian Philosophy Points of View in Buddhist, Jaina, and Advaita Vedānta Traditions [1 ed.] 9780429275982, 9780367226138

618 218 2MB

English Pages [222] Year 2020

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Dialogue and Doxography in Indian Philosophy Points of View in Buddhist, Jaina, and Advaita Vedānta Traditions [1 ed.]
 9780429275982, 9780367226138

Table of contents :
Acknowledgements ix Preface xi
Introduction 1
I.1 The form of doxography: systematic, dialectical, ahistorical 2
I.2 A universal definition of doxography 13
I.3 The content of Indian doxography: views 14
I.4 The function of Indian doxography: spiritual/
yogic exercise 18
I.5 The context of Indian doxography: a dialogical path
to knowledge 20
Notes 22
1 The beginnings of Mādhyamika doxography: Bhāviveka’s MHK 25
1.1 The context of the MHK 25
1.2 The doxographical content and discourse of the MHK 32 1.3 The doxographical structure of the MHK 38
1.4 Table of contents of the MHK 43
1.5 The dialectical strategy of Chapter 9 48
1.6 The rhetorical and thematic progression of Chapter 9 54 1.7 Irony in Chapter 9: peace by polemics 71
Notes 76
2 The beginnings of Jaina doxography: Haribhadra’s ṢDS 91
2.1 Haribhadra’s doxography in context 91 2.2 Haribhadra’s ṢDS in context 103
2.3 ṢDS Chapter 6: the Jaiminīya 125 Notes 126
viii Contents
3 The beginnings of Advaita doxography: Śaṅkara’s SSS 133
3.1 Advaita doxography in context 133
3.2 The hierarchical context of Advaita doxography:
above the above 142
3.3 Reconsidering the SSS: authorship, pedagogy,
and doxography 155
3.4 SSS Chapters 7 and 8: review and reflection 166 Notes 169
Conclusion 178 Notes 183
Bibliography 184 Index 200

Citation preview

Dialogue and Doxography in Indian Philosophy

This is the first book fully dedicated to Indian philosophical doxography. It examines the function such dialectical texts were intended to serve in the intellectual and religious life of their public. It looks at Indian doxography both as a witness of inter- and intra-sectarian dialogues and as a religious phenomenon. It argues that doxographies represent dialectical exercises, indicative of a peculiar religious attitude to plurality, and locate these ‘exercises’ within a known form of ‘yoga’ dedicated to the cultivation of ‘knowledge’ or ‘gnosis’ (jñāna). Concretely, the book presents a critical examination of three Sanskrit doxographies: the Madhyamakahṛdayakārikā of the Buddhist Bhāviveka, the Ṣaḍdarśanasamuccaya of the Jain Haribhadra, and the Sarvasiddhāntasaṅgraha attributed to the Advaitin Śaṅkara, focusing on each of their respective presentations of the Mīmāṃsā view. It is the first time that the genre of doxography is considered beyond its literary format to ponder its performative dimension, as a spiritual exercise. Theoretically broad, the book reaches out to academics in religious studies, Indian philosophy, Indology, and classical studies. Karl-Stéphan Bouthillette is currently an FWO post-doctoral researcher in Ghent, Belgium. He received his PhD at the Institute for Indology and Tibetology at Ludwig-Maximilians-University, Germany.

Dialogues in South Asian Traditions: Religion, Philosophy, Literature and History Series Editors: Laurie Patton Middlebury College, USA

Brian Black

Lancaster University, UK

Chakravarthi Ram-Prasad Lancaster University, UK

Face-to-face conversation and dialogue are defining features of South Asian traditional texts, rituals, and practices. Not only has the region of South Asia always consisted of a multiplicity of peoples and cultures in communication with each other, but also performed and written dialogues have been indelible features within the religions of South Asia; Hinduism, Buddhism, Jainism, Sikhism, and Islam are all multi-vocal religions. Their doctrines, practices, and institutions have never had only one voice of authority, and dialogue has been a shared tactic for negotiating contesting interpretations within each tradition. This series examines the use of the dialogical genre in South Asian religious and cultural traditions. Historical inquiries into the plurality of religious identity in South Asia, particularly when constructed by the dialogical genre, are crucial in an age when, as Amartya Sen has recently observed, singular identities seem to hold more destructive sway than multiple ones. This series approaches dialogue in its widest sense, including discussion, debate, argument, conversation, communication, confrontation, and negotiation. Opening up a dynamic historical and literary mode of analysis, which assumes the plural dimensions of religious identities and communities from the start, this series challenges many outdated assumptions and representations of South Asian religions. Shared Characters in Jain, Buddhist and Hindu Narrative Gods, Kings and Other Heroes Naomi Appleton In Dialogue with Classical Indian Traditions Encounter, Transformation and Interpretation Edited by Brian Black and Chakravarthi Ram-Prasad Dialogue and Doxography in Indian Philosophy Points of View in Buddhist, Jaina, and Advaita Vedānta Traditions Karl-Stéphan Bouthillette For more information and a full list of titles in the series, please visit: www.routledge.com/Dialogues-in-South-Asian-Traditions-Religion-PhilosophyLiterature-and-History/book-series/ASTHASIAREL

Dialogue and Doxography in Indian Philosophy Points of View in Buddhist, Jaina, and Advaita Vedānta Traditions Karl-Stéphan Bouthillette

First published 2020 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 52 Vanderbilt Avenue, New York, NY 10017 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2020 Karl-Stéphan Bouthillette The right of Karl-Stéphan Bouthillette to be identified as author of this work has been asserted by him in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. 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: 978-0-367-22613-8 (hbk) ISBN: 978-0-429-27598-2 (ebk) Typeset in Times New Roman by Apex CoVantage, LLC

À trois joyaux, Renaud Paquette, André Couture, et Louis Painchaud.

Contents

Acknowledgements Preface Introduction

ix xi 1

I.1 The form of doxography: systematic, dialectical, ahistorical  2 I.2  A universal definition of doxography  13 I.3  The content of Indian doxography: views  14 I.4 The function of Indian doxography: spiritual/ yogic exercise  18 I.5 The context of Indian doxography: a dialogical path to knowledge  20 Notes  22 1

The beginnings of Mādhyamika doxography: Bhāviveka’s MHK

25

1.1  The context of the MHK  25 1.2  The doxographical content and discourse of the MHK  32 1.3  The doxographical structure of the MHK  38 1.4  Table of contents of the MHK  43 1.5  The dialectical strategy of Chapter 9  48 1.6  The rhetorical and thematic progression of Chapter 9  54 1.7  Irony in Chapter 9: peace by polemics  71 Notes  76 2

The beginnings of Jaina doxography: Haribhadra’s ṢDS 2.1  Haribhadra’s doxography in context  91 2.2  Haribhadra’s ṢDS in context  103 2.3  ṢDS Chapter 6: the Jaiminīya  125 Notes  126

91

viii  Contents 3

The beginnings of Advaita doxography: Śaṅkara’s SSS

133

3.1  Advaita doxography in context  133 3.2 The hierarchical context of Advaita doxography: above the above  142 3.3 Reconsidering the SSS: authorship, pedagogy, and doxography  155 3.4  SSS Chapters 7 and 8: review and reflection  166 Notes  169 Conclusion Notes  183 Bibliography Index

178

184 200

Acknowledgements

The completion of this project would not have been possible without the financial support of Distant Worlds: Munich Graduate School for Ancient Studies and the frequent funding from the Munich Promovierendenförderung GCCW. To my colleagues and the support staff at Distant Worlds, I can’t express my gratitude enough. The book edition itself was kindly supported by the Jan Gonda Foundation of the Royal Netherlands Academy of Arts and Sciences (KNAW), administered by the International Institute for Asian Studies (IIAS). I am grateful for the time I spent in beautiful Leiden and to the people there who supported me. A word of thanks to Brian Black, editor of the present series, Dialogues in South Asian Traditions: Religion, Philosophy, Literature and History, for inviting me to publish this book in his collection, and for the brilliant conversations we exchanged on precious occasions. Surely, the result would not have been the same without the kind supervision of my Doktorvater, Professor Robert Zydenbos, who was the first to believe in the project and with whom I had numerous and meaningful discussions on the subject. Equally instrumental was the supervision of Professor Jens-Uwe Hartmann, who generously accepted to take me under his wing during this challenging journey. Thank you. Another word of thanks to the esteemed scholars who either exchanged articles with me, made meaningful recommendations, conveyed encouragements, and/ or commented on some version of the preliminary drafts. I especially think here of Vincent Eltschinger, who kindly reviewed a portion of my thesis manuscript. I would like to express my gratitude to Daniel Stuart, who commented on several sections of the same. I am particularly indebted to Philipp Maas for having read and kindly helping to improve the translation of and reflection on the materials which I present. I also think of Douglas Berger for his generous support and critical contributions. A word of gratitude to Christophe Vielle, Matthew Kapstein, Anil Mudra, Piotr Balcerowicz, Olle Qvarnström, Chakravarthi Ram-Prasad, and Johannes Bronkhorst for their severe critique, their friendly encouragements, and the precious information which they shared with me. It has been such a long journey that most of them undoubtedly forgot what kindness informed their responses to my queries. But I did not. Thank you.

x  Acknowledgements Within a closer circle, to my friends and colleagues who shared their views and tolerated my rambling on this project and its theoretical implications, Patrick McCartney, Henry Albery, Agnieszka Rostalska, Peter Romaskiewicz, Charles DiSimone, Guttorm Norberg Gunderson, Constanze Pabst von Ohain, and AnnaMaria von Parseval, a heartfelt word of gratitude. May we meet more often. Thank you. Naturally, the eternal support of my family, father and mother, but especially the love, patience, and precious insights of my wife and dearest friend, Laura von Ostrowski, ensured that my heart and mind remained on course throughout this dialogical venture. I thank you all.

Preface

This book is a revised version of my doctoral thesis, defended in Munich in November 2018. Several changes have been brought to the original manuscript in order to give more coherence and concision to its exposé. The result is a dense and multilayered analysis of doxography in general and of three early Indian doxographies in particular. It should be noted that, at the unfortunate risk of potentially alienating the more philologically driven colleagues, I decided not to include my translations of the doxographical chapters on which rests my analysis. Whenever meaningful, I comment on my reading of verses, in notes, in comparison with existing translations. I believe that scholars will nonetheless find scope for serious critical discussion within the present book. My decision was motivated by various considerations. First, translations of these texts are already accessible in book format. I do believe that it would be meaningful to refresh them, as some have not aged well. But, then, the translation of a single chapter of each text would be of little help. What is rather needed, I suggest, and this is particularly the case for Bhāviveka’s Madhyamakahṛdayakārikā (MHK), is a complete translation of the entire work, collecting in one place the resources put together by the many scholars who worked on different portions of the text, along with the available editions and commentaries existing in various Asian languages: Sanskrit, Tibetan, and Chinese. So far, no translation of the entire MHK exists in any European language. In my opinion, the best format for this enterprise would be one similar to the excellent Bibliotheca Polyglotta (BP), hosted online and edited by the University of Oslo’s Norwegian Institute of Palaeography and Historical Philology. Having already discussed the idea with one of its chief editors, Professor Jens Braarvig, I know that this project is an old dream, and I can only wish for its careful undertaking and due funding. Perhaps a comparable database could be dedicated to regrouping all known Indian philosophical doxographies. This would undoubtedly facilitate their study. In any case, adding a new book translating single chapters of these known texts, in itself, would be a marginal contribution. I thus chose to concentrate my efforts on what I believe to be most important at this point. It is time for Indian doxographical literature to come out of the shadow. To be analyzed in its multiple dimensions. Its odd heterogeneous composition reflects a

xii  Preface complex dialogical phenomenon. By stressing the importance of the genre, for our understanding of the practice of philosophy in India, regarding both its conceptual developments and its sociohistorical context, I believe that the present book better serves the cause of promoting research in doxographical literature, potentially paving the way to new textual editions. My insistence on the performative function of such texts, which I also read as spiritual exercises, rather seeks to open new analytical and comparative avenues, in tune with contemporary trends in philosophy and religious studies. This new perspective on doxography will rightfully be criticized. Yet, in the process, if its refreshing view on a misunderstood philosophical material succeeds in stimulating a renewed interest in doxography, it will have achieved its main goal. Studies in doxography, in my view, require one to consider the multidimensional networks of exchange in which evolved competing philosophical practices, complex styles of interrelated hermeneutics, and colourful and powerful mythological discourses, claiming the best of ancient India: the best of existence. Doxographies give voice to the sectarian ideologies underpinning the logic of doctrinal identity in Indian philosophy. What may appear as mere metaphysical lists of doctrinal constituents, enshrined in long-forgotten lore, had a life of its own, through customary rituals of competing sects, and developed an ultimate ideal of ‘freedom’ which continues to transform our world, in thoughts and deeds. Doxographies open a rare fold. They shed light onto an obscure world of elite scholasticism. Though elitist, scholars are seldom mighty. Some of their teachings may yet survive longer than fame. This book is the analysis of three significant models of doxographical compositions found in medieval India. I argue that doxography is a dialogical literature, and I engage the genre in a sustained dialogue with various streams of scholarship, likely to interest a broad range of scholars. I particularly invite colleagues of other cultural areas, in classical studies for example, who are well aware of the richness of doxographical literature, but who might not know of the various dialectical forms taken by doxography in India. Because I rather seek to reopen an old dialogue, this book does not claim to close the discussion on Indian doxography. It is but an invitation.

Introduction

The present study is a critical examination of three Sanskrit doxographies: the Madhyamakahṛdayakārikā (MHK) of the Buddhist Bhāviveka, the Ṣaḍdarśanasamuccaya (ṢDS) of the Jain Haribhadra, and the Sarvasidd­ hāntasaṅgraha (SSS) attributed to the Advaitin Śaṅkara. Based on the Mīmāṃsā chapter of these three early examples, my analysis reflects on the nature and purpose of Indian doxography in general, a literary genre of philosophy which has received little attention in scholarly circles. Of primary interest are the dialectical and religious attitudes adopted by doxographers while engaging with the plurality of views prevalent in the dialogic world of their days. If the whole content of this book were to be summarized into one sentence, at the risk of looking trivial, it should be the following: there is more to Indian philosophical doxography than a mere listing or summary of competing views. The nature of that ‘more,’ however, pertaining to soteriology and sectarian rivalries, can hardly be stated in one go. It is the purpose of this book to examine it. As the following pages unfold, the reader is led to consider how, in order to produce a general account of the Indian doxographical genre, one needs to bring together numerous hermeneutical threads rooted in various scholarly disciplines and to browse through a history crossing several time periods and sectarian divisions. To begin with, the development of doxography in India is intimately related to the process of philosophical systematization already well under way by the fifth century CE, itself a phenomenon inseparable from a series of socioreligious changes slowly transforming the cultural and political landscape of the subcontinent; at least until Buddhism is expelled from India around the twelfth century CE. None of these developments are fully accounted for in scholarly literature. The whole picture remains to be painted, and the present work seeks to contribute to its refinement. But what exactly is ‘doxography’? The definition of the genre is not as obvious as it may seem and requires some explanation. It should be known, at the outset, that the literary genre of doxography has initially been framed and studied within Western classical philosophy. Although their materials apply to an altogether different cultural environment, the theories developed by classicists are of significant value when it comes to reflect on the doxographical production of India. They have

2  Introduction surely been instrumental in producing the definition of doxography entertained within these pages. If, throughout this book, I maintain that the systematic classifications of views found in Indian doxographies is meaningful in the context of a particular religious training, involving the student in a transformative dialectical study designed to help one discriminate genuine truth from its numerous ersatz, it is after having examined and criticized the informed opinions of classicists and Indologists on the subject. It is thus necessary, in the present introduction, to briefly review this scholarship, so as to clearly identify three critical features of doxography upon which rests my interpretation of the nature and function of the genre: namely, (1) the systematic, (2) the dialectical, and (3) the ahistorical forms of doxography.

I.1 The form of doxography: systematic, dialectical, ahistorical I.1.1  Doxography in classical studies I.1.1.1  Birth of the concept Before its recent appearance in Indological studies, the term ‘doxography’ was brought to life and raised among the classicists. Under their meticulous care, it acquired an identity of its own and prolific offspring. Once transplanted into Indian soil, as a fragile offshoot into an ancient tree, it carried with it a range of newly formed assumptions which influenced the perception of the related Indian texts. To better understand these developments, the present discussion will formally introduce the little-known literary genre of philosophical doxography. The term ‘doxography’ is a neologism. It is a substantive derived from the Latin doxographi, a noun in plural number which could be translated as “tenet, doctrine, dogma, opinion, or view-writers.” Jaap Mansfeld, leading scholar of classical doxography, observes that “[t]he terms for tenets, or views, in ancient Greek are doxai or dogmata, in Latin opiniones; those for doctrines are, in Greek, areskonta, translated into Latin as placita. But note that these designations were used interchangeably.”1 Thus, before the term ‘doxography,’ Latin was used to coin doxographi, designating ‘doxographers,’ a singular group of philosophical writers involved in the compilation of informed opinions. It is the German classical philologist Hermann Alexander Diels, in his 1879 authoritative Doxographi Graeci,2 who was the first to coin the term ‘doxographi,’ meant to describe a specific type of ‘litterateurs.’ Diels’s research strictly focused on philosophical sources dedicated to physical speculation, or “philosophy of nature.”3 It includes principles, theology, cosmology, astronomy, meteorology, biology, and aspects of medicine. Diels’s research insight came from the hypothesis of his Bonn Doktorvater, Hermann Usener, about linguistic similarities in reports of pre-Socratic and other philosophical doctrines contained in two late compendia, the Placita (Tenets) attested in the manuscripts of Plutarch, and the Eclogae of Stobaeus. Diels modified the initial theory of his Doktorvater to make it

Introduction  3 more complex.4 In brief, he searched for the many predecessors of pseudo-Plutarch and Stobaeus and elucidated how they successively relied upon one another in compiling their own doctrines of philosophers. The exact details of his hypothesis, along with the list of authors of doxography in physics which Diels traced back to Aristotle’s pupil and successor, Theophrastus, need not be discussed here.5 What is important to notice is the predominant historiographical motive in Diels’s enterprise. Early Indologists, like Max Müller,6 and later ones, like Surendranath Dasgupta and Sarvepalli Radhakrishnan, were also tempted to read Indian doxographical7 materials as historiographical documents – a trend which carried on. Eventually, ancient sectarian doxographical classification schemes came to be used by Indologists as constitutive benchmarks delineating the history of Indian philosophy. A quick browse through the various existing encyclopaedias of Indian philosophy is enough to notice how many are built along thematic lines borrowed from ancient doxographies. Surely, there is something practical about well-defined conceptual categories ordering the enunciation of philosophical systems, as one finds them in doxographies, but one problem remains while using such schemes. These ancient doctrinal classifications were eminently sectarian and, by extension, political. None of them gives a neutral account of Indian philosophy. If neutrality is at all desirable in the context of dialectic, or even possible within the writings of an author, are other questions of concern. Within these pages, I argue that every doxography is designed to convey a specific message and, within the Indian philosophical context, to promote a definite path to truth. This is what makes such writings fascinating and worth examining from the perspective of religious dialogue, for example, for doxography is the natural offspring of a dialogical world. Through doxography, one may glimpse how ancient Indian authors conceived of and negotiated with plurality, in its ideological dimension. Back to Diels, his historiographical method of textual criticism, aiming to produce a cladistic reconstruction of the doxographic tradition starting from surviving witnesses and going back to the lost common archetype, followed the textual stemmatology of Lachmann (1793–1851), popular in Diels’s day. According to Mansfeld, Diels’s method gave his theory an “aura of absolute certainty” and ensured its dominating position in the study of ancient philosophy, until today. In philosophical circles, however, such doxographical writings, perceived as uncritical and merely descriptive, soon developed an aura of inferiority. Similar ideas also influenced the reception of Indian doxography in modern scholarship. What brought ancient doxographical writings back in the spotlight, by challenging previous assumptions, was the revisionist work of scholars such as Jaap Mansfeld, David Theunis Runia, Han Baltussen, and André Laks. In a three-­volume study titled Aëtiana, initiated in 1997,8 Mansfeld and Runia reviewed Diels’s hypothesis regarding the structure and development of the so-called doxographical tradition. Their work launched the revisionist movement that, according to Baltussen, was to transform the way the transmission and shape of the evidence for Greek philosophy, the Presocratics in particular, is being viewed.9 One of the main contentions with Diels’s thesis is that he forgot that Theophrastus, too, had a

4  Introduction Doktorvater in no less than Aristotle himself and that the latter happened to champion the same dialectical method which can be seen to underpin the dialectical teleology of later doxographies.10 Mansfeld stresses, Aristotle, in his treatises, as a rule lists and discusses the opinions (doxai) of men in general and of the experts (who often are philosophers) in particular, concerning an issue in metaphysics, or physics, or psychology, and sometimes ethics, before embarking on his own investigations. These opinions are ordered according to the method of diaeresis, or division: a classification according to sets, sub-sets and sub-sub-sets with specific differences. Aristotle checks to what extent these opinions are in agreement among themselves or contradict each other, and then tries to establish to what extent one of the available options may prove acceptable, at least as the starting-point for further inquiry, or whether some option may be available which others have failed to consider. We call these exegetical and evaluative overviews ‘dialectical’, in the Aristotelian sense of the word, of course.11 In the Topics (1.14), Aristotle laid down his dialectical method on how to deal with any particular problem (πρόβλημα). He gave specific instructions on how to select and classify propositions (προτάσεις) and problems (προβλήματα). Because of this procedure, a wide range of opinions on various subjects, but especially in physics, can be organized. Aristotle’s practice has been followed and developed further by later Peripatetic authors, Theophrastus taking the lead. But the practice of compiling doxai soon became widespread, and diverse collections can be found to have been exploited for various purposes among other schools and individual philosophers, like the Academics and the Sceptics, but also among later Islamic and Christian theologians. Today, what remains of this early doxographical tradition is chiefly found in the imperfectly preserved Placita of Aëtius, an extensive collection dated to the first century CE. So far, the theories of the revisionist movement initiated by Mansfeld and Runia have been well received, with some controversies raised by Leonid Zhmud12 and dismissed by Mansfeld.13 For our purpose, the primary significance of this enterprise is the definition of ‘doxography’ which it produced, insisting on the dialectical nature of such writings. As shall be seen, Indian doxographies also tend to reproduce specific dialectical strategies which can be traced back to famous founding figures and narratives. I.1.1.2  Towards a definition of classical doxography Based on his observations on the Placita of Aëtius, Runia enumerated three typical features of texts related to the doxographical genre as a whole: 1 2

The work is divided into books and chapters, which represent a systematically organized whole. Individual chapters are usually structured by means of diaereses, whether as disjunctions or in the form of lists.

Introduction  5 3

To each opinion a name-label is attached, i.e. the philosopher who represents the view, but the views take priority over the names. For this reason, historical and chronological aspects play but a minor role. To put it crudely, doxography is more systematic than historical in orientation.14

These three considerations are strictly formal, whereas the last comment is an assessment of the general value of doxographical data. The first criterion touches upon the text’s main divisions and general organization. It answers the following question: How is a doxographical text presented? It takes the form of a book or many books, divided into chapters. The second criterion deals with the internal logic of the general structure. It answers the following question: How is the doxographical content organized? Diaereses are the divisions of topics according to sets and sub-sets with specific differences, following the Aristotelian method. These sets can be divided based on their disjunctions, which means that the philosophers’ opinions (doxai) would be divided according to their concordance or discordance. The last criterion defines the way in which doxographical data is being internally labelled. It answers the question: How is the doxographical content identified? The answer suggests that the identification or labelling of the data is not a main factor of classification. Regardless of the name of the philosopher, the internal logic of doxographical classification focuses on the view, not on the holder. Therefore, concerns such as historicity and chronology are secondary, if relevant at all. In line with the Aristotelian heritage, doxographies offer an ideal ‘dialectical setup’ somewhat abstracted from the temporal context in which the philosophers’ opinions were originally expressed. Their motive is to offer a systematic overview of opinions on given topics. If we were to paraphrase the criteria suggested by Runia in a single sentence, we could say that, according to him, a doxography consists of a text divided into chapters subdivided into topics under which specific opinions are organized either randomly or systematically, each named after their author. Now, that general definition may create problems when it comes to precisely identifying what text is to be included or excluded in a finite list determining the extent of doxography as an established genre of literature. In ancient Western philosophy, as in the Indian context, there is no lack of texts divided into topical chapters regrouping various views. Also, the idea that doxography may constitute a literary genre of its own is a debated question. Runia’s colleague, Mansfeld, sensitive to the issue, identifies two approaches to doxography: one that defines it in a “narrow” sense and another that offers a “broad” perspective. His narrow definition of doxography follows the parameters established by Diels in his Doxographi Graeci. It is derived from the nature of the majority of the sources that he used and could be said to be a normally very brief presentation according to theme, or subject, of contrasting (or even bizarre, or compromise) tenets in natural philosophy (or science, if you wish), which in itself fails to provide a decisive answer to the issue involved although it may assist you to find a way out.15

6  Introduction This definition is narrow in the sense that it limits the scope of doxography to “natural philosophy,”16 understood in its classical sense of physics (φύσις). Lists of topics detached from physics, like ethics for example, are thus excluded from potential doxographical candidates. The second part of the definition is also relevant. In the Aristotelian dialectical context, for which (narrow) doxographies are being compiled, the purpose is not to bring about certainty on any given topic, but merely to provide food for thought. These two aspects of the narrow definition would be problematic if projected on the Indian context. The limitation of topics to physics, for example, could hardly be maintained throughout. The second part of the definition would also be off target. Every early systematic Indian doxography studied within these pages, and likely most of their successors, dialectically indicates what it considers the best ‘opinion,’ the best darśana.17 Coming back to Mansfeld, he has taken note of a recent trend, notably in the field of ethics, to extend the definition of doxography to include topics not strictly related to physics. He remarks that some scholars even find doxographies in the dialogues of Plato or in treatises of Aristotle, regardless of the fact that these works are dedicated to specific philosophical issues where external views only play an ancillary function. To account for this practice, Mansfeld articulated a “broad” definition of doxography: Broadly speaking, doxography encompasses those writings, or parts of writings, in which the author presents philosophical views of some or other of the ancient philosophers or schools, in some or other areas, or on some or other topics, of philosophy, whether with or without presentation of the argumentation or analysis through which they offered philosophical support or reasons in favour of their ‘tenets’, and whether or not they also include critical evaluations and comments of the author’s own. In other words, these are works (or sections of works) taking as their subject matter the tenets or doctrines of the philosophers, rather than independent works of philosophy in which the author addresses in the first instance issues or topics of philosophy, with ancillary discussion along the way of the opinions of other philosophers.18 This broad definition is of considerable interest here since it corresponds more or less to what Indologists have so far recognized in Indian doxography, namely that doxography is, to paraphrase Mansfeld, (1) either a whole text or a part of a text in which competing views of philosophers or schools are presented; (2) where the original argumentative support of such views may or may not be given; and (3) where the author’s own view and arguments may or may not be criticized. Mansfeld stresses that doxographies are always fully dedicated to the presentation of competing views on a given topic and are not independent compositions where philosophers would formulate their own view. Doxographical contents are either ‘fragments’ or ‘testimonies’ dependent on a certain tradition of transmission. André Laks, who discussed the implications of this feature of doxography, observes, Le couple fragment/témoignage fait partie de l’appareil critique fondamental de tous les historiens de la littérature ancienne, quel que soit le domaine

Introduction  7 considéré. . . . Le fragment provient de l’œuvre elle-même, c’est une citation littérale, dont l’auteur est, en premier ressort, la source (même si la transmission suppose l’intermédiaire de multiples scripteurs). Le témoignage, par contraste, n’est pas le fait de l’auteur, mais d’un lecteur – un lecteur qui dit quelque chose de l’œuvre en son propre nom. Au lieu d’une citation littérale, il nous livre une élaboration secondaire, qu’il s’agisse d’un résumé, d’une paraphrase, d’un commentaire, d’une allusion.19 Indian doxographies consist mainly in testimonies, sometimes accompanied by limited identifiable fragments. This may explain the frustrations of some Indologists who, overlooking the nature and function of the material they were dealing with, accused Indian doxographers of not being rigorous in their descriptions of competing philosophical tenets. In Bhāviveka’s MHK Chapter 9, for example, the irony in the dialogue occurring between the author and his Mīmāṃsā opponent can give a humorous tone to some arguments, but logical rigour remains throughout. This brief overview of studies in classical doxographies has highlighted (1) the systematic, (2) the dialectical, and (3) the ahistorical nature of doxography, three essential features shared by Indian and Western sources alike. It also provided the foundation for a clear definition of the literary ‘genre,’ a definition which could generally be applied onto varying doxographical materials, of divergent historical and cultural contexts. Before proceeding to elucidate a universal definition of doxography able to encompass both the classical and the Indian kind, it is now proper to review the opinions of Indologists on the subject. I.1.2  Doxography in Indological studies I.1.2.1 The dissemination of an undefined concept: Indian doxography Indologists directly borrowed the term ‘doxography’ from the classicists. The earliest Indologists to apply the term doxography onto specific Indian sources, Wilhelm Halbfass (1988) and Olle Qvarnström (1989), did so without investigating further the genesis of the term. For example, justifying his linguistic choice, Qvarnström simply explained that he was following Halbfass in borrowing the concept from the classicists: “availing myself of a concept familiar to historians of Greek and Roman philosophy, I subscribe to Halbfass’ linguistic usage and call these doxographies.”20 Back in Halbfass, besides a brief reference to the thirdcentury doxographer Diogenes Laertios in a paper on darśana,21 one finds little to clarify his understanding of the term ‘doxography.’ Its connotations appear obvious. The first occurrence of the word ‘doxography’ in Halbfass’s India and Europe is accompanied by a laconic description: An obvious basis for the semantic association between “philosophy” and darśana is given by the fact that darśana is a familiar and characteristic term in Indian doxographic literature, i.e. in that literature which summarizes and

8 Introduction classifies the main schools or systems of what is commonly called “Indian Philosophy.”22 This brief definition equates the form of ‘doxography’ with its function. Being concise and systematic, doxographies are meant to summarize and classify. Or, so they appear to. Halbfass’s definition remains silent about the origins of the term doxography. It ignores the actual composition of the genre along with its dialectical and rhetorical purposes. Although a full chapter of India and Europe is dedicated to the subject of Indian doxography,23 one finds no historical analysis or any inquiry into the term ‘doxography’ itself. It is assumed to be self-explanatory. A doxography summarizes and classifies. As will be shown within these pages, this oversimplification is a recurrent misunderstanding of Indian doxographical sources. If these two features are present within doxographies, summarizing and classifying, it is because they structure a dialectical project through which the author seeks to establish the superiority of his worldview. Prior to Qvarnström and Halbfass, A. K. Warder (1970a) had briefly discussed similar Indian sources, but he referred to them as “encyclopaedic writings”24 and nowhere used such terms as ‘doxography’ or ‘doxographers.’ The same can be said of Kendall W. Folkert (1993) who, some years after Qvarnström, discussed Jaina ‘philosophical compendia’ without ever using the term ‘doxography.’25 There is no trace of the word either in such giants as Moriz Winternitz26 (1920), Satis Chandra Vidyabhusana27 (1920), or Fedor Ippolitovich Stcherbatsky28 (1970), who discussed related materials, or in Bruno Petzold’s (1995)29 major work on the classification of Buddhist doctrines. This alone suggests that the concept of ‘doxography,’ as convenient as it might be, may not always have been obvious to everyone. In any case, following the practice of Halbfass, the term ‘doxography’ found echo in subsequent pieces of scholarship, notably in the Buddhist studies of Jeffrey Hopkins30 (1996) and Malcolm David Eckel31 (2008). In fact, the term seems to have been rapidly popularized in Buddhist study circles, particularly in Tibetan studies. For example, ‘doxography’ is mentioned early in the writings of David Seyfort Ruegg (1981),32 Katsumi Mimaki (1982),33 and Geshe Sopa (1984).34 Mimaki made a survey of the abundant Tibetan doxographical literature that he identified with the grub mtha’ type of Tibetan treatises. In the field of Chinese Buddhism, Chanju Mun35 (2006), who retraced the history of the various Chinese systems of doctrinal classification known as panjiao, never referred to the concept of ‘doxography.’ It is only with Andrew J. Nicholson (2010) that the concept of ‘doxography’ is finally traced back to its nineteenth-century origins. I.1.2.2  Towards a definition of Indian doxography OLLE QVARNSTRÖM

Besides the history of the term itself, formal definitions of ‘doxography’ in Indological circles, if formulated at all, have also been rather scant. Qvarnström

Introduction  9 was the first to establish some formal criteria by dividing the available materials he knew into three types. He suggested that: The first two types consist of texts which utilize the already specified device customary since the time of the Upaniṣads and the early Jaina and Buddhist scriptures, namely, to phrase one’s doctrinal position in relation to that of an opponent. What distinguishes the two types is that, in the first type, the doxographer composes a fictitious dialogue between himself and his opponent, forming part of the running text, whereas in the second type, he assigns the presentation of the opponents’ teachings to the first part of the chapter, the pūrvapakṣa, and his own answer, i.e., his critique, to the second part of the same chapter, the uttarapakṣa. The third type of doxographies consists of texts that simply outline one system after the other in separate chapters, without refutations of the positions of which they give an account.36 The first type is a section of a more extensive text. The second one is a full text divided into two recurrent parts: the opponent’s views and the author’s reply. The third type consists in a single text merely listing views without refutation. Qvarnström also distinguishes three generic properties of Indian doxography: Irrespective of type and affiliation, the different doxographical texts not only demonstrate features that differentiate them but embody generic properties as well. The most salient one is their lack of a historical point d’appui; the various philosophical stands are treated as once and for all fixed systems (siddhānta) and not as products of history, a tendency that is evident even in modern exposés of Indian philosophy.37 Akin to their Western counterparts, the fact that Indian doxographies have little interest in historiography suggests their dialectical nature. A lack of historiographical concern, even if it can be explained by the predominantly systematic orientation dictated by the dialectical nature of doxographical texts, is not a feature peculiar to doxography alone. It is rather typical of the overall Indian intellectual tradition, philosophical or otherwise. Even though historiography might not be a feature of doxography per se, this does not mean that such literature cannot provide information of a historical nature at all. In fact, this book intends to show that political and religious concerns are simultaneously intertwined within the dialectic teleology of Indian doxographies. Such concerns, in turn, reflect the sociohistorical context of their time, against which it is also necessary to interpret doxographical projects. Qvarnström observes two other common denominators of Indian doxography: their lack of criticism directed (1) towards Patañjali’s yoga, as a separate system of philosophy, and, with the exception of the Madhyamakahṛdayakārikā/Tarkajvālā of Bhavya, (2) towards Vedānta. These last two features of Qvarnström will be problematized within these pages. At this point, it suffices to say that I consider both claims partial towards a certain type of doxography. For, Yoga, as a philosophical system, is present in early Indian doxography but sometimes not where

10  Introduction it is expected to be. As for Vedānta, it is not only present in Bhāviveka’s MHK, but the Vedāntins themselves also produced numerous doxographies asserting the superiority of their worldview. Hence, these claims are questionable, unless one limits one’s definition of doxography to some early Jaina sources alone. As we shall see, Jaina studies have been particularly influential in projecting a ‘neutral’ image of doxography. ANDREW J. NICHOLSON

After Qvarnström, the most serious attempt to formally describe Indian doxography came from Andrew J. Nicholson. He criticized his predecessors for being too broad in scope and chose to limit his own definition to Qvarnström’s third type of doxography alone. His reasoning goes as follows: I choose to restrict Qvarnström’s definition somewhat and use the word “doxography” only for the third type he presents. My reasons for doing this are threefold. First, I think that if we wish to retain a meaning of the word “doxography” close to the sense used by contemporary historians of Western philosophy, it makes the most sense to count only those texts that provide a naturally neutral overview of the opinions of multiple schools. While these texts often contain subtle criticism of the schools presented, their primary function is to summarize, not to critique. Second, my interest in the study of Indian doxography arises primarily from the value these texts provide in helping us understand the underlying logic behind systems of doctrinal classification in premodern India. Qvarnström’s interest, by contrast, appears to be the data that Buddhist and Jaina authors can provide about the early doctrines of their opponents. Many texts can be said to contain “doxographical information” interesting to the modern historian, but not all of these texts are doxographies. Third, if we construe the word “doxography” too broadly, it will not be useful as a descriptive term. The majority of philosophical authors in premodern India write texts using the convention of pūrvapakṣa and uttarapakṣa, and by accepting Qvarnström’s first and second types of doxography, every text containing pūrvapakṣa might qualify as doxography.38 Such reasoning supports Nicholson’s effort, but it needs not restrict my approach. Moreover, at least two elements of Nicholson’s reasoning are questionable – first, his claim that “it makes the most sense to count only those texts that provide a naturally neutral overview of the opinions of multiple schools.” As already noted, historians of Western philosophy are aware of a doxographical literature which is not ‘neutral.’ That is why Mansfeld felt justified to leave room for ‘critical’ material in his broad definition of the genre. In fact, I argue that no Indian doxography is purely neutral, but that all follow a dialectical scheme coherent with their author’s worldview. Second, when Nicholson suggests that the primary function of doxography is to summarize, and not to critique, he is falling into the same oversimplification

Introduction  11 already observed in Halbfass. Even if Indian doxographies essentially present doctrinal summaries, one should not take it for granted that this structural pattern alone determines their function: merely to ‘summarize.’ I rather suggest that such a systematic summary is but a formal feature, perhaps of mnemonic import, subordinated to another predominant function, i.e. the dialectical and therapeutic function of studying opposing views within a given religious context. Nicholson clearly states his interest in studying Indian doxography: to “understand the underlying logic behind systems of doctrinal classification in pre-modern India.” He equally mentions the interest doxography had in the eyes of Qvarnström: “the data that Buddhist and Jaina authors can provide about the early doctrines of their opponents.” Although my research is surely informed by these motivations, it looks beyond. I am primarily interested in assessing the socioreligious motives of doxographers, and the dialectical methods they designed, for engaging with the plurality of philosophico-religious views prevalent in their days. For example, I examine how doxographical writings, by their structure, method, and content, inform us about the dialectical strategies their authors devised to deal with ideological opposition. I am also attentive to the rhetorical processes by which apologetic endeavours are embedded within a soteriological narrative. Thus, I am not merely attending to the Indian dynamics of doctrinal classification, but also to the significance of studying ‘views’ within the worldviews that produced doxographies. I am particularly concerned with what I characterize as a philosophical-cum-therapeutic motive underlying doxographical writings, at least perceivable within the metaphors of philosophical language. For this purpose, I consider all recorded attempts to deal with competing views as worthy of consideration, especially if they assume a doxographical format. I need not be too strict in my definition of doxography, but I need levels of precision, nuances. Doxography is a philosophical language after all, not a science. Every book has its limits, however. The present inquiry is restricted to analogous Mīmāṃsā chapters from three of the earliest Indian systematic doxographical writings. Classicists insist that dialectic is the “Sitz im Leben” of doxography.39 Could the same be said of Indian doxography? To assess this possibility, a critical study of the genre must examine the dialectical methods developed by their authors to engage with competing views. It should also probe the function that these dialectical texts were intended to serve in the intellectual and religious life of their intended public. This concern happens to be one of the three important questions that Mansfeld felt were missing in the pioneering work of Diels: “He never asked the question what purpose the lists of opinions (placita) might have served, and why they had undergone continuous additions and changes.”40 Similarly, we may ask ourselves what purpose did Indian doxographies serve and why this genre of literature continued to develop, even beyond the borders of India. If Indian doxographies were not intended as some sort of historiographical texts, could they, in their listing of competing views, inform a dialectical exercise indicative of a peculiar religious attitude to plurality? I suggest that it is the case.

12  Introduction The dialectical study of competing views in medieval India should not be conceived as a mere historiographical interest in philosophy, something which would be rather anomalous within the overall ancient Indian ethos. Similarly, it is unlikely that doxography be a mere training in debate. Yet, it surely comes from the same environment. Debate itself is not an isolated phenomenon within the religious life of ancient India. Just like doxography, it is but one medium of an effervescent socioreligious activity. By systematizing positions of debate, doxographers engage their students in a dialectical practice to discriminate the true from the false. Naturally, their systematization is coherent within the doctrinal and educational environment in which they gravitate. In fact, such dialectical exercises only make sense within their intended institutional context. As a pure product of scholasticism, doxographies are born of an institutionalized mode of intellectual formation, already transmitting centuries of contemplative information to monks and ascetics. To distinguish the true from the false, in this scholastic context of intellectual training, sublimates two universal religious rituals, that of the ‘conversion from’ and the ‘confirmation of’ beliefs. With scholasticism, through doxography, knowledge is ritualized. As will be discussed, many scholars have emphasized the significance of debates in the social and religious life of ancient and medieval India. This environment naturally shaped Indian doxography. According to the famous Chinese Buddhist monk and pilgrim, Xuanzang (602–664), debate was essential to a successful scholarly career. On the subject, Malcolm David Eckel noted, An effective debater had to be familiar not just with different Buddhist traditions, but also with non-Buddhist rivals, including the Lokāyatas, Jains, Śaivas, Sāṃkhyas, and Vaiśeṣikas. Scholars moved around the country, studying with experts in other traditions and debating with their opponents. Preparation was important.41 Although one may consider whether doxographies have been used as ancillary manuals to train for debate, one must admit that the knowledge of views gained therefrom is rather scarce and often inaccurate. A debater only equipped with such a minimal understanding of the ideological systems defended by his opponents is unlikely to win over any serious contender. At best, if training in debates was part of the motivation behind doxographical writings, these short texts could merely serve a mnemonic function, summarizing views of opponents into a few crucial points. Perhaps a teacher could use this framework as a basis for more elaborated commentaries and training.42 But, doxographies would appear redundant in the face of broader treatises where similar topics were given a more in-depth and accurate discussion. Bhāviveka’s MHK, along with its commentary, is perhaps an exception here. Contrary to other early doxographies, the text is complex, lengthy, and well argued. Nevertheless, I will explore new interpretative avenues. What if, just as Mansfeld and Runia intuited within the Greco-Roman context, Indian doxographies had a philosophical function of their own within the wider practice of philosophy? What if, to borrow from Baltussen on fragment editions,

Introduction  13 instead of treating doxographical materials like precious stones to be excavated from their environment, their original context was to receive considerable attention?43 Since we know that the three traditions which produced doxographical texts, namely Madhyamaka Buddhism, Jainism, and Advaita Vedānta, trained their followers in a disciplined life oriented towards the fruit of religion, generally framed within the narrative of liberation, it is logical to consider whether doxographies could be a literary technology44 designed to contribute to that same narration. But how would that work? How does the study of views different from one’s own contribute to one’s religious experience? To answer such questions, one must inquire about the function of dialectic in itself and within a given tradition. Therefore, each chapter of this book, dedicated to the early doxography of a distinct tradition, begins by a preliminary assessment of the function of dialectic within that tradition. To summarize this brief overview of the theorization of doxography in Indological studies, one may retain that the definitions of ‘doxography’ produced in Indological circles have so far been either too broad or too narrow. To improve upon this situation, I now suggest a universal definition which could define both the formal aspects of the genre and the nature of its content. For this purpose, it is most practical to combine Mansfeld’s and Runia’s definitions, elaborated within the classical field, to portray a generic literary genre, found in numerous philosophical cultures, which came to be referred to as ‘doxography’ from the late nineteenth century onwards.

I.2  A universal definition of doxography Looking back at Mansfeld’s broad definition, where doxographies are described as (1) either a whole text or a part of a text where competing views of philosophers or schools are presented; (2) where the original argumentative support of such views may or may not be given; and (3) where the author’s own view and arguments may or may not be criticized; one notes that it leaves out two formal elements from Runia’s definition: (1) the diareses structure and (2) the labelling of the chapters. The observation that doxographies tend to be more systematic in nature, whereas historical concerns are secondary, if present at all, is also missing. These points can be added without problems to Mansfeld’s definition. In fact, by combining the definitions of Mansfeld and Runia, both formulated out of extensive scholarship in classical philosophical literature, I can now proceed to establish the definition of ‘doxography’ which will be adopted in the present research. A doxography is: 1 2 3

either a whole text, or a part of a text: where competing views of philosophers or schools are presented following a division of topics organized into sets and sub-sets with specific differences to which a name-label is attached in most cases; where the original argumentative support of such views may or may not be given;

14  Introduction 4 5 6

where the author’s own view and arguments may or may not be criticized; where the content consists either in literal or in non-literal renderings of sources; and where the overall concern is primarily systematic, dialectic, with little or no historiographical character.

This definition outlines with concision and precision the formal structure of doxographical writings, accentuating the dialectical disposition of their content, philosophical views. The fact that it can be applied to a broad range of materials across cultures and periods makes this definition a valuable comparative apparatus; a research avenue deserving further attention. Having thus discussed the formal features of doxography, it is now time to pounder the nature of its content, to assess the significance of ‘views’ within the Indian philosophical context.

I.3  The content of Indian doxography: views Although the form adopted by Indian philosophical doxographies may vary significantly, their content has essentially one object: philosophical views. It is thus no surprise that the common Sanskrit term for ‘view,’ darśana, is often found in the titles of Indian doxographies. Scholars regularly translate darśana as ‘philosophy’ or ‘philosophical system.’ Yet, the metaphor of sight and the semantics of ‘perception,’ of ‘seeing,’ of ‘perspective,’ or again of ‘opinion’ carried by related Sanskrit terms is too rich and often exploited by Indian philosophers to be abandoned. Terms like dṛṣṭi, naya, and mata, conveying varying nuances of an epistemic order, frequently appear in philosophical discussions. The notion of ‘worldview’45 might be better suited to translate darśana. In any case, doxographies are no doubt dealing with ‘views.’ In order to understand the nature and function of Indian doxography, it is crucial to reflect upon the nature of ‘views’ in plural, and of ‘right view’ (samyag-darśana) in particular. The Indian authors discussed within these pages repeatedly adopted the semantic of ‘sight’ and of ‘medicine’ in relation to their dialectical and hermeneutical activities. Could doxographies also be some kind of programmatic ‘vision therapy,’ where dialectic is used as a transformative tool? My analysis suggests that, indeed, doxographies are designed to transform one’s ‘view’; that they seek to redirect it from the unreal (asat) to the real (sat), as with the transcendental aspiration found in the famous pavamāna mantra of the Bṛhadāraṇyaka Upaniṣad (1.3.28): asato mā sad gamaya.46 To explore this avenue, I examine the doxographical impetus of three competing traditions, namely Madhyamaka Buddhism, Jainism, and Advaita Vedānta, and I formulate what could be three distinct, though correlated, answers to the legitimate question: What is the point of views? It appears that the transformation of one’s view is a pivotal component of the religious path (mārga) promoted by the three traditions which preoccupy us. For instance, a decade ago, Paul Fuller examined the notion of diṭṭhi (the Pāḷi equivalent of the Sanskrit dṛṣṭi – ‘seeing’) in early Theravada Buddhism. His opinions were recently reformulated within the wider Buddhist context, by Albert Charles

Introduction  15 Muller.47 Both observed that the notion of ‘view’ and particularly of ‘right view’ is of major significance in the context of the Noble Eightfold Path which defines the Buddhist religious ethos. Therein, the entire path rests upon the cultivation of ‘right view’ (Pāḷi: sammā-diṭṭhi) and the abandonment of ‘wrong views’ (Pāḷi: micchā-diṭṭhi). The nature of ‘views’ and one’s cultivation of them is thus a fundamental concern for Buddhists. Muller even suggested that: The notion of views is treated with decidedly more importance in Buddhism than in other religious traditions.48 However, Muller’s assertion needs to be mitigated. While ‘seeing’ (dṛṣṭi), ‘view’ (darśana), and ‘opinion’ (naya/mata)49 are certainly a capital locus of transformation along the Buddhist path, they were equally important for the traditions of all the doxographers about to be introduced within these pages. Jainism, with its notorious theory on the many-sidedness of reality (anekānta-vāda), and Advaita Vedānta, with its teachings on superimposition (adhyāropa) and its removal (apavāda), were equally concerned with the impact of views on one’s relation to reality. In fact, the spiritual exercise of critically engaging with one’s view seems not only to be a pan-Indian concern, but one generally shared among those engaged in philosophy, from the earliest days of the classical tradition, as highlighted by Pierre Hadot, whose theories we shall discuss in a moment. Beyond mere intellectual games, views, and the knowledge which they serve to reveal or hide, depending on their nature, being possibly right (samyag) or wrong (mithyā), affect behaviour (cāritra) either positively (resulting in puṇya – merit) or negatively (resulting in pāpa – demerit). Hence, for anyone anxious about the consequential fruit (phala) of one’s action (karma), as the Indian ascetics of old, reason imposes that the dynamics of the occurrence (vṛtti) and cessation (nirvāṇa) of views be fathomed. Fuller explained why the notion of ‘view,’ perceived as an obstacle to ‘seeing things as they are’ (Pāḷi: yathābhūtadassana), is fundamental to Buddhist thought. His thesis clarified what is meant, in this context, by holding a ‘right view’ and rejecting ‘wrong’ ones: Wrong-view is neither a wrong proposition requiring correction – the adoption of right-view; nor is it entirely a form of craving requiring rejection – the practising of no-views. It combines both what is untrue and harmful. On the other hand, right-view is not the adopting of a correct doctrine and the rejection of an incorrect doctrine; nor is it the rejection of knowledge – the abandoning of all views. The realization of the way things are is itself the cessation of craving. It combines the notion of ‘is’ and ‘ought’ and in so doing reflects both ‘what is’ and what has ultimate value.50 In the context of doxography, one may wonder whether it is possible that, through a dialectical presentation of views (darśana-s), such philosophical material hints

16  Introduction at a realization of ‘the way things are,’ transcending mere ‘seeing’ (dṛṣṭi). Beyond the minute debates which it embodies, could a doxography be used within one’s tradition as a kind of contemplative text conveying an “exercice spirituel” or, again, as a “technique du soi,”51 having a therapeutic function, the alleviation of a form of ‘suffering’ (duḥkha) caused by distorted views (mithyā-darśana-s), rooted in “metaphysical ignorance” (avidyā)?52 It is worth examining. When looked at in their performative dimension, doxographies appear as spiritual exercises. As will be seen, for our three doxographers, which can be said to promote a knowledge-oriented path (jñāna-mārga), in contrast to a path of action (karma) or devotion (bhakti), views, the result of seeing, can only be partial, conveying but a limited perspective on reality. Adopting a meta-philosophical perspective extracted from the three traditions under study, it can be said that views superimpose various distinctions upon reality which are not part of its inherent nature but are rather inbuilt within the seer and his seeing act, through the various norms with which one interacts. In other words, views are tainted mental activity and not pure reality. They are caused by and cause further tainted activity. Thus, they actuate the dynamics of saṃsāra, the conventional world of perpetual ‘reaction,’ and are ultimately to be abandoned by the jñāna-yogin. According to such thinkers, liberation cannot be caused by any means (yoga), otherwise it would be impermanent, as anything produced. Ultimate freedom (niḥśreyasa/mokṣa), then, becomes the very cessation (nirvāṇa) of such activity and of its associated disturbance (duḥkha). Freedom ‘re-appears’ through the ‘re-establishment’ of clear sight (samyak-darśana), obscured through recurrent tainted activity (karma). This tainted activity itself is the mark of a defective sight, of ignorance (avidyā). Views are ultimately deceiving. They never fully or solely capture the actual seen (dṛṣṭa), but they rather mingle the seer’s (draṣṭṛ) own mental activity with the perceived object, the ultimate ‘object’ of interest being the self (ātman) or truth (tattva). Views tend to hide reality. At best, if they are ‘correct’ (samyak), and this distinction is fundamental, they may directly point at reality (tattva) through some pedagogical instructions (anuśāsana/śikṣā/śāstra). But, being conventional by nature, views are never ‘reality’ itself. Thus, for the philosopher (tārkika) as for the mystic (yogin), which are but two sides of the same coin, what Bertrand Russell presented as two interrelated human impulses towards science and mysticism,53 the practice of distinguishing views (darśana-s) from the real (sat)54 is an essential component of the scientific (śāstrika) and spiritual (ādhyātmika) method. Both methods combined, in the Indian context which preoccupies us, constitute the jñāna-mārga, or the practice of jñāna-yoga, the yogic cultivation of wisdom. The practice of distinguishing views recalls the old jñāna-mārga (path of knowledge) of the ancient Sāṃkhya school, which sought to distinguish between puruṣa (consciousness) and prakṛti (prima materia). Therein, one would learn to identify a set of realities (tattva-s) only to negate them later on, by an analytical process of discrimination (viveka) designed to separate pure consciousness from its antithesis. It is not too farfetched to suggest that discriminating between tattva-s and puruṣa or between the various components of philosophical views (darśana-s) proceeds

Introduction  17 from a similar ‘intellectual exertion.’ Within these pages, I elucidate how the earliest doxographies of three competing Indian religious traditions have designed different propaedeutics to support this kind of ‘intellectual exertion’55 (jñāna-yoga) within their respective worldview. As shall be seen in the section on Jaina doxography, a systematic understanding of ‘view’ (darśana) came to be generally adopted around the seventh century CE, likely devised within Jaina circles, distinguishing the views of established philosophical systems (siddhānta) from the mere ‘seeing’ (dṛṣṭi) of the commoner, or from the ultimate sight (samyag-darśana) of the Jina. Indian doxographies, be they Jaina or other, are interested in established systems, worldviews. In brief, their focus is on what the Greeks would call endoxa (ἔνδοξα),56 a well-known consensual view held by an expert or by a group of people. But the Greek term mainly conveys the shared dimension of such ‘opinion,’ whereas the term darśana also came to carry an implicit definition of the content of ‘views.’ As discussed in the chapter on Jaina doxography (2.2.3), a darśana is an opinion held by an authority (āpta). But this is not enough. It also comes with a set of principles (tattva-s) established through a given epistemology, amounting to various means of knowledge (pramāṇa-s). These three elements (āpta + tattva + pramāṇa) are what is generally found in Indian doxographical lists. One could say that they constitute the very backbone of Indian philosophy. To conclude this brief overview, let us notice how, for our three doxographers, even though the views (darśana-s) of experts (āpta-s) are more systematic and elaborated than mere common ‘seeing’ (dṛṣṭi) and are transmitted by authoritative traditions, they retain a ‘conventional’ status. This is why philosophers claimed the right and the soteriological necessity to approach views through reason (yukti) and inferential means (anumāna). This is especially made clear by Bhāviveka, the author of the Madhyamaka doxography about to be introduced. The great Advaita theologian, Śaṅkara, also relegated the Veda and related scriptures (āgama-s), the very āpta of his system, to the realm of non-ultimate truth (aparāvidyā), based on his reading of the Muṇḍaka Upaniṣad (1.1.4) for example. In this view, only brahman is the ultimate truth (parāvidyā). Śaṅkara’s understanding of the scope of reason, however, differs significantly from that of Bhāviveka. This being said, there seems to be a consensus about the fact that conventional ‘views’ (darśana-s) are not truth (sat) itself, in the sense of being the actual reality towards which they would rather point, if valid. Nevertheless, ‘right’ views serve as effective ‘indicators’ or ‘inferential proofs’ (liṅga-s), if they survive the fire of critical examination (parīkṣā). In essence, views (darśana-s) are but an authoritative set of ‘principles’ (tattva-s), which must be analyzed with reason (yukti) by the ‘discriminator’ (vivekin), the aspiring yogin on the path to ultimate freedom. Only then can the ‘nectar of truth’ (tattva-amṛta), an expression dear to Bhāviveka, be extracted from views, as through a kind of dialogical torsion (tarka), an extraction of their linguistic and logical components. In that optic, Indian philosophers continuously argued over the dubious logic of opposing views and, in the process, to support their intellectual exertion (jñānayoga),57 composed what we now call Indian doxographies.

18  Introduction

I.4 The function of Indian doxography: spiritual/ yogic exercise In order to probe the function that doxographical texts were meant to serve in the intellectual and religious life of their intended public, I propose to read them as ‘spiritual exercises.’ The expression is borrowed from the historian of classical philosophy Pierre Hadot (1922–2010), who was ‘Directeur d’études’ at the École pratique des Hautes Études (EPHE) and professor at the Collège de France. Breaking away from the dominant type of historiography of philosophy, which consists mainly in the analysis of the genesis and structures of philosophical literary works, assessing the rational edifice and the internal coherence of their systematic exposition; Hadot rather sought to show that what the ancients considered as pure theory, as abstraction, was in fact eminently practical, both in regards to its mode of exposition and to its finality. From this practical perspective, Hadot suggests that philosophy appears, in its original form, not as a mere theoretical construct, but as a formative method, leading to a new way of life, to a new way of seeing the world. In brief, philosophy is a means to transform man.58 The work of Hadot, along with reflections on related thematics by Michel Foucault, soon gathered interest from scholars of Buddhist studies. Matthew T. Kapstein and Sara McClintock took the lead by exploring Hadot’s ideas in the context of the eighth-century Tattvasaṅgraha of the Buddhist Śāntarakṣita, and its commentary from Kamalaśīla.59 Responding to their praiseworthy efforts, Vincent Eltschinger presented a critical reply60 in which he summarized the theoretical framework structuring the work of Hadot and sought to reflect on its heuristic potential in Buddhist studies, along with the problems it raises. Since then, replies to Eltschinger’s remarks have been drawn, notably from Kapstein,61 and the idea of philosophy as therapeutic62 and of Buddhist spiritual exercises63 continued to germinate in edited volumes on the subjects. My purpose, here, is not to summarize once more what others already dutifully compiled. There are some points, on which I nonetheless want to insist, gathered from Eltschinger’s analysis of Hadot’s work: 1 2 3

4

If the ancient philosophical discourse aims at transforming the disciple it is because, as with Platonic dialogues, it tends to form rather than to inform.64 Thus, a philosophical discourse is almost always addressed to a public belonging to a specific ‘school.’ It is within these schools that such discourses are produced and transmitted among initiates.65 In this sectarian context, philosophy becomes a call to ‘conversion.’ Every school engages its disciples in a new mode of being. Hadot explains this religio-philosophical conversion as a transformation of a mental order, which ranges from the simple modification of an opinion to the total transformation of one’s personality.66 In the end, ‘spiritual exercises’ seek to operate a radical change of being, a metamorphosis of personality, to guide one towards a higher level of

Introduction  19 perception, a universal perspective. In this way, these exercises act as transformative vectors of change in one’s worldview.67 These remarks are particularly well suited for the doxographical materials about to be studied here. Concerning point 1: Indian doxographies have often been shunned and negatively reviewed by commentators. Historians of ideas repeatedly pointed out the misleading nature of the information Indian doxographers provide on the doctrines they portray. By moving away from a historiographical reading of these materials, however, and by rather insisting on their dialectical nature, I suggest that the point of Indian doxography is primarily transformative, before being informative. Doxography is designed to direct the student in a precise doctrinal direction signposted by rhetoric and dialectic, not by historicity. The point is not merely to inform students about philosophical exotica. Doxography engages students in a dialogue. The dialogical nature of such texts appears particularly obvious in Bhāviveka’s MHK, for example, where the form of pūrva- and uttarapakṣa (the enunciation of a thesis and its refutation) recalls Socratic dialogues, not without irony. Being dialogic, and by times dramatic, doxographies tend to take some rhetorical liberty with ‘historical reality.’ Doxography is but a means to an end: converting/confirming. And the end justifies the means. In brief, doxographies are intended to shape one’s view. As for point 2, the advantage of the present book is to illustrate how Indian doxography, as a spiritual exercise, adopts different dialectical forms, dependent on the sectarian filiation in which it is embedded. Three different dialectical forms are analyzed within these pages. It is essential, in order to understand the dialogical dynamic of Indian doxographies, to firmly establish each of them in its dialogical soil. I do so within each separate chapter of this book. Only then can their doxographical scheme reveal its full pedagogic and psychagogic value. Uprooted from their doctrinal sectarian context, doxographies are literally insignificant. On points 3 and 4, my analysis of each of the three doxographies under review seeks to elucidate the dialectical teleology through which each doxographer is guiding his student, towards the adoption of right view (samyag-darśana). In so doing, the author engages each student in a process of conversion and confirmation, depending on one’s initial station within the maṇḍala of views. In that sense, doxographies can engage with a broad array of students of different capacities. Within every chapter, I clarify how every view listed in doxographies is taken to be part and parcel of the conventional world. Only by establishing the right view indicated by the doxographer can a student’s view move from its conventional status to the position of truth, the ultimate perspective. This is the ultimate outcome of the spiritual exercise. At this stage, it should be emphasized that, by referring to Hadot’s terminology, the point is not to suggest that the Greco-Roman spiritual exercises he analyzed are the same as the ones found in the three Indian traditions under discussion here. Perhaps this could be an area of comparison, but this is not my purpose. Rather, the expression ‘spiritual exercise’ serves as a reminder that the doxographical texts under study, though appearing to be strictly intellectual and apologetic, are

20  Introduction embedded in a therapeutic context, determined by specific doctrinal configurations. To avoid the confusion of associating too closely the Greco-Roman world with the Indian one, I also adopt a terminology closer to home, and locate these ‘exercises’ within a known form of ‘yoga,’ or ‘method,’ dedicated to the cultivation of ‘knowledge’ or ‘gnosis’ (jñāna). I was invited in that direction by the notorious notion of jñāna-yoga: an expression which was explicitly used by some doxographers (Haribhadra, Śaṅkara) to describe their own path (mārga). The expression has a long history in Indian literature.68 It already appears in the Bhagavad Gītā and the Upaniṣads. At a time when yoga studies are enjoying a due momentum, where yoga, modern and ancient, is re-examined in its pluralistic expression, the history of the variegated practices which presented themselves as yoga-s has become a central locus of attention. No history of yoga could be complete, however, without the contribution of those yogin-s who look onto philosophy as a medium of transmission and transformation: let’s call them the nerds of yoga; intellectuals and scholars. In this context, doxography is a yogic exercise meant to establish oneself in a given position, a given view, in order to see reality through that specific perspective, delineated by tradition. Adopting a view, like adopting a posture, defines the way one stands. And one invariably stands in relation to something, an object: the ultimate object being the self, or reality. One also often stands against something, in opposition to it, as in the face of an opponent objecting to one’s view. In order to win over adversity, to arrive at the ultimate station, one needs to discriminate which position withstands all challenges. One must learn the analytical tools which sustain discrimination (viveka), the various epistemological means (pramāṇa-s) entertained by tradition. The practice of discriminating between various dialectical options is an essential component of every philosophical yoga. Even if one claims that one has already established oneself in a given posture, one must also be able to maintain that same position, in the event of a particular challenge to its stability, for example, as in debate. Indian doxographies are dialectical exercises designed to train oneself in establishing and maintaining one’s view in the face of controversy. In that sense, doxography is a particular form of yoga binding one’s mind to the insight of tradition.

I.5 The context of Indian doxography: a dialogical path to knowledge To better discern the preoccupations of the jñāna-mārga, it is skillful to contrast it with its main rival, the karma-mārga, the path of ritual action. This, in part, explains my decision to analyze the Mīmāṃsā chapter found within each of the doxographies used for the present exercise. It allowed me to embed the development of doxography within a long-lasting ideological rivalry between experts of a ritual-centred scriptural hermeneutics (the Mīmāṃsakas) and advocates of a rational and contemplative form of conventional analysis (the Tārkikas). The socioreligious factors and historical consequences of this power negotiation cannot be fathomed in full within these pages, otherwise the focus on doxography would

Introduction  21 be lost, but they remain an essential background against which the selected portions of the sources examined herein need to be read. These developments must be accounted for through the emergence of reasoning and dialectic as appealing vectors of socioreligious authority, but also as part and parcel of the kind of transformative praxis which required the formulation of doxographies in the first place. Thus, again, the picture complexifies as it gains in analytical depth. The dialectical strategies of early Indian doxographies are such a product of their time that they manage to simultaneously project a contemplative course to ultimate ‘truth,’ which constitutes their ‘gnostic’ and ‘otherworldly’ dimension, and a religio-political vision of the status and function of the doxographers’ tradition within the world in which they gravitated. In fact, the reflection of one dimension upon the other is so intimate that it appears as a mirror play. One is thus left wondering which guided which, spinning once more the infamous chicken and egg dilemma. The sacred and the profane cannot be easily separated. Seen in this light, doxographies are a goldmine for the historian, the philosopher, and the theologian. To recall the expression of the eclectic Russian philosopher, literary critic, and semiotician Mikhail Bakhtin (1895–1975), the world of Indian doxography is a ‘dialogic world.’ Michael Holquist briefly defined the concept of ‘dialogic world’: A dialogic world is one in which I can never have my own way completely, and therefore I find myself plunged into constant interaction with others – and with myself. In sum, dialogism is based on the primacy of the social, and the assumption that all meaning is achieved by struggle.69 Dialogic literature is in communication with multiple works. It does not merely answer, correct, silence, or extend previous writings, but informs and is continually informed by what is done. The importance of ‘dialogism’ within the Indian ethos has already been meaningfully reported by Amartya Sen, who maintains, The long tradition of arguing also has considerable bearing on the reading of India’s past, along with the understanding of contemporary India. Recognizing the history of heterodoxy in India is critically important for coming to grips with the cross-current of ideas, including intellectual processes and scrutinized convictions, that have survived through the turbulence and turmoil of Indian history.70 Dialogue is certainly not the preserve of intellectual speculations but is rather enshrined within the very fabric of Indian society. It is thus somewhat natural to observe it not only at the intersection of the religious and political discourses but also at the very junction of the conventional (vyāvahārika-satya) and the ultimate (pāramārthika-satya) truths, the pivotal axis of liberation (mokṣa). These different perspectives must be considered when reading Indian doxography. With doxographical exercises, one is trained to elevate one’s view, through a systematic dialogue with competing opinions. In the end, the abstract dialectical setup of this

22  Introduction dialogue indicates that doxography is a dialogue with oneself and one’s tradition, an inner reflection of one’s own view on the world.

Notes 1 See introductory remarks, before “1. Introduction and overview,” in Mansfeld 2016. 2 Diels 1879. 3 The Blackwell Dictionary of Western Philosophy, under the entry on the “philosophy of nature,” gives the following information: “This study was the main preoccupation of the pre-Socratic philosophers who originated Western philosophy. Philosophy of nature is also called ‘physics’ [from Greek phusis, nature], and one of Aristotle’s major philosophical works is entitled Physics. Philosophy of nature has been a standard part of Western metaphysical systems, although it has been increasingly superseded by the empirical study of nature.” For more details, see Bunnin and Jiyuan 2004: p. 525. Additional information is also given under the entry on “natural philosophers,” p. 457. 4 For more details on Usener’s theory and its development in Diels, see “Chapter Two: The Theory of Diels’ Doxographi Graeci” in Mansfeld and Runia 1997. 5 One can find all the relevant details on Diels’s research and hypothesis summarized in Mansfeld 2016. 6 In 1899, Max Müller published his own general introduction to Indian philosophy under the title of The Six Systems of Indian Philosophy, a title reminiscent of the Ṣaḍdarśanasamuccaya (Compendium on the Six Views) of the seventh-century Jain logician Haribhadra Sūri, discussed within the present book. See Müller 1899. 7 However, these authors never used the term ‘doxography.’ 8 Mansfeld and Runia 1997. 9 See Baltussen 2005: p. 3. 10 For a summary in three points of the weakness of Diels’s hypothesis, see Mansfeld 1999a: p. 25–26. 11 See “3. Objections to Diels’s theory: the Aristotelian background,” in Mansfeld 2016. 12 Zhmud 2001. 13 Mansfeld 2002. 14 Runia 1997: p. 94. 15 See “6. Doxography broad and narrow,” in Mansfeld 2016. 16 See note 3 on “philosophy of nature.” 17 Daniel Stuart kindly brought to my attention that it might be relevant here to consider earlier Buddhist (proto) doxographical sources, such as the Kathāvatthu and Mahāvibhāṣā. According to him, these early accounts do not always definitively indicate the best answer in their listing. Stuart’s assessment requires verification. 18 See preliminary remarks, before “1. Introduction and overview,” in Mansfeld 2016. 19 Laks 2007: p. 27–29. 20 Qvarnström 1999: p. 171. 21 Halbfass 1979. 22 Halbfass 1988: p. 264. 23 See Halbfass 1988, Chapter 19: “The Sanskrit Doxographies and the Structure of Hindu Traditionalism.” 24 Warder 1970a: p. 9. 25 Folkert 1993. 26 Winternitz 1920. 27 Vidhyabhusana 1920. 28 Stcherbatsky 1970. 29 Petzold, Hanayama, and Ichimura 1995. 30 Hopkins 1996. 31 Eckel 2008.

Introduction  23 32 Ruegg 1981: p. 59–61. 33 Mimaki 1982. 34 Sopa 1984. 35 Mun 2006. 36 Qvarnström 1999: p. 174. 37 Qvarnström 1999. 38 Nicholson 2010: p. 148. 39 As per Mansfeld’s formula, quoted in Laks 1999: p. 49. 40 Mansfeld 1999b: p. 25–26. 41 Eckel 2008: p. 15. 42 The question comes back again: a ‘training’ into what? 43 Baltussen 2005: p. 6. 44 For more explanation on the expression ‘literary technology,’ see note 36 in Chapter 1. 45 See further discussion on the concept of ‘worldview’ at note 48 in Chapter 1. 46 Literally, “Guide me from the unreal to the real.” The mantra was popularized by the soundtrack of the 2003 movie The Matrix Revolutions, in a song titled “Navras,” written by Don Davis and Ben Watkins, performed by Juno Reactor. 47 Muller 2011. 48 Muller 2011: p. 164. 49 On the development and use of these terms in Sanskrit philosophical literature, see Halbfass 1979, 1988. 50 Fuller 2005: p. 11. 51 I am borrowing the expressions “exercice spirituel” from Pierre Hadot 1993, 2001, and “technique du soi” from Michel Foucault 2001, 2013. 52 By the term “metaphysical ignorance,” I refer to Hulin 1994. 53 In his essay “Mysticism and Logic,” Bertrand Russell discusses the interaction between two human impulses, which he labels as the “urge towards mysticism” and the “urge towards science.” It is the combination of both, he suggests, which gave birth to metaphysics. In his view, philosophy reaches its greatest potential when the mystic’s apparent insight into a higher reality and a hidden good is combined with a more scientific kind of philosophy, paying attention to the details of nature. A philosophy where the two impulses are divorced, according to him, is lifeless and barren. See Russell 2014. 54 It is to be noted that one of the many forms of the name given to Bhāviveka is Bhāvaviveka, from the compound bhāva-viveka, meaning the “clear distinction of the real” or “he who clearly distinguishes the real/being.” Bhāva, here, has the same meaning as sat. An early discussion on the rendition of the name in Tibetan, Chinese, and Sanskrit is to be found in La Vallée Poussin 1903: p. 582. 55 The notion of ‘intellectual exertion,’ coined by Olle Qvarnström to translate Haribhadra’s jñāna-yoga, will be discussed more elaborately in the chapter on Jainism. For now, it is worth noting that the meaning of ‘intellectual,’ in this context, is better understood when read along Plotinus’s understanding of the nature of the ‘Intellect’, which Eyjólfur Kjalar Emilsson clarified in the following way: “For him the Intellect is at once the locus of the real or real being, and the locus of perfect knowledge and understanding. Its knowledge, being the highest possible kind of knowledge, is naturally of the real. Everything else is an image, an external act of the real (or an image of an image)” (Emilsson 2007: p. 124–125). It is worth contrasting this ancient understanding of the nature and function of the ‘Intellect,’ and thus of things ‘intellectual,’ with the modern perception of ‘intellectual activity’ as a purely cervical activity, a mere reasoning. If the latter understanding were to be projected upon the notion of jñāna-yoga, the spiritual discipline would lose all significance. 56 In a conference held at LMU, in Munich, on 6–8 March 2018, titled Received Opinions: Doxography in Antiquity and the Islamic World, Han Baltussen explained that ἔνδοξα (endoxa) pedagogically informed the dialectic of Aristotle’s Analytics. Properly

24  Introduction speaking, Aristotle himself was not a doxographer, argued Baltussen, but an ‘endoxographer.’ As with Simplicius after him, what these particular classical philosophers were doing was rather a form of ‘critical endoxography’ and not a mere collecting of ‘opinions’ (δόξα), as some later doxographers began to do. Indian doxographies, I argue, also adopted a critical (dialectical) form of endoxography. They were not mere collections or summary of common opinions. 57 I will further explain my use of the expression jñāna-yoga in the section on Jainism (2.1.2.3 The Yoga of Knowledge: Reasoning on Anekānta-Vāda), where I refer to Olle Qvarnström’s translation of the term, found in Haribhadra’s work. In brief, I suggest that Haribhadra’s understanding of jñāna-yoga, as a form of intellectual asceticism necessary to free oneself form the bondage of saṃsāra, is similarly shared by Bhāviveka and Śaṅkara. 58 This summary of Hadot’s hermeneutical enterprise follows the careful review of Hadot’s work compiled by Vincent Eltschinger (2008: p. 487). 59 Eltschinger refers to the doctoral thesis of McClintock (2002). The book format of the thesis was to appear eight years later (2010). 60 Eltschinger 2008. 61 Kapstein 2013a, 2013b. 62 See Ganeri and Carlisle 2010. 63 See Fiordalis 2018. The book gives a substantial overview of the debates surrounding the use of the term in Buddhist studies to date and contributes further original reflections on the problematic, engaging with Eltschinger’s criticism, notably by McClintock. 64 Eltschinger 2008: p. 491. 65 Eltschinger 2008: p. 492. McClintock also dedicated a full paper to the scholarly context surrounding spiritual exercises. See McClintock 2018. 66 Eltschinger 2008: p. 499. 67 Eltschinger 2008: p. 504. 68 For a brief introduction to the expression, see Cush, Robinson, and York 2008: p. 392; Klostermaier 2007: p. 156–165; Raju 1985: p. 8. 69 Holquist 1990: p. 38–39. 70 Sen 2006: p. 21.

1

The beginnings of Mādhyamika doxography Bhāviveka’s MHK

1.1  The context of the MHK Fortunately, since Eugène Burnouf (1801–1852) first introduced the name of Bhāvaviveka,1 much has been said about the Indian philosopher’s life and work.2 Such lengthy and precious toil provides us with the opportunity not to delve too deep into these issues but to focus, instead, on our primary target, the doxographical nature of the MHK and its purpose. Nevertheless, for the uninitiated, and because there is always something to add, especially since the development of Indian doxography has not been a popular research topic, a general introduction is appropriate. Compared with other such Indian philosophical figures, putting a date on the life of Bhāviveka is a relatively easy task. It is facilitated by the fact that his works involved many well-known Buddhist thinkers such as Dharmapāla, Buddhapālita, Diṅnāga, and Sthiramati. His life is believed to have taken place around 490–570 or 500–570 CE. Pinpointing a precise location for the author’s life is a less certain enterprise. As usual, different accounts disagree with one another. The Chinese pilgrim Xuanzang indicates that Bhāviveka lived in South India, in what is now Andhra Pradesh. Another Chinese scholar, Falin, the translator of Bhāviveka’s Prajñāpradīpa, dubiously mentions Magadha, whereas Tāranātha suggests that the author comes from a princely family in South India, from where he travelled north, for studies, before returning south and achieving a successful scholarly career while supervising over fifty monasteries. Recently, Chien Y. Hsu concluded, from a review of the many sources at hand, “Bhāviveka was born into a royal family in Dhānakataka,3 which is presently a western neighbor of Amarāvatī.” By the second century already, the latter city was one of the greatest Buddhist foundations of the eastern Deccan,4 one of the three major centres of Buddhist studies in sixth-century India, along with Nālandā in the east and Valabhī in the west. The name of the author varies depending on sources. One can find Bhavya, Bhavyaka, Bhavyakāra, Bhavyavivikta, Bhavyaviveka, Bhāvivika, Bhāvivikta, Bhagavadviveka, and Bhavavadviveka.5 Tibetan sources mainly gave the name Bhavya,6 though it is not found in any known Sanskrit text. Chinese translations consistently use a set of characters matching the name Bhāviveka, from the Sanskrit bhā, for ‘light,’ ‘brightness,’ and so on; and viveka, for ‘discrimination,’

26  The beginnings of Mādhyamika doxography ‘discernment,’ and so on.7 Hence, what looks like a laudatory title refers to a man of ‘brilliant intellect,’ a suitable name for the sharp and articulated debater who is the first known Indian philosopher to systematically compile, present, and criticize all the predominant philosophical views of his time, including those alien to his Buddhist tradition, within a coherent dialectical compendium, the MHK. Although seven other titles are attributed to the famous dialectician,8 only two more are acknowledged to be from his hand: the Prajñāpradīpa (PP) and the Karatalaratna (KR). The first is Bhāviveka’s commentary on Nāgārjuna’s Mūlamadhyamakakārikā (MMK) and is available in Tibetan and Chinese.9 The second is Bhāviveka’s latest and shortest work of the three and appears to be a concise and more accessible version of the MHK. It is available only in Chinese (Dàchéng zhǎng zhēn lùn/大乘掌珍論).10 Thus, only the MHK has been found in Sanskrit,11 while its commentary, the Tarkajvālā (TJ), is preserved in Tibetan. Whether the whole commentary is from the hand of Bhāviveka is uncertain. In a familiar scenario, Yasunori Ejima speculates that a shorter Ur-TJ, written by the famous philosopher, might have pre-existed the present revised and enlarged recension.12 Concerning the significance of Bhāviveka’s work within the history of Buddhist philosophy, to better grasp the impact and method of his thought, one must locate the author within the Madhyamaka trend of Mahāyāna Buddhism chiefly initiated by Nāgārjuna. In India, according to reports of Chinese pilgrims, Bhāviveka’s influence was so important that he came to be considered as the legitimate successor of Nāgārjuna. Based on the fact that Nāgārjuna and his pupil Āryadeva did not seem to use the term Mādhyamika to designate their school of thought, but that it was mainly due to Bhāviveka that it became an established philosophical system referred to even by its opponents, modern scholars have argued that, in fact, most of the later Mādhyamikans belong to Bhāviveka’s lineage.13 It is known that, prior to Bhāviveka, as the speculations of the Mahāyāna developed, based on a new set of sūtra-s, moving away from the Abhidharma, in an effort to distinguish their commentaries from those of the Ābhidharmika-s, scholars of the great (mahā) movement, including the Mādhyamikans, came to adopt the generic term śāstra to label their philosophical compositions. On this phenomenon, Hsu makes a significant observation for the present enquiry, aiming to contextualize the birth of Mādhyamika doxography: Those who compose śāstras to promote Buddhism are addressed as śāstramasters [ācārya-s]. In their compositions, in order to impose their religious authority, they intensively challenged different scholars or schools in both Buddhism and non-Buddhism. Thus, debates with other philosophers are the essential parts of those śāstras.14 Bhāviveka’s MHK is a perfect example of that dialectical ‘śāstric’ trend which, it should be noted, was not peculiar to Buddhism. Though the logician followed in the footsteps of Nāgārjuna and Āryadeva – the latter, in his Catuḥśataka and Śataśāstra, criticizes the views of the Sāṃkhya, Vaiśeṣika, Jaina, Lokāyata, and

The beginnings of Mādhyamika doxography  27 even the Vaiṣṇava and Śaiva schools15 – he adopted the hetu-vidyā (logical science) method of Diṅnāga (480–540 CE) and designed his own strategy of presenting independent (svatantra) logical demonstrations (anumāna) in defence of the Mādhyamika doctrine, thus breaking a long-standing taboo of Madhyamaka methodology, which so far limited itself to a reductio ad absurdum of the opponent thesis without presenting any counter-thesis. Qvarnström argues that Bhāviveka’s major inspiration for the composition of a doxography may have been Diṅnāga’s Pramāṇasamuccaya (PS), dealing with Nyāya, Vaiśeṣika, Sāṃkhya, and Mīmāṃsā within the ‘pramāṇa tradition.’ Bhāviveka’s innovation would thus have been to follow Diṅnāga in criticizing the various schools known to him within the ‘vāda tradition,’ the tradition of debate.16 Naturally, besides philosophical developments, the MHK is also a reflection of the sociohistorical context in which Bhāviveka intervened, a context where debates significantly influenced religious life. In a previous article,17 while contextualizing the work of Bhāviveka within the agonistic sociohistorical background of the sixth century, wherein historical records reveal the increasing occurrences of a violent pattern of Brahmanical usurpation and political agitation, if not of outright revolts and massacres, against rulers and institutions supporting an alternative to Brahmanism, I described how, from its early inception in India, Buddhism has been surrounded by various groups with which it had constant interactions on more or less peaceful terms. The Brahmajālasutta, the first sutta of the Dīghanikāya, lists sixty-two theories regarding the existence of the self advocated by other ascetics, among which, of course, are Brahmins. Bhāviveka himself, in the MHK, mentions a list of 363 views (dṛṣṭi-s), a number which appears as early as the second or third century in Jaina literature.18 It is well known that Buddhism’s main rivals came from the Vedic milieu. The anti-Vedic sentiment is already felt in Buddhist canonical literature and never disappeared from Indian Buddhist treatises altogether, for reasons going beyond mere philosophical disputes. What is more important to remember here is that although Buddhist authors often mention various competing sects since early times, disparagingly, actual debates directly engaging the views of their non-Buddhist opponents, and not only listing and condemning them, as in the Pāḷi sutta literature, are slow to appear. To that effect, Albrecht Wezler noted, Buddhist literature is characterized by the fact that most, if not all, debates carried on in the texts start from and centre around internal Buddhist or even Hīnayānistic difference of views and Abhidharma points of controversy. In Buddhist literature it is only gradually that heterodox doctrines are taken notice of.19 Vincent Eltschinger indicated how the Savitarkasavicārādibhūmi part of the late fourth century CE Yogācārabhūmiśāstra (YBhŚ), in its Paravāda section, provides us with one of the earliest testimonies of the use of parīkṣā, ‘critical examination,’20 in a scholastic context, by presenting and argumentatively refuting sixteen allodoxies. Nevertheless, we must wait yet another century for a systematic

28  The beginnings of Mādhyamika doxography tradition of debate to emerge out of these first departures. Thus, as observed by Eltschinger: Brahmanical orthodoxy and Buddhism had been familiar with each other for many centuries, but in spite of the development of eristic-dialectical (vāda) traditions in both of them and the narratives of debates between their representatives, there are very few textual or otherwise documentable hints at there having been a sustained philosophical confrontation before the (end of the) fifth century.21 Yet, by the fifth century, this picture is changing, and systematic philosophical confrontations with the ‘outsiders,’ in a formal scholastic (śāstric) format, emerge in full bloom. It is within this context that Bhāviveka began his dialectical career. This is not to say that debates between the proponents of competing worldviews did not happen before the fifth century. For, we do know that this is the case, as Johannes Bronkhorst observed: Debates between proponents of different currents of belief or practice took place long before the beginning of classical Indian philosophy. . . . We know that in classical India kings might oblige representatives of different movements to participate in public debates, in which much might be at stake, e.g. the life or freedom of the participants, or the well-being of their movement. . . . And these same public debates appear to have inspired thinkers to revise and improve their positions, thus creating the schools of classical philosophy.22 To come back to the second half of the fifth century CE, where we notice the beginning of a sustained philosophical confrontation between various competing groups, the changes taking place come in the form of a systematic approach to debate (vāda) and critical examination (parīkṣā), making use of new developments in the field of argumentative reasoning (hetu-vidyā) which, in time, developed into a coherent epistemological method (pramāṇa), used and debated throughout the Indian cultural sphere. One of the richest accounts of the philosophical impact of the development of the ‘vāda tradition’ of public debates in India comes from John Clayton. He observed how these dialectical encounters shaped the formulations and development of every philosophical system which came to be labelled as darśana. The main passages of his remarks are worth quoting: [V]āda became a public discourse for defining defensible difference between competing groups able to survive public scrutiny. To say that political advantage was more randomly distributed does not of course mean that political factors were absent from the Indian debating tradition or that debates were not a means of one local group gaining political advantage over another. This is especially true of court debates, where royal patronage and prejudice were often decisive to the outcome of the disputation. . . .

The beginnings of Mādhyamika doxography  29 Representatives of any philosophical perspective – Brahmanic or Buddhist or Jaina or even Cārvāka – had free access to the forum of public debate, providing, that is, that they were willing to have their school’s claims publicly challenged according to the rules of vāda. Contestability, not neutrality, was the price of entry to the public arena in the Indian vāda tradition. . . . Preparation for debate forced each competing darśana to reflect systematically on its own position and also on the opponent’s, since they would have to defeat the opponent on his or her own grounds. In the process, it became clear that their difference – their alterity – was not simply a difference about this or that issue; it was systemic. Their difference was darśanic; it was perspectival. But the darśanas that survived this process gained credibility and earned thereby a place in the set of darśanas. They were regarded as worthy opponents in debate.23 Without excluding the significance of internal philosophical factors, which certainly laid the foundations for the nature and form of the discussions about to take place, the reasons for the occurrence of such new philosophical developments might not entirely lie within the sole realm of philosophy. Why would philosophers suddenly feel ‘obliged’ to strengthen their position and to engage with the views of others, and not only with those circulating within their own tradition? Although the trend was already set in the second to third centuries, by the works of Nāgārjuna and Āryadeva, for example, argumentative philosophy reaches a new systematic height in Bhāviveka, which is worth examining. Eltschinger suggests that these dialogical innovations, though introduced by new developments in thought, were also influenced by a given sociohistorical context: The factors responsible for this sudden outburst of philosophical confrontation cannot be seriously looked for within the competing tradition themselves, since here the reasons are most likely to be of a non-philosophical and sociohistorical character.24 According to Eltschinger, the sine qua non factor of this philosophical shift, towards inter-sectarian confrontation, is ‘a dramatic increase of hostility towards Buddhism as the Gupta dynasty starts to crumble.’25 If doctrinal issues could fuel such animosity, one can safely assume that, as elsewhere, Indian religions and politics became intimately intertwined in an explosive mix. This is Giovanni Verardi’s main argument. According to him, Doctrinal debates became hot political issues, and the very development of Buddhist logic and Brahmanical critique can be construed as functions of the political confrontations characterising the Indian scene.26 Hostility brews in Brahmanical circles, now engaged in a dynamic and successful process of reassertion and self-redefinition. On one side, polemical figures, such as the famous Mīmāṃsaka Kumārila Bhaṭṭa, direct a systematic attack on

30  The beginnings of Mādhyamika doxography nāstika27 philosophical positions; while, on the other side, new theistic popular religious movements – such as Vaiṣṇavism and Śaivism – gain in prestige. These new movements value piety (bhakti) and symbolic rituals, while claiming Vedic affiliations and the wealthy patronage that ensues. Spiritual liberation through knowledge, especially of the Buddhist kind, is not the only concern. One might suggest that this antagonizing mood affected every group perceived as heterodox from a Brahmanical perspective, and not only the Buddhists. Certainly, the Jainas came to feel the wrath of the Brahmins, particularly in the guise of Vīraśaivism. At times, however, the Jainas also contributed towards antiBuddhist animosity. The decision to select the Mīmāṃsā chapter of the MHK for the present study was in part motivated by a historical concern: to illustrate the literary expression of an enduring ideological struggle between Brahmanism and its śramaṇa opponents.28 The post-Gupta era, rife with conflicts on all fronts, shattered the relative peace and religious coexistence which was somewhat maintained under the Guptas. Suresh Chandra Ghosh, commenting on the declining state of learning and the rise of endogamy, commensality, and craft-exclusiveness in seventh to ninth century India, observes, Peace and stability so vitally needed for the growth and development of learning took its flight partly under frequent influx of the foreign invaders including the Arabs through the North-West and partly under the impact of an almost incessant warfare among the ambitious northern potentates for the prized possession of Kanauj.29 Along the lines of Ghosh, Shotaro Iida also suggests that the intellectual decline which followed what V.V. Gokhale describes as “the most fertile and progressive period so full of daring attempts,” the period of Bhāviveka, is “partly caused by the Moslem Invasions.”30 Challenging the causality of this reasoning, which he attributes to a common nationalist distortion in India, Verardi argues for another explanation. He insists that this biased historical perspective, blaming the outsiders and the warrior clans for the social tensions and cultural decline of the era, does not fully reflect the testimony of the available sources and the archaeological evidence which, instead, point to serious political oppositions between the proponents of a casteless open society, supported by the śramaṇa-s and centred around cities fuelled by the prosperity brought about by trade, and the conservative orthodoxy of a Brahmanical elite, imposing by all means the caste system and its agrarian anti-urban values. Verardi’s scenario, though it may not provide the only answer, is better suited to historically reflect on the polemical developments witnessed in philosophy from the fifth century onwards, to which Bhāviveka contributed significantly. As for Ghosh’s statement, peace and stability might be ideal for growth, yet many suggest that war, or necessity, is the mother of invention.

The beginnings of Mādhyamika doxography  31 For example, Randall Collins argues, Social conflicts affected the fluctuating strengths of Vedic, Buddhist, Brahmanical, and devotional religions; inside the intellectual networks, conflicts among the subfactions of these religions shaped the pattern of philosophical creativity.31 Similarly, Eltschinger argues that the social dynamics of the sixth century onwards brought about two major innovations in Indian Buddhism: Buddhist esotericism32 and epistemology. He notices how both “legitimate themselves in a structurally homologous way, and in quite martial terms, as that which, by defeating the outsiders, removes the obstacles to the path towards liberation.”33 The work of Bhāviveka gains in depth and historical significance when framed within that context, the emergence of a new apologetic strategy of Buddhism, making use of epistemology to defend itself and to attack the position of others. Few attempts have been made so far to reflect on Bhāviveka’s work in the light of the turbulent sociohistorical context of the sixth century CE. Adopting a strict philosophical, if not philological, perspective, most scholars focused on interpreting his positions within the doctrinal framework of Buddhism alone, as if Bhāviveka had worked en vase clos. For example, there has been ample discussion on the reasons that led Tibetan doxographers to classify Bhāviveka as a proponent of Svātantrika-Madhyamaka, and more precisely of Sautrāntika-SvātantrikaMadhyamaka, for his criticism of Buddhapālita’s position on the first chapter of Nāgārjuna’s MMK, raised in his PP. There, Bhāviveka argues that it is insufficient to simply state the absurd consequences (prasaṅga) that follow from the position of an opponent, but that one should equally state his own position through an autonomous inference (svatantra-anumāna). Candrakīrti eventually came to the defence of Buddhapālita, whence the Tibetans saw in this split the birth of two separate lines of interpretation. While much attention has been given to the philosophical implications of both positions, little attention was paid to the social context which Bhāviveka was facing and which might have contributed to his stance on inference. As Verardi rightly stressed, “Indian Buddhism should not be studied per se, but in counterpoint with Brahmanical theorisations.”34 Taking into consideration the pan-Indian context of vāda, where one is required to assert his own position prior to engaging in the fray of debate, a prerequisite already stated in the Nyāyasūtra-s of Gautama, and knowing how crucial debates had become for the financial and physical security of the debater and of his following, one might better appreciate the desire of Bhāviveka to elaborate and defend a solid dialectical position of his own, in the name of the Middle Way (Madhyamaka). Malcolm David Eckel, in the broad introduction to his book Bhāviveka and His Buddhist Opponents, where he translated Chapters 4–5 of the MHK, together with the TJ commentary, did mention the pan-Indian concern for debate and its influence on Bhāviveka’s autonomous inference. But he limited his discussion

32  The beginnings of Mādhyamika doxography to the problem of vitaṇḍā – a form of reasoning denounced in the Nyāyasūtra. His concluding remarks on the topic, stressing the limited historical knowledge available on the social dimensions of philosophical debates in ancient India, are worth quoting: We do not know much about the way he lived. We do not know how much he was involved in debate. We do not know who his patrons were, when he won his greatest victories or suffered his most embarrassing defeats. But we can see from the innovations in his commentary on Nāgārjuna that he was committed to bringing Madhyamaka arguments into the mainstream of Indian debate. By formulating his “independent” syllogisms, he could post his own positions on the gate, figuratively or literally, alongside the positions of his competitors, and he could demand a response. This made him not only an innovator in Madhyamaka tradition but very much a man of his time.35 Debate, indeed, seems to have highly preoccupied the sixth-century dialectician. The gamble of the present enquiry is to examine how this dialectical endeavour, captured in the MHK, was not a purely worldly pursuit after fame and scholarly career, nor a mere apologetic strategy, but was also conceived by Bhāviveka and the likes as an essential dimension of the spiritual life. And, to that end, the composition and study of doxographical materials were developed as a contemplative form of ‘literary technology.’36 For this end, one must take a closer look at the MHK as a whole before examining its dialectical strategy in more detail within Chapter 9: the Mīmāṃsātattvanirṇayāvatāra.

1.2  The doxographical content and discourse of the MHK The Sanskrit manuscript of the MHK was first discovered and photographed by the adventurer Pandit Rāhula Sāṅkṛtyāna, in August 1936, in the archive of Shalu (zhwa lu) monastery, over a kilometre away from the juncture between Shigatse (gzhis ka rtse) and Gyantse (rgyal rtse), in Tibet.37 Iida informs us that the manuscript, ascribed to Bhavavadviveka, was written in Rañjanā script.38 The MHK consists of roughly 927 anuṣṭubh verses and is divided into eleven chapters. It is the first Indian comprehensive and systematic study of the siddhānta-s of all the known schools of philosophy and was composed at a time when these philosophies were still in the process of formation. Hence, the MHK is an invaluable historical document, even if its purpose is dialectical and not historiographical. At the same time, or rather in the first place, as an independent work, it presents an original defence of Bhāviveka’s own Madhyamaka system. In that sense, it is also a genuine philosophical reflection, decorated with numerous witty metaphors and poetic allusions to a wide array of pan-Indian literary and cultural lieu communs. The eloquence of the MHK, gilded with subtle allusions and humour, at times incredibly scathing,39 also makes it a literary masterpiece. On several occasions Bhāviveka is able to skillfully appropriate non-Buddhist tropes and rechristen

The beginnings of Mādhyamika doxography  33 them as Buddhist suggestions. This ingenious ability, contributing to the challenge of translation, is evidenced in the concluding verses of the second chapter (Munivratasamāśraya). Therein, Bhāviveka describes the wise ascetic as a Bodhisattva who has taken the compassionate vow and who devotedly engages in his daily practice. Depending on how one translates the verses, however, one can also hear a veiled call to disbelievers, to realize the ultimate truth hidden behind their ‘unhealthy convictions,’ through the power of ‘thorough examination.’ They are invited, by Bhāviveka, to sacrifice such views as oblations to fire, instead of less worthy offerings.40 Wearing the black antelope skin of compassion,41 the one carrying the pure water jar of faith, who’s sensory doors are guarded by mindfulness, sits on the reed rug of resolution; takes refuge at the penance grove of great delight, the Mahāyāna; feeds on the fruits of the rapture of meditation, having the ‘establishment of mindfulness’ (smṛtyupasthāna) as sphere of activity. He whose offenses are destroyed by reciting the entire sūtra-s, eloquent and deep; mutters the Sāvitrī [prayer] of dependent origination, because he resorts to the two truths; worships daily the sun who is the one who has attained complete awakening, with the blossoms of the path, so rich in words, exuding fragrances throughout all of space; having burned unhealthy convictions (saṃkalpa) [as sacrificial oblation] in the fire of thorough analysis (pratisaṃkhyāna); thus may he practice the vow of the ascetic to reach the ultimate abode.42 With a brilliant poetical wit of profound resonance, these verses express the purpose of the MHK, its method, and its target: Bhāviveka calls upon his brethren and his opponents to investigate their own beliefs through sound reasoning, in order to realize the two truths, the resplendent Sāvitrī43 of dependent origination. This was the impetus of the first known Mādhyamikan doxography. The MHK is not a mere historiographical account of sixth-century Indian philosophy. It is the exposition of a dialectical training, aimed at converting the views of those who hold erroneous ones, eradicating doubts in minds confused by competing claims, and consolidating the insight of others, who, having given birth to knowledge already, seek to see ‘nothing.’44 For Bhāviveka, wisdom is gained through the thorough analytical process of negation, which leads one to see through ‘seeing.’ True seeing, it appears, ultimately leaves one with no view. Bhāviveka clarifies this idea in his definition of ultimate reality: Ultimate wisdom negates the entire network of concepts, and it moves without moving into the clear sky of reality, which is peaceful, directly known, nonconceptual, non-verbal, and free from unity and diversity.45 As already elucidated by his inspirational master, Nāgārjuna, ultimate reality (paramārtha), in Bhāviveka’s system, relies on relative (saṃvṛti) truth.46 In terms of Madhyamaka practice, the crux of the matter is to know and explain how to transfer one’s conventional view into an ultimate one, the non-view, to break into

34  The beginnings of Mādhyamika doxography liberation. In other words, how to move from a conceptual knowledge to a nonconceptual one? Through what pedagogy? For that purpose, Bhāviveka innovated by integrating logical tools, inherited from Diṅnāga, into his therapeutic methodology. To better understand how the process is said to operate, one must become familiar with the three possible ways Bhāviveka understood the meaning of ‘ultimate reality.’ Conveniently, Hsu already summarized how, for Bhāviveka ultimate reality contains three connotations: 1) from an ontological perspective, ‘ultimate reality’ means the ultimate object which is understood as the true nature of things; 2) from an epistemological perspective, ultimate reality is non-conceptual wisdom, which refers to how the Buddhist sages view reality; 3) the teachings in accord with non-arising is ultimate reality.47 The first two meanings transcend language and concepts. From this perspective, nothing possesses intrinsic nature (svabhāva), and truth cannot be designated linguistically or captured through concepts. Hence, the ultimate truth is not a ‘view,’ in the conventional sense of ‘worldview,’ or ‘Weltanschauung,’ which always involves ideation.48 Truth is seen through sight, taken in the full semantic range of immediate perception (pratyakṣa). But truth is not mere sight. The distinction between the conventional and the ultimate remains mutually exclusive and no relation between the two, no crossing, can be established. Yet, sight must be posited. For, seeing implies seeing something. How then does one go about becoming enlightened if truth is not something? The third definition, however, allows for a secondary ultimate reality, a wisdom (prajñā)49 involving conceptuality and language (prapañca), acting as a lever to pass from the conceptual to the non-conceptual. This wisdom, “the teachings in accord with non-arising,” which “is ultimate reality,” is obtained from hearing (śruta-mayī). The word of the Buddhas must first be heard. At this primary level, conceptuality remains. Scriptures are spoken in narrative forms. Once their story is heard, it must be analyzed with reason, in order to perceive their meaning. Through critical examination of scriptures, a second wisdom is produced. It plays a pivotal role in philosophical yoga, and the Buddhists call it cintā-mayī-prajñā. This second level of wisdom, born of thinking, could be described as a critical dialogue with, or a study (svādhyāya) of, tradition. It represents a second and subtler conceptual realm of activity meant to support a third and last layer of insight, a third wisdom, non-conceptual: the wisdom of contemplative insight (bhāvanamayī-prajñā). This process of insight appears to follow an intuitive pedagogical method. The student is given a lesson on a topic which he must commit to memory. This is the seed of hearing the teaching. It generally amounts to memorizing a text. The student is then given a commentary on that text to elucidate its content. With commentaries come critics and intertextuality. This is the seed of thinking about the teaching. As the content begins to reveal itself to the student, through dialectical analysis, it transforms the student’s understanding of the architecture of

The beginnings of Mādhyamika doxography  35 his mind and language. This is a common expectation of pedagogical formation: to be assimilated intimately in order to transform one’s way of seeing and doing. Theoretical insight is said to come from an experiential knowledge which pervades theory and reality. It is a code, a symbolic medium of mental transformation. It is the seed of contemplation. A contemplative insight coming to life is one that survived the many storms of life. It is experienced. Through it, the student’s learning takes roots in reality. It quiets the fires of conflict and doubts. The teachings serve as the support object of this continuous rumination, like the food of sacred cows. By ruminating the teachings, the divide between the conceptual and the nonconceptual is said to be progressively crossed over by a bridge whose entry is rooted in firm concepts (heard – śruta – from scriptures) and which ends in pure non-conceptual void, immediate reality, tied at its centre by the ever-subtler junctions of reasoning. This theoretical framework, outlining the pivotal role of reasoning and hermeneutics in Mahāyāna religious praxis, was already enunciated in the well-known third-century Yogācārabhūmi.50 Bhāviveka applied its methodology to his notion of ultimate truth, to strengthen the soteriological resonance of his Diṅnāga-inspired practice of inference. This gradual threefold mode of insights into the nature of reality has also been asserted by other prominent Buddhist thinkers. Eltschinger has pointed out how Diṅnāga, Dharmakīrti,51 and Kamalaśīla “consider the traditional sequence of insights as a self-sufficient means for securing enlightenment once the wrong notions spread and argued for by the outsiders have been discarded.”52 This suggests that, although they perhaps did not attribute any soteriological value per se to the science of reasoning (hetu-vidyā), these authors saw it at least as a necessary step to refute possible wrong views one might entertain, before engaging in hermeneutic praxis proper, understood here as coinciding with hearing and reflecting upon the teachings of the Buddhas. In order to cultivate the non-view of the Buddhas, these authors considered that worldly conventional views had to be removed, like cataracts from the eye, with the appropriate antidote of reason. Along Bhāviveka’s therapeutic journey, the diseases plaguing one’s existence and their remedy are to be found within the conventional realm. It is through the conceptual analysis of conventions that Bhāviveka invites one to move beyond conceptuality altogether. The conventional alone allows the meaning of scriptures to point at ultimate reality. For, in and by itself, reality has no indicator and has nothing to indicate. But scriptures are composed according to conventions, assembled with meaningful concepts. They can only be understood through conventional means (inference), designed to investigate their meaning. Hence, a correct (samyak) hermeneutical analysis is essential for the path. This is what Bhāviveka argues with his Mīmāṃsā opponent in Chapter 9: If, according to you, the word is understood [simply] through hearing, how can you believe it manifests as sound? Understanding, in fact, is contingent upon the convention. The word [as found in scriptures] is that [understanding]. If the convention would not exist in the first place, understanding would not be

36  The beginnings of Mādhyamika doxography logical. Because, like saṃsāra, being beginningless, [understanding derives] from the corroboration of the convention.53 As an endless knot, knowledge, the understanding of anything, is but a convention gained in conformity with the conventional world. This cycle, going on forever, without beginning, the conventional production of conventions, is the root of every understanding. It is saṃsāra. Ironically, scriptures (āgama) themselves, made of words, are mere nets of conventions, like saṃsāra. Thus, valid knowledge is not merely transmitted through hearing scriptures, but by repeatedly analyzing their very nature, their conventions. Even if, as Bhāviveka suggests, “the teachings in accord with non-arising is ultimate reality,” the way to go about assessing such teachings can only be conventional, saṃsāric. Thus, in a Nāgārjunian twist,54 are saṃsāra and nirvāṇa forever bound in truthful scriptures. A truth statement and its understanding (pratipatti) act as indicators, epistemological positions, provided to perception through reasoning (yukti). In the immediately following verse, Bhāviveka reminds his opponent that, as a means of knowledge, scriptures and inference (anumāna) are essentially the same: Moreover, in the case of inference, its status as a means of knowledge is not distinct from the nature of that (āgama).55 To the Mīmāṃsā proponent who claims that only scriptures can reveal the apūrva, Bhāviveka asserts the epistemological necessity of inference, which he sees as mediating the knowledge born of scriptures through the knowledge produced out of reasoning. Indeed, how could scriptures, in themselves, produce any knowledge without one reasoning upon them, according to some conventions? And with conventions come rules, acting as the binding knots of logic, tied in grammar. For a statement to be sound, the reasoning it produces must be valid, states Bhāviveka. Concretely, it must be validated by others through some means. Because contradicting scriptures are promoted by different traditions, their traditional transmission and validation is no universal proof of authority. What needs to be examined, in the end, is what is ‘universally’ true (tattva). According to Bhāviveka, this is the task of reason. In Chapter 9, Bhāviveka begins his uttarapakṣa by claiming, [If one holds that] valid verbal testimony [merely] results from being verbal testimony and from the continuation of its transmission, then all verbal testimony is valid. Therefore, one must establish what the truth is. If a verbal expression can be investigated with reason, then it is valid verbal testimony. That is what one must examine first, thereafter what was said by it, of course.56 Since reason mediates scriptures, the ground is set for Bhāviveka to submit all and any scriptures to the test of fire: critical examination (parīkṣā). If a reason (hetu) is fallacious, it is not logical (ayukti) and cannot be true. Hence, it is false (asat), and a knowledge (tattva-jñāna)57 system relying upon such error is not authoritative

The beginnings of Mādhyamika doxography  37 (asiddhānta), not accepted as āgama, and soteriologically impotent (aśakta). For, as Bhāviveka argues in the immediately following verse, the knowledge producing liberation is the antidote to ignorance: Then, it should be granted that, since it is the antidote to that [ignorance], liberation comes from knowledge, just as freedom from sickness comes about through medicine, because it is the antidote to illness.58 In other words, the reasoning process producing valid knowledge, the path of knowledge (jñāna-mārga), is therapeutic. As I have argued elsewhere,59 for a jñāna-mārgin, like Bhāviveka, knowledge is the cure of saṃsāric diseases. According to him, knowledge is obtained from the corroboration of conventions (v. 9.49v.: saṃketasyānuvādataḥ), the corroboration of the causes and conditions of a phenomenon; an act reminiscent of the mythical churning of the heavenly ocean which produces the nectar of immortality (amṛta) coveted by gods (deva-s) and demons (asura-s). For Bhāviveka, it is the ‘nectar of truth’ (tattva-amṛta), born out of the labour of inference (anumāna), which conveys immortality, liberation.60 The allusion to the mythical nectar is of much significance within the MHK. It appears on several occasions. Lindtner and Gokhale suggested that, before the MHK acquired its actual form, it consisted only in the first three chapters and was titled Tattvāmṛtāvatāra (Entry Into the Nectar of Truth).61 The metaphor of the mythical nectar (amṛta) appears for the first time in the MHK immediately following the opening verses of Chapter 1, after a three-verse long praise to the Buddha taking the form of a discussion on the nature of truth. Therein, Bhāviveka uses the metaphor of the mythical ambrosia (amṛta) to designate the kind of liberating knowledge which the Bodhisattva must cultivate, and which corresponds to the one dispensed by the MHK. A little should be said, according to one’s ability, so that those whose resolve is fixed on the Great Awakening, in order to bring about the benefit of others, may enter the nectar of truth.62 In verses 13 and 14 of the same chapter, Bhāviveka carries the metaphor further, to emphasize how precious, rare, and hard to obtain is this liberating knowledge: What could be more precious for a Brahmin, for Indra or for a universal emperor than the means to quench the perpetual thirst common to all, which is not overpowered by quarrel, decay, annihilation, sorrow and so on, the nectar of the knowledge of the nature of things, completely pacifying suffering?63 And the metaphor comes back again in Chapter 3: The nectar of wisdom, bestowing satisfaction, is a lamp of unobstructed light, the stairway to the palace of liberation, the oblation-fire which fuels on afflictions.64

38  The beginnings of Mādhyamika doxography This nectar of truth (tattvāmṛta) is the ‘three wisdoms,’ at the centre of which, as we saw, acting like Mount Meru in the mythical tale, the pillar of reasoning ‘churns’ the conventional, saṃsāra. Quenching one’s thirst with this knowledge extract amounts to obtaining liberation, the end of craving. Discussing the metaphor of ‘churning’ in association with tarka, Halbfass mentioned that, in Brahmanical context, when related to ānvīkṣikī, a similar terminology “is applied to questions of an Upaniṣadic type, and we hear about the ‘churning’ of an Upaniṣad for the purpose of the ‘supreme ānvīkṣikī’ (ānvīkṣikī).”65 The metaphor of ‘churning’ in relation to tarka appears to have circulated before Bhāviveka. As will be discussed further in Chapter 2 of the present book, on Jaina doxography, Gerdi Gerschheimer’s study investigated the notion of tarka and explained its etymology, derived from the Indo-European root of ‘twisting,’ ‘tordre’ in French, ‘torquère’ in Latin.66 In association with tarka, in the context of debate, it metaphorically portrays ‘deliberation,’ ‘examination,’ and ‘reflection’ as a mental ‘churning’ or ‘grinding’ of conventional knowledge. This investigative act, squeezing out the essence of things through mental torsion, typifies the mode of philosophical inquiry in Indian literature. The practice of inference, reasoning, is the essence of hermeneutics, under which Bhāviveka, loyal to the Buddhist epistemological tradition,67 subordinates the authoritativeness (pramāṇatva) of scriptures (āgama). Here, inference is not only a means of debates, but also one of self-transformation, moving from textual hermeneutics to the fundamental analysis of saṃsāra, of conventional existence. Within Bhāviveka’s worldview, it is inappropriate to reduce the praxis of dialectic to a mere intellectual enterprise exercised on the side of the path to liberation, or to mere apologetics.68 Rather, it is part and parcel of his soteriological methodology. The therapy of the contemplative dialectician seeking liberating truth consists in uncovering his own ignorance, the fallacy of his own reasoning, to uproot it from his mind stream.69 As a compassionate Bodhisattva, his aspiration is also to remove such wrong views in others,70 like an eye doctor.71 This is the dialectical battleground of the MHK, the continuous examination of the conventional views held by self and others. It set the tone of Indian doxography.

1.3  The doxographical structure of the MHK To witness a manifold arrangement of competing views within a single text is not a rare occurrence in Indian literature, and certainly not a phenomenon confined to the realm of philosophy. Accordingly, Upinder Singh emphasized how Many ancient Indian texts are polysemic; many ideas, sometimes contradictory ones, jostle with each other within a single text. This is partly because of their complex compositional and transmission history, but it also tells us something about the nature of the ancient Indian intellectual tradition. Vitriolic debate and diatribe are present in systems of thought that view themselves in oppositional terms; but within traditions, what is visible is a tendency to juxtapose different views, rather than to reject and replace them.72

The beginnings of Mādhyamika doxography  39 This fitting remark not only naturalizes the polysemy encountered in Indian doxographical writings, but most of all, it singularizes such compositions for a noteworthy particularity. Indian philosophical doxographies juxtapose not only the views which are promoted within the broad tradition of their authors, but even, if not especially, views coming from the outside. But, why so? An important task of the analyst deciphering such literature is to decode the peculiar pattern of juxtaposition elaborated in each text. As a rule, this pattern is not random but in tune with the worldview assumed by its author. It is indicative of the propaedeutical strategy behind each composition. Bhāviveka’s MHK offers a good example. The MHK opens on a brief discussion about “safeguarding the aspiration towards enlightenment” (bodhicittāparityāga), followed by a reflection on the significance of monastic vows (munivratasamāśraya) and a revision of foundational Abhidharma doctrinal expositions (tattvajñānaiṣaṇā) examined in a way which reveals the ‘no-view’ approach of the Madhyamaka propounded by Bhāviveka. Chapter 3 perpetuates the analytical pattern initiated by Nāgārjuna, submitting familiar Abhidharmic ontological categories to an analysis which argues that none of them has svabhāva (inherent nature), a hermeneutical strategy found in the Lokātītastava, the Pratītyasamutpādahṛdayakārikā, and many other similar texts.73 With Bhāviveka, one is taught that truth (tattva) is to be achieved through prajñā, which itself is divided into what Lindtner translates as “conventional/practical” and “intuitive/theoretical” truths: the conventional and the ultimate. Ultimately, the Buddha is tattva, explains Lindtner. He also noticed that the MHK’s first three chapters altogether form an independent unit designed to “present a curriculum (caryā) guiding the intelligent student of Madhyamaka to sarvārthasiddhi (accomplishment of all aims).”74 Similarly, Gokhale suggested that the first three chapters were originally composed by Bhāviveka, under the title Tattvāmṛtāvatāra,75 to form the core of the MHK, while the remaining chapters were added later. However, since the whole MHK is accepted as coming from the hand of the same author, Gokhale’s remarks need to be moderated. Judging by its length, the whole MHK must have taken a certain amount of time to produce and could not have been written down all at once. It is quite imaginable that the first three chapters, which present Bhāviveka’s own exposition of the path, cultivating the ‘mind of enlightenment’ (bodhicitta) through virtue (munivrata) and knowledge (tattva-jñāna), were first composed. However, within these sections, as has been shown, there are numerous allusions to non-Buddhist tropes. This suggests that the author already had in mind to engage with non-Buddhist materials. This becomes especially plausible if we consider that, as I have argued elsewhere,76 Bhāviveka’s ideal Bodhisattva is a skilled dialectician, a debater. His main duty is to refute the mistaken views of others. This conception of the Bodhisattva’s career is not original to Bhāviveka. In the first century CE, Aśvaghosa, in his Buddhacarita, already portrayed the Bodhisattva in similar terms, alluding to debates. As noted by Eltschinger: By refuting the other systems and by argument he [the Bodhisattva] caused men to understand the meaning which is hard to grasp. . . . Proclaiming the

40  The beginnings of Mādhyamika doxography final truth of impermanence, selflessness and painfulness does not go without critically examining and overcoming (*Nigrah?) competing religio-philosophical claims.77 The MHK is likely to have been conceived with this exact design in mind: critically examining and overcoming competing religio-philosophical claims. In other words, it is plausible that the MHK was planned from the beginning to be doxographical. It will rather be the task of Bhāviveka’s latest work, the KR, to summarize the MHK, by focusing on the Madhyamaka approach.78 If we establish the MHK as an early example of a historically extended and broad literary genre of philosophical writings, generically referred to as ‘Indian doxography,’ then the overall structure of the text becomes a key element to interpret its philosophical method and the way its worldview is articulated in relation to competing claims. Hence, it is significant to notice that the structure of the whole text develops out of a primary section presenting the author’s own position elaborated through an emptying out of Abhidharmic categories. This stands in sharp contrast to the doxographies composed by the competing philosophical systems which will be presented in the subsequent chapters of this study. For example, it was already observed by Halbfass that the structure of Jaina doxographies does not follow any explicit hierarchical schema, whereas the Advaita Vedānta ones tend to present a hierarchical classification at whose apex stands the Vedānta.79 In the case of the MHK, every view is refuted from the beginning, based on the no-view principle of the Madhyamaka, radically adopting the perspective of ultimate truth, which negates being a view. Along the line of Jan Westerhoff, it is important to insist that the position of the Madhyamaka, here, is not standing on its own, nor does it develop out of the “theory-free or unsophisticated conventionalities of the ordinary man on the street.”80 Rather, it is carved out of the intricate analytical framework of the Abhidharma, which it accepts as a somewhat reliable depiction of the conventional world. This is to say that the Madhyamaka does not elaborate its own systematization of conventional categories. It only analyzes those existing in order to point out the ultimate. Metaphorically, it employs the conventions of others as cannon fodder to assert the ultimate truth, the no-view. Signaling this strategy within a composition of doxographical nature, like the MHK, reveals an ingenuous rhetoric where apologetic and soteriological motives are intimately intertwined. It also indicates that, for Bhāviveka, all views are taken as more or less problematic conventions to be analyzed and ultimately relinquished by those seeking truth. In this sense, the MHK offers an inverted reflection of the later Advaita compositions, wherein each view (darśana) refutes a preceding inferior one, leaving the Advaita throning alone on top of the philosophical pyramid, freed from any fear of predation. Previously, I discussed how doxographical compositions usually follow a precise dialectical teleology. Concerning the MHK, the sequence of presentation of each consecutive view suggests that the further away a system stands from the initial exposition of the author’s own perspective (a Mādhyamikan analysis of Abhidharma), the more it falls from grace and stands in contrast to the

The beginnings of Mādhyamika doxography  41 Madhyamaka of Bhāviveka.81 This overall scheme confirms Westerhoff’s observations to the effect that: [W]hen the Mādhyamika deliberates between different systems of categories at the conventional level he is considering two main criteria: a) its general philosophical usefulness and b) the extent to which the system can be made ‘svabhāva-free.’ This allows him to rank such systems and provides criteria to choose between systems, none of which have any claim to truth at the absolute level.82 This might explain the particular virulence of Bhāviveka’s criticism towards the Mīmāṃsakas, in the last chapter, a people which he discourteously introduces as the ‘shameless’ (anapatrapa) who value ritual acts above meditation and knowledge. The insistence of the Mīmāṃsā on the effectiveness of ritual acts (karma) sanctioned by the Veda in matters of salvation, by endorsing a pure karma-mārga, is indeed diametrically opposed to the jñāna-mārga promoted by Bhāviveka. After having presented his own position in the first three chapters of the MHK, Bhāviveka launches his doxographical conquest. Casting himself as the cakravartin of world philosophy, he begins his dialectical campaign on his own turf, by directing his attacks at the Buddhist śrāvakas (MHK Chapter 4: śrāvakatattvaviniścayāvatāra) and the yogācārins (MHK Chapter 5: yogācāra tattvaviniścayāvatāra). Proceeding like Alexander the Great, who first subdued Macedonia and Greece before proceeding with his dream of repaying the Persians in kind for their earlier invasions, through the corrective critique of its author, Bhāviveka’s MHK strategically unifies the realm of Buddhism under a single nonbanner, made ready to march into Brahmanical lands. Bhāviveka begins his foreign conquests by expanding his criticism to Sāṃkhya (MHK Chapter 6: sāṃkhyatattvāvatāra), portrayed as entertaining two different theories concerning the process of knowledge, which, according to Qvarnström, are “reminiscent of the doctrine of two truths (dvisatya)” found in Madhyamaka Buddhism and in Advaita Vedānta.83 In the following chapter (MHK Chapter 7: vaiśeṣikatattvāvatāra), the theory of liberation of the Vaiśeṣika becomes the main target. Unfortunately, only the last two kārikā-s of this section are preserved in Sanskrit, but the remaining is extant in Tibetan.84 Onward, Bhāviveka becomes the first critic of vedānta (MHK Chapter 8: vedāntatattvaviniścaya), particularly criticizing its theory of the self as having an intrinsic nature, and suggesting that all that is well said in Vedānta was stolen from the Buddha.85 Finally, the author kept his utmost wrath for the Mīmāṃsakas (MHK Chapter 9: mīmāṃsātattvanirṇayāvatāra), whose ways he seems to have despised the most among all. This last chapter, however, as will be discussed more thoroughly in these pages, presents a hodgepodge of Brahmanical views going beyond what is traditionally designated as Mīmāṃsā, ridiculing the belief in such deities as Śiva and Kṛṣṇa, for example. In the end, after having trampled the Brahmins, Bhāviveka finally turns his gaze towards the notion of omniscience (sarvajñasiddhinirdeśa), a central issue in Jainism, Yoga, and Śaivism, on which he will spend a few verses,86

42  The beginnings of Mādhyamika doxography as in a postscript, before setting the final boundary of his doctrinal conquest. This overall dialectical strategy conquers the entire realm of philosophy known at this period. It sets the tone for later Indian philosophical compendia. One cannot discuss the general structure of the MHK without mentioning the specific number of systems organized as separate units. Bhāviveka’s division, without specifically using the label, predates the darśana typology which became a standard of its own, particularly under the impulse of Haribhadra’s ṢDS, as we shall see in the next chapter. The reason for this more or less ‘canonical’ division, however, remains elusive. Nevertheless, just as with Haribhadra’s ṢDS, the MHK singles out six different systems and adds a brief extracurricular discussion on another (Śrāvaka, Yogācāra, Sāṃkhya, Vaiśeṣika, Vedānta, Mīmāṃsā + sarvajña. The whole picture amounts to an enumeration of six tattva-s, ‘truth,’ or ‘perception of reality,’ presented by competing systems, plus one siddhi (accomplishment). The discussion on omniscience is not introduced as related to tattva, but to siddhi. It should thus be seen as an addition to the preceding unit (of the six tattva-s), leaving us, in total, with a ‘6 + 1’ kind of enumeration. In the MHK, instead of referring to the later notion of ‘philosophical view’ (darśana), Bhāviveka distinguishes his chapters based on their different exposition of ‘reality’ (tattva). Similarly, although Haribhadra uses the darśana typology, he also mentions that each darśana is to be identified based on its peculiar alignment of tattva-s, adding the discriminative notion of devatā as an additional mark of a darśana. In the end, for both authors, what distinguishes ‘views’ is strictly theoretical.87 With doxography, we are dwelling in the realm of pure knowledge, of texts composed for and by jñāna-mārgin-s, where to know is to be. The dialectical nomenclature devised by such texts is not interested in mundane divisions, possibly more sociohistorical, such as those which could reflect ritual practices, monastic codes, or any other perceptible distinctions known to fragment the sectarian landscape of India. The structure of the doxographical texts of our jñāna-mārgin-s is precisely built to discriminate between various systems of beliefs, variegated depictions of reality, regardless of other possibly more concrete sectarian means of distinction. This, alone, should have served as a sufficient warning to anyone tempted to reify such doxographical categories as actual historical realities, as if they would portray well-defined institutions existing on their own in the human realm. As we know, a single ascetic rule can unite a variety of beliefs, just like a single worldview allows for numerous interpretations. It is not impossible to imagine, however, that, with time, individuals came to identify with a particular set of beliefs presented within known doxographies and started to distinguish themselves from others accordingly, within or outside their communities. But this categorical identification would most likely not have determined their whole sectarian affiliation and daily practice. We know, for example, that Bhāviveka himself was an ardent devotee of the Bodhisattva Maitreya.88 Yet, such ritual detail is irrelevant within the MHK. The principles of division used by Bhāviveka carried on in later doxographies, particularly of a Jaina filiation. The number six, which became a recurrent

The beginnings of Mādhyamika doxography  43 doxographical pattern, has already drawn much attention. Gerdi Gerschheimer discusses the doxographical relevance of the number in the most extensive way.89 He argues that it appears to be an aporia, since, as we noticed already, even the works that mention six systems in their titles actually present more than six, as in the six plus one pattern of Bhāviveka. Whatever the reason for this sixfold division might be, Eckel observes that Bhāviveka does not treat the number has having any particular significance.90 On the other hand, one could argue that the number six has shaped the reception of Indian philosophy in the West, at least since 1730, when the French Jesuit Jean François Pons, who pioneered the study of Sanskrit in Europe, explained that Indian philosophies are divided into six different sects, the books of which form six sciences known as darśanaṃ.91 In other words, with Pons already, an ancient doxographical classification scheme which was designed by Indian dialecticians for a specific dialectical teleology, varying within each tradition and/or each doxography, was adopted by early Indianists as a valid historiographical presentation of Indian philosophy, with its correlate terminology of ‘schools’ and ‘systems,’ reproducing a pattern familiar to Western classical philosophy. In fact, the classification pattern into ‘schools’ is so convenient that it is practically impossible to find an encyclopaedia or a general introduction to Indian philosophy, nowadays, without seeing it reappearing in one way or another. Coming back to the MHK, a quick look at its general table of contents, browsing through the many subdivisions of its numerous chapters, does leave the impression of surveying a kind of scientific encyclopaedia. The discourse is appealing. Its eloquent style, claiming the enlightenment of rationality, cannot fail to remind one the subversive role of Diderot’s eighteenth-century French Encyclopédie.92 As with Diderot’s project, behind the veil of reason, subtle rhetoric reveals the trials of the days.

1.4  Table of contents of the MHK Before moving on to discuss Chapter 9, it is worth looking at the table of contents of the MHK, in order to have a clear picture of the entire text and to see where the discussion on Mīmāṃsā stands. The whole of the MHK has never been translated integrally in any European language in one single publication. Hence, until now, no general descriptive table of contents has been assembled. The following one has been produced by bringing together what has been done so far by various scholars. In the cases where no detailed table of contents subdividing the chapter has been presented by the translator, I take the liberty to provide subdivisions based on the given translation. As for Chapter 9, I divided it according to my own translation and interpretation. 1  Bodhicittāparityāga93 a

Cultivating the mind of wisdom by training in the view of the void, the wise remain in the world for the benefit of others. (1–33)

44  The beginnings of Mādhyamika doxography 2  Munivratasamāśraya a

Superior ascetic practice is dedicated to the welfare of others, guided by knowledge and compassion. (1–12)

3  Tattvajñānaiṣaṇā a b c

The Meaning and Aim of the Knowledge of the Highest Truth (1–13) Samādhi (14–23ab) Introduction to Examinations (23–24) i Saṃskṛtadharma-s (Conditioned Dharmas) ii Skandha-s iii Rūpaskandha (25) iv Mahābhūta-s (26–39) v Rūpa (40–44) vi Indriyaṃ (v. 45–65) vii Vedanāskandha (66–68ab) viii Saṃjñā and Saṃskāraskandha-s (68cd–69) ix Vijñānaskandha (70)

d e f g h i j k l m n o p q r

Dhātu-s (71) Āyatana (71 ab) Lakṣaṇa (72–76) Gati (77–85ab) Mokṣa and Bandhana (85cd–89) Pudgala (90–92) Ātman (93–99ab) Rāga and Rakta (99cd–108) Nirvāṇa (109–116) Viparyāsa (117–118) Dveṣa and Moha (119) Svabhāva (120–128) Conclusion (129ab) Asaṃskṛta (129cd-136) Introduction to the Non-Production of All Dharma-s (137–138) i Non-Production From Itself (139–146) ii Non-Production From Others (147–158) iii Criticism of the Four Pratyaya-s (159) 1 2 3 4

Hetupratyaya (160–162) Ālambanapratyaya (163–166) Samanantarapratyaya (167–169) Adhipatipratyaya (170–175)

iv The Problem of v Criticism of the vi Non-Production vii Non-Production

Opposition by Pratyaya and Pratīti (176–181) Theory of the Saṃkhyā (182–191) From Itself and Others (192–193) From Ahetu (194–213)

The beginnings of Mādhyamika doxography  45 v iii Criticism of the Lokāyata-s (194–214) ix Criticism of Īśvara (215–223) x Dṛṣṭi (224–229) xi Duḥkha (230–233) xii Buddha (234–239) xiii Seeing Pratītyasamutpāda and Seeing the Buddha (240–246) xiv Conclusion (247–265) s t

Niḥsvabhāvatā and śūnyatā (257–266) The Buddhas and the Bodhisattvas (267–360)

4  Śrāvakatattvaviniścayāvatāra94 a b

Introduction (1) Pūrvapakṣa: i The Śravaka-s’ Objection (2–7) ii Miscellaneous Objections (8–14)

c

Uttarapakṣa: Bhāviveka’s Response (15–740) i Introduction (15) ii The Buddha’s Cognition Is Non-Conceptual (16–25) iii The Path Is Not Real (26–33) iv The Mahāyāna Is the Buddha’s Teachings (34–58) v Ordinary Objects Are Relative (59–63) vi There Is No Identity (64–66) vii The Nature of Ultimate Reality (67–68) viii Sense Media Are Knowable and Exist in a General Sense (69–71) ix The Relative Nature of Mind and Phenomena (72–74)

5  Yogācāratattvaviniścayāvatāra95 a b c

Introduction (1) Pūrvapakṣa: Yogācāra Objection (2–7) Uttarapakṣa: Bhāviveka’s Response (8–114) i Introduction (8–9) ii The Ultimate as an Object of Cognition (10–16) iii The Ultimate as Consciousness (17–54) iv Imagined Identity (55–68) v Dependent Identity (69–84) vi Absolute Identity (85–114)

6  Sāṃkhyatattvāvatāra96 a

Pūrvapakṣa: The Sāṃkhya’s Propositions of Puruṣa and Prakṛti (1–4) i The Characteristics of Prakṛti and Puruṣa (1) ii The Relationship Between Buddhi and Puruṣa (2) iii The Sāṃkhya Theory of Liberation (3–4)

46  The beginnings of Mādhyamika doxography 1 2 b

Puruṣa’s Liberation (3) Prakṛiti’s Liberation (4)

Uttarapakṣa: Bhavya’s Critique of the Above Positions (5–65) i Critique of Puruṣa’s Characteristic of Having Consciousness and the Theory of Its Liberation (4–24) ii Critique of the Sāṃkhya Proof of the Existence of Prakṛti and Puruṣa (25–61) 1 2 3 4

Critique Critique Critique Critique

of of of of

the the the the

Existence Existence Existence Existence

of of of of

Prakṛti Puruṣa Prakṛti Puruṣa

(1) (1) (2) (2)

(25–42) (43–46) (47–51) (52–61)

iii Epilogue (62–65) 7  Vaiśeṣikatattvāvatāra97 a

Introduction: The theories of the Vaiśeṣika (D Dza 242a7–244a6) i The Characteristics of Ātman (2242a7–242b2) ii The Theory of the Six Padārthas (242b2–243b4) iii The Theory of Liberation (243b4–244a6)

b c

Pūrvapakṣa: The Vaiśeṣika Theory of Liberation (k.1, 244a6–244b2) Uttarapakṣa: Critique of the Theories of the Vaiśeṣika (kk.2–28, 244b2–250a4) i Critique of the Relationship Between Attributes. Manas, and Ātman (kk.2–14, 244b2–247a1) ii Critique of the Existence of Ātman (kk.15–22, 247a1–148b7) iii Critique of the Vaiśeṣika Theory of Liberation (kk.23–28, 248b7–250a4)

d

Conclusion: The Vaiśeṣika View Is Erroneous (k.29, 250a4–251a1)

8  Vedāntatattvaviniścaya98 a b

Introduction (1) Pūrvapakṣa: Vedānta Philosophy According to Bhāviveka i The Characteristics of the Self (2–17)

c

Uttarapakṣa: Bhāviveka’s Critical Examination of Vedānta Philosophy i Introduction (18) ii Refutation of Liberation as a Result of the Perception of the Self (19–24) iii Refutation of the Self as Agent (25–35) iv Refutation of the Self as Bound or Liberated (36–39) v Refutation of Cognition as the Intrinsic Nature of the Self (40–53) vi Refutation of the One Self as Having a Manifold Nature (54–58)

The beginnings of Mādhyamika doxography  47 vii Refutation of the Self as a Substance and a Support (59–64) viii Refutation of the Oneness and Existence of Space and Self (65–70) ix Refutation of the Self as Agent and Enjoyer (71–72) x Refutation of the Self as One, All-Pervasive, Permanent, Inexpressible and Inconceivable (73–85) xi The Vedānta-Darśana – A Completely Heterogeneous System (86–104) 9  Mīmāṃsātattvanirṇayāvatāra99 a b

Introduction (1) Pūrvapakṣa (2–17)100 i The Veda, Superior in Every Point to Any Knowledge, Entertain the Only Valid Path to Liberation: The Sacrificial Path (2–5) ii Because of the Inherent Quality of the Vedic Word (śabda), Its Unequalled Authority (pramāṇatva) Stands on Its Own (6–8) iii Liberation, Like Dharma, Is Produced by the Rituals (kriyā) Enjoined in the Veda (9–10) iv The Veda Are the Sole Privilege of the Godlike Male Twice-Born (11) v Anything Worthy of Interest Belongs to These Men (12) vi Reasoning Is a Dangerous Abuse of Knowledge Spoiling the Veda (13–14) vii Anything Which Objects to the Veda Is Spurious and Devoid of Authority (15–17)

c

Uttarapakṣa (18–167) 18 Purpose and Methodology 19–22 Rejoinder to (2): The Ritual Activity Prescribed in the Scriptures Is the Only Path to Liberation 23–24 Rejoinder to (3): Because the Veda Is Authorless, Not Having the Defects of Human Speech, It Is Reliable 25–42 Rejoinder to (4): The Veda Has No Remembered Author and Is of Uninterrupted Transmission 2 3 43

The Veda Has an Author and He Is Deplorable (25–31) Mantras Cannot Protect One From the Effect of Misconduct Even if Ritually Enjoined (32–42)

Rejoinder to (5): Without the Veda, How Is One to Know What Lies Beyond Sense Perception? 44–49 Rejoinder to (6–7): The Word, United With Meaning, Is Eternal/ Four Reasons for the Eternity of the Word 50–54 Rejoinder to (8): The Word Is a Separate Means of Knowing 55–58 Rejoinder to (9–10): Vedic Rituals Reveal the Apūrva and Cause Liberation 59–151 Rejoinder to (11): The Vedic Path of Gods and Ṛṣi-s Is to Be Avoided by Women and Śūdra-s

48  The beginnings of Mādhyamika doxography 1 2 3 4

The Gods, Their Path and Followers Are Stained by Misconduct (Pāpa) (59–73) The Yoga Invoking Manifestations of God Is Irrational and Helpless (74–93) The Gods Are as Impotent and Helpless as the Yoga Based Upon Their Worship (94–119) Other Impotent and Helpless Practices Found in the Veda (120–147) a b c d e

Bathing (10–126) Jumping Into Fire (127–128) Fasting to Death (129–131) Vegetarianism (132–138) Worshipping Trees and Inanimate Objects (139–147)

5

Even the Words of So-Called Gods and Seers Must Be Analyzed Based on Their Arguments (148–151) 152–154 Rejoinder to (12): Everything Comes From the Veda, Even the Four Goals of Life 155–158 Rejoinder to (13–14): The Inferential Reasoning of Debaters Spoils the Vedic Path and Inference Dangerously Leads the Blind in Unwarranted Directions 159–166 Rejoinder to (15–16): No One Is Omniscient, the Buddha’s Omniscience Is a Deception and The Buddha Is Human Hence His Discourse Is No Pramāṇa and He Is Not Omniscient 167 Rejoinder to (17): Any Discourse Objecting the Veda Is No Pramāṇa, Be It From the Buddha or the Jainas 10  Sarvajñasiddhinirdeśa101 a

Considering Omniscience (1–12)

11  Stutilakṣaṇanirdeśa a

In Praise of the Buddha (1–3)

1.5  The dialectical strategy of Chapter 9 The perspective I elaborate in the present analysis of MHK Chapter 9 aims at understanding the internal coherence of the chapter and its significance within the whole of the MHK, as a dialectical unit designed to communicate a precise knowledge, possibly of soteriological relevance. I insist that Bhāviveka was not merely interested in listing idiosyncrasies of religious competitors nor in displaying his knowledge of world literature but that he had an overarching design in mind: (1) to categorically refute the soteriological validity of the karma-mārga and its numerous ersatz (rituals, mantras, deity worship, object worship, Tantra, yoga, and so on)

The beginnings of Mādhyamika doxography  49 and (2) to stress the importance of reasoning (anumāna) and knowledge (jñāna) in regard to liberation. This design can be observed by examining the internal dynamic of the chapter along with its rhetorical strategies and doctrinal content. The underlying question of the present enquiry is as follows: Did Bhāviveka transmit any soteriologically relevant knowledge to his students through the dialectical exercise performed in Chapter 9, or did he only carry a debate for the sake of quieting an imagined opponent and displaying his dialectical skills? To ponder the previous question, one needs to analyze the overall structure of the MHK and retrace its argumentative development. In accordance with the vāda tradition, and in continuity with every dialectical unit of the MHK, Chapter 9 is divided into a pūrvapakṣa and an uttarapakṣa section: a prima facie argument of the opponent followed by the reply from the author. This pervasive structure of the MHK is typical of the dialectical model of exegetical analysis originally designed by the Mīmāṃsā, where an adhikaraṇa (the chapter of a text) is divided in a fivefold manner, recalling a dialogue between teacher and disciple: 1 the 2 the 3 the 4 the 5 the

presentation of a viṣaya (topic) to be discussed; raising of a viśaya or saṃdeha (doubt); pūrvapakṣa (prima facie argument); uttarapakṣa (the reply of the author); and nirṇaya (the final decision).102

The topic of Chapter 9 is the Mīmāṃsā worldview, valuating ritual acts above knowledge and meditation. The doubt is raised indirectly in the introductory verse, which qualifies the Mīmāṃsakas as the “shameless ones” (anapatrapa) who revile meditation and knowledge as the right path to emancipation. Naturally, one assumes that, from Bhāviveka’s perspective, there is a reason to doubt such a stance. But, beforehand, one needs to hear the reasons behind it. Thus, the pūrvapakṣa voices the claims of the Mīmāṃsā in sixteen propositions (pakṣa), from verse 2 to 17. Then, the uttarapakṣa assesses these claims in the remaining segment of the text, from verse 18 to 167. However, no separate nirṇaya is formulated at the end. One must find it finely threaded in the lines of the uttarapakṣa. For, this is the whole purpose of the discussion, as signified already in the title of the chapter: mīmāṃsātattvanirṇayāvatāraḥ (Entry into the assessment of reality within the Mīmāṃsā). While introducing the uttarapakṣa, at verse 18, Bhāviveka insinuates that his ‘assessment’ of the Mīmāṃsā’s position will be conducted methodically, without bias: Those seeking facts as they are will examine that [critical discussion], after having abandoned the poison of partiality, being experts in hermeneutic rules applicable to words and their meaning.103 This method of critical examination, engaged in for the sole purpose of elucidating an uncertain thesis, corresponds to the ideal Indian mode of debate, vāda, as

50  The beginnings of Mādhyamika doxography enunciated already in the first verse of the second chapter of the Nyāyasūtra of Gautama: Vāda consists in the formulation of a position and counter-position (pakṣapratipakṣaparigraha), in establishing and refuting through means of knowledge and reasoning (pramāṇatarkasādhanopālambha); it does not contradict the teachings of [one’s] school (siddhāntāviruddha), and it involves the application of the five members [of the inference] (pañcāvayavopapanna).104 Bhāviveka, however, adopts the threefold inferential method of Diṅnāga and not the fivefold one promoted by Gautama. Nonetheless, he invites his student to witness a debate, over the tenets of the Mīmāṃsā, conducted in a manner consistent in a Brahmanical context. The whole scene is staged as a critical but neutral dialogue. One can almost imagine a dramatic setting where a Mīmāṃsaka is actively discussing the validity of his worldview with Bhāviveka. Of course, as with Platonic dialogues, the author of the script has his own dialectical strategy already well in mind from the beginning. His fictitious interlocutor, though voicing legitimate views and objections, which add both philosophical substance and an appealing historical resonance to the scene, is masterfully made to admit the tragic conclusion that his views are ultimately inconsistent. As with Socratic dialogues, irony is to be expected. Casted within this narrative setting, the plot of the chapter, and of the entire MHK, is not a mere intellectual debate among dull scholars, but a dramatic process of conversion, of guided insight into truth, by which Bhāviveka’s student is initiated into the path of critical examination. The whole art of such dialectical śāstra-s is to veil the obvious fact that the opponent’s position is strategically presented in a way that ends up supporting the author’s own view, by repeatedly calling for expected replies to objections which gradually reveal the extent of the opponent’s shortcomings.105 This dialectical scheme can only succeed if the opponent is actually presenting a credible defence of his viewpoint. In such a case only can the alert student, or jury, confirm the defeat and be established in his conviction of cultivating a superior view. Perhaps one can also imagine a skilled debater realizing similar rhetorical prowess in front of a jury and appreciate the impressive level of intellectual mastery potentially involved. Yet, even in a literary format, not everyone can claim the same dialectical success. Poison is not skillfully turned into medicine by everyone. As Bhāviveka’s Mīmāṃsā caricature (avatāra) enters (avataraṇa)106 the stage, its pretentious (anapatrapa) attitude is immediately placed under the spotlight, as if the learned regisseur was showcasing a circus buffoon, intended to become the locus of mockery. Thereupon the mocked speaker pronounces his inaugural speech in sixteen points. Shinjo Kawasaki, serious as a pope, has summarized the given pūrvapakṣa in eight points.107 He was confirmed in his scheme by ‘discovering’ a topical terminology parallel to his own in Śāntarakṣita’s polemics against Kumārila. However, he was baffled that none of Bhāviveka’s buffoon’s actual utterances found resonance either in the Tattvasaṅgraha (TS) or in the Ślokavārttika (ŚV);

The beginnings of Mādhyamika doxography  51 not even in Śabara’s commentary on the Mīmāṃsāsūtra (MS). Thus, unamused, his historiographical habitus led him to suggest a third, albeit unknown, figure who would connect our knowledge of Mīmāṃsā with Bhāviveka’s simpleton. However, by shifting the paradigms of Kawasaki’s ‘philologico-centric’ model, finding meaning through external literature and linguistic norms, towards a rather hermeneutical one, relating parts of the text with the whole, and by postulating that Bhāviveka might have been a somewhat creative author after all, the ‘third figure’ of Kawasaki starts to behave as the equant in the Ptolemaic system. Though it might have some explanatory potential on its own, it is unnecessary to explain and understand the unfolding drama. By insisting on what should be expected of a ‘good-student’s’ refutation of the Mīmāṃsā, Kawasaki’s summary misses the rhetorical play of Bhāviveka’s own introductory sequence. On my part, I will investigate the nature of the pretentiousness (anapatrapatva) which Bhāviveka accuses his Brahmin ‘fool’ of. Under the pen of Bhāviveka, the Mīmāṃsaka states, 1 2 3 4 5 6 7

The Veda, superior in every point to any knowledge, present the only valid path to liberation: the sacrificial path (verses 2–5). Because of the inherent quality of the Vedic word (śabda), its unequalled authority (pramāṇatva) stands on its own (verses 6–8). Liberation, like dharma, is produced by the rituals (kriyā) enjoined in the Veda (verses 9–10). The Veda are the sole privilege of the godlike male twice-born (verse 11). Anything worthy of interest belongs to these men (verse 12). Reasoning is a dangerous abuse of knowledge spoiling the Veda (verses 13–14). Anything which objects to the Veda is spurious and devoid of authority (verses 15–17).

Confronted with the unblinking ‘truth’ (tattva) of the twice-born, even the most learned and benevolent doctor could only hope for a miracle to improve such eyesight (darśana). Reassuringly, Bhāviveka acquired suitable means in the Madhyamaka dispensary. Filled with commiseration, having polished his sharp analytical scalpel, he is ready to operate on the Brāhmaṇa. As a comforting gesture worthy of a seasoned health professional, Bhāviveka immediately engages in bridge building with his wise patient (verse 18), introducing his impartial therapeutic methodology as an interreligious dialogue suiting those who, like the Mīmāṃsakas, are ‘experts in reasoning on the meaning of words’ (śabdārthanyāyakovida). For, surely, the Brahmin opponent also seeks to know facts as they are (yathābhūtagaveṣiṇa), to gain that knowledge which, alone, is the active ingredient in the serum (v. 9.21: auṣadha) which Bhāviveka is about to introduce. However painless, the cure will not be any small affair. Besides Bhāviveka’s own disquisition on truth (tattva) in Chapter 3, the present dialogue is the longest section of the whole MHK. Aux grands maux les grands moyens!108 Fortunately, the chapter is well divided and can be followed with a minimum of guidance.

52  The beginnings of Mādhyamika doxography Different ways can be adopted to go about summarizing and analyzing Bhāviveka’s reply. Hyoung Seok Ham presented a threefold division of the uttarapakṣa which regroups its verses in three main arguments: 1 2 3

the Veda is not moral, the Veda is not rational, and the Veda is not omniscient.

These three counter-arguments are certainly part of Bhāviveka’s strategy to militate against the Mīmāṃsaka’s plea, but they miss a crucial element. Essential to Bhāviveka’s case against his opponent is his claim that the path of rituals, of any such activity (karma), be it prescribed in the Veda or not, is soteriologically impotent. As mentioned, for a jñāna-mārgin like Bhāviveka, liberation comes from knowledge. Hence, Brahmanical rituals, because they rest upon mere mechanical activity, are not emancipatory. The reason is clearly stated in verse 9.22: An activity, because it is [mere] activity, like ploughing, is not suitable [as a means] for attaining liberation. Either because, being devoid of intelligence, it consists of [mere] utterance, or because it is limited in time.109 Bhāviveka argues that rituals, by nature, are mindless repetitions of actions and that their results, born of a limited cause, are equally limited. In no way could such repetitive toil, like ploughing, cultivate liberating wisdom. It is crucial to understand this decisive argument, for it justifies numerous discussions which have puzzled scholars, because they engage with doctrinal elements which are not naturally connected with the Mīmāṃsā. These discussions are predominantly related to verse 11, where Bhāviveka’s opponent argues that the Vedic path is the sole privilege of the godlike male twiceborn. As can be noticed by looking at the content of the chapter, Bhāviveka’s reply to this assertion is the longest of all, going from verse 59 to verse 147, counting for more than half of the whole chapter. This section can further be subdivided into several important subtopics, a quick look at which suffices to indicate that Bhāviveka is moving far beyond what is traditionally designated as Mīmāṃsā territory. Here is my own division of the section: 1 2 3 4

The gods and their path and followers are stained by misconduct (pāpa) The yoga invoking manifestations of god is irrational and helpless The gods are as impotent and helpless as the yoga based upon their worship Other impotent and helpless practices found in the Veda a b c

Bathing Jumping into fire Fasting to death

The beginnings of Mādhyamika doxography  53 d e 5

Vegetarianism Worshipping trees and inanimate objects

Even the words of so-called gods and seers must be analyzed based on their arguments

It is known that the Mīmāṃsā defends an atheistic view. It is thus certainly not concerned with such gods as Viṣṇu, Śiva, and Brahmā, all of which prominently feature in this section. It is also not particularly invested in all the five practices listed here, like jumping into fire or even vegetarianism. Perhaps more importantly, it is not usually so directly associated with yoga. Therefore, it is legitimate to wonder why such heterogeneous elements are mentioned in this chapter. To account for this ‘anomaly,’ besides his earlier-mentioned threefold division of the text, Ham further breaks the whole chapter into an ‘official pūrvapakṣa,’ five ‘independent discussions,’ and ‘contextual discussions.’110 According to Ham, the ‘official pūrvapakṣa’ regroups the verses which directly discuss and refute the Mīmāṃsā. The five ‘independent discussions,’ all found within the reply to verse 11, do not aim at the Mīmāṃsā as it is traditionally known. They are introduced by a play on the verse’s wording. Whereas verse 11b states, “It is suitable that the Threefold [Veda] be avoided by women and śūdra-s, excluded from the meaning of the Veda,” these five arguments extend the meaning of “it is suitable that the Threefold [Veda] be avoided” (yuktaṃ yat tyajyate trayī), to suggest that is proper that the three Veda be avoided by all. Here again, Bhāviveka skillfully displays his irony. Concerning Ham’s last subdivision, the ‘contextual discussions,’ he presents them as further ‘rejoinders.’ They are introduced by words such as cet, yadi, and atha, a common practice in śāstra-s. Their target cannot be unproblematically assumed to be the Mīmāṃsā. Although this additional division of the chapter suits Ham’s analytical purpose, it has the disadvantage of suggesting that parts of the text are mere excursus. The advantage, however, is that by separating and presenting these discussions as subordinated to a main ideological debate, Ham manages to convincingly argue that the whole chapter can be said to mainly target the Mīmāṃsā, although it deviates on several occasions to include a heterogeneous group with which the Mīmāṃsakas share their trust in the Veda. On this point, I agree with Ham. The whole chapter is directed at Brahmins vested in Vedic lore and in ritual activity. However, fragmenting the chapter into discrete ‘independent’ sections has a consequence. It dissolves the rhetorical strategy of the dialectician, which is made unrecognizable by an analytical scheme dictated, again, by a historiographical need to clearly identify a legitimate target of criticism. I suspect that this ‘dissolution’ explains why Ham did not insist on the fact that the uttarapakṣa, in arguing that the Veda is not moral, not rational, and not omniscient, stresses that it is fact soteriologically irrelevant. As will be seen within these pages, Indian philosophical doxographies are explicitly concerned with the soteriological efficacy of the views they list. Thus, I argue that all the discussions engaged in within the ninth chapter work as many

54  The beginnings of Mādhyamika doxography dialectical streams and rivers culminating in a single rhetorical wave, resulting in the total discrediting of the path of activity (karma-mārga) prescribed by the Veda and, in extensio, of anything claiming Vedic filiation. It is a direct blow to the ‘pretentiousness’ of the Brāhmaṇa. According to Bhāviveka, not only is the Brāhmaṇa path obscured by self-­ interest, but it is also sterile. Being the product of a pious monk, the MHK does not engage in idle chatter (sambhinnapralāpa). Its author is not interested in collecting exotica or in cataloguing the strange views and habits of his opponents. Bhāviveka is not a museum keeper. Every element analyzed within his discussion has a function. It is meant to contribute to an overall rhetorical strategy. Ham’s ‘independent discussions’ may not be so independent after all. I read them as logical developments in a discussion which aims to expose and to ridicule the total impotency of Vedic ritual activities and the groundless arrogance of their patrons. Whether the accusation is true, at this point, is rhetoric.

1.6  The rhetorical and thematic progression of Chapter 9 Before moving forward to analyze the reply to verse 9.11, let us first move backward and follow the rhetorical path signposted by Chapter 9 itself. The point here is not to enter every discussion encountered along the way, as inviting as they might be, but to provide an overview precise enough to observe Bhāviveka’s rhetorical strategy at work. Significant details of the dialectical panorama will nonetheless be pointed out whenever relevant. An economical approach to survey the entire uttarapakṣa is to proceed along the lines of the seven-point summary I have just given. It subsumes the logical and natural progression of the argument. More precisely, making use of the chapter’s table of contents, before presenting my analysis, I list the section’s headings one by one, based on each verse of the pūrvapakṣa which the section aims to refute. My analysis comments on meaningful elements within these rebuttals. Pūrvapakṣa i

The Veda, superior in every point to any knowledge, entertain the only valid path to liberation: the sacrificial path (verses 2–5)

Verses summary: (2) The ritual activity prescribed in the scriptures is the only path to liberation (3) Because the Veda is authorless, not having the defects of human speech, it is reliable (4) The Veda has no remembered author and is of uninterrupted transmission (5) Without the Veda, how is one to know what lies beyond sense perception? Analysis of the Uttarapakṣa (v. 19–43) Regrouping verses 19 to 43 of Chapter 9, this section occupies a significant portion of the entire uttarapakṣa. It refutes crucial doctrinal claims of the Mīmāṃsā

The beginnings of Mādhyamika doxography  55 proper: (1) the authority of scriptures (āgamatva); (2)the validity of Vedic activity (kriyā); (3)the non-humanity (apauruṣeyatva) of the Veda; (4)which are a revelation without an author (akartṛtva); (5)the uninterrupted transmission (saṃpradāyānupaccheda) of the Veda; and (6)its epistemological superiority, said to perceive apūrva.111 In this section, Bhāviveka engages his opponent head on, like a hunter, aiming straight at his core beliefs. Like a garuḍa, he seeks to first behead the snake, to leave the remaining parts of its doctrinal body without much venom. Most of the reply to (2) has already been discussed within these pages. Therein Bhāviveka points out that if his opponent argues that liberation is conditional to the execution of what is prescribed in scriptures, then the problem arises as to what determines valid scriptures. A more reliable criterion, Bhāviveka suggests, would be to first determine what is truth (tattva), based on reason. Thus, the validity of scriptures can be ascertained through their reasonability, the meaning of which is to be further investigated through reasonable means (anumāna). The dialectical tactic here is to subordinate the authority of scriptures (āgama), the only asset of the Brahmin, upon which rests his doctrinal edifice, to the one of reasoning (yukti), the privileged weapon (śastra) of the Mādhyamikan dialectician (śāstrika); engaging the Brāhmaṇa on a battlefield which allows him no strategic advantage. By insisting on the fact that liberation comes from knowledge and that ritual acts are devoid of intelligence, Bhāviveka proceeds to symbolically emasculate his opponent. Among Brahmins, by the time of Bhāviveka, a distinction came to be established between those texts which insisted on the ‘action-part’ (kriyā-kaṇḍa) or on the ‘knowledge-part’ (jñāna-kaṇḍa) of the Veda, respectively distributed between the Brāhmaṇas and the Upaniṣads. Chapter 8 of the MHK, on Vedānta, already dealt with those given to the jñāna-kaṇḍa: the Vedāntins. In Chapter 9, the whole focus lies on those who favour the kriyā-kaṇḍa, the part of the Veda dealing with ritual activity: the Mīmāṃsā. Overall, by discrediting the ideological (jñāna) and religious (karma) business of Brahmins, in the MHK, Bhāviveka ridicules their entire livelihood and social significance. The consequences of his accusations, when translated in their sociohistorical context, are severe. They unambiguously challenge the authority of a powerful community: on intellectual, ritual, and moral grounds. Surely, not everyone found Bhāviveka’s rhetoric amusing. Here, history and philosophy intersect within the MHK, at the frontline of an uncompromising dialectical struggle for truth (tattva). At least philosophically, Bhāviveka’s rhetoric amounts to a declaration of war. With his rejoinder to (3) and (4), Bhāviveka proceeds with his logical skirmishes, one argument after the other. In tune with the Indian dialectical and logical tradition of Nyāya, but more particularly in the footsteps of the Buddhist Diṅnāga, Bhāviveka points out the logical fallacy in each argument of the Brāhmaṇa; leaving it defeated, collapsing on its own deficient reasoning (vitarka).112 For Bhāviveka, such a fate is but a tragic testimony of the blunt intelligence of the Brahmins’ headmasters, the Veda and their seers (ṛṣi-s). Bhāviveka’s reasoning is categorical and coldly logical. If the proof offered by the opponent does not resist investigation, it is defeated.113 Logic, unlike grammar, knows no compromise. It cuts in two: true or false. This is the principle of discrimination (viveka).

56  The beginnings of Mādhyamika doxography For example, at verse 24, Bhāviveka argues, Else, if the validity of the Veda is to be established on account of having no author, then either because the reason is too narrow, or because its meaning is unestablished, it should be [refuted].114 Here, Bhāviveka points out that the reason (hetu) provided by the Mīmāṃsā – ­linguistically recognizable by its ending in pañcamī-vibhakti (apādāna/­ablative) – “because of having no agency,” is defective on two counts. The assertion of the opponent is meant to support the claim that the Veda are reliable. Thus, the Mīmāṃsā resorts to this argument to prove that the Veda are reliable because they are not like human utterances: prone to error. The problem with the given reason (hetu), according to Bhāviveka, is twofold. (1) First, it is too narrow. It does not pervade the entire scope of ‘reliability,’ since there are instances of reliable utterances which are not authorless (akartṛ), as in the case of the speech of goldsmiths and so on – mentioned at verse 23. Not all reliable speech is authorless. Hence, the reason does not pervade all instances of ‘reliable speech.’ Or, again, the reason is faulty because it is (2) a false evidence. Its reasoning is unwarranted (asiddha), since, for those who are not partial to the Mīmāṃsā, there is no obvious connection between ‘authorlessness’ and ‘reliability.’ Highlighting flaws in the relation between the reason (hetu) asserted by the opponent and what is to be proven (sādhya) by him is the nerve of dialectical disputation (tarka), the yoga of philosophers. If the relation is defective in one way or another, according to established logical criteria, then it is unfit (ayukti), and what needed to be proven is not established (asiddha).115 The asserter (vādin) of such erroneous pronouncements (vāda-s) loses his case. Not every Indian philosophical doxography engages in these minute refutations. Many, like Haribhadra’s ṢDS, for example, only list the reasoning supporting various views without criticizing them. Another interesting feature of this section is found in the second half of the rejoinder to (4), discussing the alleged protection of mantras and scriptures. To those who suggest that mantras or scriptural approval prevent the undesirable consequences ensuing from the performance of reprehensible actions, like violence or drinking alcohol, Bhāviveka replies that it is not so. An action (karma) invariably produces a fruit (phala). At verse 33, Bhāviveka compares those who use mantras to protect themselves from the unwholesome consequences of killing, to rapists who subdue their victims by means of mantras, extortion, powders, and so on (mantrakarṣaṇacūrṇādya). The protection of mantras does not cover such harmful activity, agues Bhāviveka. Thus, the reason is too broad. Further, Bhāviveka points out that scriptural authority does not alleviate the consequences of action, just like it does not prevent one from being drunk when enjoying liquors. Using mantras to protect oneself from the consequences of murder, an extreme indeed, amounts to being murderous. It brings destruction upon oneself. If one argues that such harm is enjoined by scriptures and is thus performed for a noble purpose, then this proposition amounts to an argument that liberation itself can be gained out of

The beginnings of Mādhyamika doxography  57 violence, which is nonsensical, at least according to Bhāviveka. The same would be the case even if all scriptures were to prescribe murder. Again, Bhāviveka is categorical. The discussion continues further, discrediting ritual killing and the alleged protection of mantras against harmful acts. Bhāviveka observes, “(Ritual) action certainly produces a corresponding effect in a visible location.”116 In other words, every act causes a reaction, affecting the actor. The same logic applies to the use of intoxicants during rituals. The rhetorical effects of Bhāviveka’s arguments are twofold: (1) they discredit further the reliance upon action (karma) for the purpose of liberation (mokṣa) and (2) they point out that not only are such acts soteriologically useless, but also they are potentially harmful to oneself, since they may carry undesirable consequences in the future. Of course, such arguments target the traditional promoters of Brahmanical sacrifices, involving ritual violence, but their consequential scope is broader. For example, they also target the emerging practices of Tantra, particularly the Śaiva-tantra-s, which Bhāviveka will single out later on (v.  62a: siddhi-yoga). Such arguments are what Ham would label ‘contextual discussions.’ However, here again, one can observe how such a discussion is legitimate within a broad refutation of the path of action (karma-mārga). Whatever is its doctrinal basis, activity alone is soteriologically impotent and potentially harmful in the realm of rebirth. Similarly, the repetition of mantras, being mere babbling, a verbal act, cannot prevent or alter the consequences of action.117 Pūrvapakṣa ii Because of the inherent quality of the Vedic word (śabda), its unequalled authority (pramāṇatva) stands on its own (verses 6–8) Verses summary: (6) The word, united with meaning, is eternal (7) There are four reasons for the eternity of the word (8) The word is a separate means of knowing Analysis of the Uttarapakṣa (v. 44–54) This section, ranging from verses 44 to 54, refutes the main reasons given to justify why Vedic utterances are a superior source of knowledge: (1) because of their alleged eternity; (2) because of being an independent pramāṇa; (3) because they reveal the apūrva. Hence, the discussion unfolds on predictable lines of debate against the Mīmāṃsā. The most relevant part of this section has been discussed already: the relation between scriptures (āgama), inference (anumāna), and conventional knowledge (saṃketa).118 The rhetorical result of Bhāviveka’s argument, here, is to demonstrate why Vedic scriptures function like any others and, thus, why they must be reasoned out in order to be understood and vindicated. In other words, the Veda are nothing special.

58  The beginnings of Mādhyamika doxography Pūrvapakṣa iii Liberation, like dharma, is produced by the rituals (kriyā) enjoined in the Veda (verses 9–10). Verses summary: (9–10) Vedic rituals reveal the apūrva and cause liberation Analysis of the Uttarapakṣa (v. 55–58) As its title indicates, this short section of four verses (55–58) refutes the relevance of rituals concerning dharma and mokṣa. The rejoinder to verse 9 stresses that apūrva is neither the result of (ritual) action, nor is it permanent. Hence, it is not instrumental in bringing about liberation. The remaining verses elaborate on why the path of action is soteriologically impotent: (1) because the result of action is impermanent (v. 56) and (2) because it is produced out of desire (v. 57). Liberation, being the permanent end of desire, suffering, and so on, cannot logically be the result of polluted and limited acts – especially of the ritual kind: always motivated by desire. Bhāviveka’s stance on action is clear: the karma-mārga does not conquer death. The ritual path is frequented by those who are submissive to others, whose mental faculty is incompetent in critical thinking (v. 58b: parapraṇeyairjuṣṭo ‘yaṃ vicārākṣamabuddhibhiḥ). Ritualistic thinkers are said to be submissive because they blindly follow tradition instead of using their judgment to critically analyze scriptures and reality, for the sake of seeing clearly. Pūrvapakṣa iv The Veda are the sole privilege of the godlike male twice-born (verse 11) Verses summary: (11)  The Vedic path of gods and ṛṣi-s is to be avoided by women and śūdra-s Analysis of the Uttarapakṣa (v. 59–151) We are now coming back to the famous reply to verse 11, the longest section of Chapter 9 (v. 59–151). To best capture the irony of Bhāviveka’s reply within those verses, it is relevant to recall the initial assertion of the Mīmāṃsaka who triggered it: Desired by the learned and practised by ṛṣi-s and gods, is the ancient propitious path. It is suitable that the Threefold [Veda]119 be avoided by women and śūdra-s, excluded from the meaning of the Veda.120 What is important to notice here is the final formula of the Sanskrit verse: “It is suitable that the Threefold [Veda] be avoided” (yuktaṃ yat tyajyate trayī). This

The beginnings of Mādhyamika doxography  59 formulation is repeated five times within this section: at verses 59, 94, 120, 127, and 139. Each of these verses refers back to the same formula of the Mīmāṃsaka and introduces a new subdivision within the refutation of the argument. Ham, based on TJ, divides the section according to these five formulaic occurrences, whereas my subdivisions rather aim at highlighting the rhetorical coherence of the whole unit by delimiting its progressive sequence of argumentation. As mentioned previously, none of these discussions are directed at conventional Mīmāṃsā doctrines or practices. They rather aim beyond; to include the broader scope of Vedic lore. This whole portion of Chapter 9 can be subdivided into three coherent argumentative units which, in turn, extend the assessment of the previous sections, which were targeting the three Veda. In other words, the strategy here is to show that what was said of the Veda proper also applies to the broader traditions which depend upon them. In brief, these traditions are accused of (1) promoting unwholesome conduct; (2) being irrational; and (3) being soteriologically impotent. The last argumentative unit lists a series of Vedic sanctioned practices which Bhāviveka accuses of being soteriologically useless. My analysis presents each denounced practice in a separate treatment. The last argument of the chapter restates the reply to pūrvapakṣa (ii), in order to stress that, even if they be from godlike beings, words must be analyzed based on the reasonability of their arguments. In categorical terms, based on anumāna. The rhetorical effect of the entire reply to verse 11 is to extend the rebuttal of the Mīmāṃsā to the entire Vedic lore. In the end, for Bhāviveka, the Mīmāṃsā is the karma-mārga, the unintelligent repetition of conventional action. Such a path of activity does not lead to liberation and is potentially harmful, even in the realm of rebirth, because of its sanction of dangerous conduct and irrational views. It is within this broad discussion that Bhāviveka’s criticism of what could be called ‘Vedic’ or ‘Brahmanical’ yoga appears. Yet, one must place every piece of the dialectical puzzle in place before arriving there. The gods, their path and followers are stained by misconduct (pāpa) (v. 59–73) Herein, the three gods of the Hindu Trimūrtī, Brahmā, Viṣṇu, and Śiva, are introduced as brahma-keśava-śūlin. They are all said to suffer from various afflictions, justifying the first ironic reformulation of verse 11: “after having seen the conduct, consisting of afflictions, logically that Vedic-Triad is to be avoided” (v. 59b). In opposition to the misjudgment of the gods, the figure of the ‘learned one’ is cast as one who burns misconduct with knowledge, as fuel with fire (v. 60a). It indicates that the gods are devoid of intelligence, and that knowledge alone, not divinity or ritual action, confers sanctity. Bhāviveka insists that all the Brahmanical gods are corrupted. Brahma lusts after women. Śiva is consumed by anger. Kṛṣṇa, also named Keśava and Hari, is nothing more than a parasite on beings. Stealing wealth and wives, indulging in liquors, the one who is said to be the victor of all foes is in fact ironically outmatched (jita) in his conduct, by savages and thieves (v. 67).

60  The beginnings of Mādhyamika doxography The word taskara (thief), employed to describe Hari, in this verse, is the same one from the introductory section of the MHK (1.10): Those having fallen asleep with the slumber of delusion after having drunk maddening liquors, whose collection of merit is ruined to naught by the thief of dubious reasoning.121 There seems to be a direct correlation between the two verses, where Bhāviveka accuses Hari, the alcoholic, of treacherously fooling his believers and inciting unwholesome conduct. If one was to apologize for such ungodly actions by pretexting that they are but mere devices meant to protect dharma, then Bhāviveka replies, However, how can the one who is a slave to lust, who protects the world out of desire, be perfect? But, if it were out of compassion, how can it then be that he has led the world astray with illusory deception?122 Fully aware of Vedic lore himself, Bhāviveka seems to amuse himself by pointing out its numerous shortcomings. In no way can the unwholesome conduct described therein be said to be that of the highest being (puruṣa-uttama). Simply put, Vedic gods are unworthy of veneration. What is more, those, like the Śaivas, who profess that the ‘yoga of power’ (v. 62a: siddhi-yoga) can protect them from the downfall of misconduct, are dangerously leading others astray: “By such claims, [made] by those who lost [sight of] the right path, others are ruined as well.”123 The gods are not simply unworthy of respect, but their teachings are also dangerous. The yoga invoking manifestations of god is irrational and helpless (v. 74–93) After having substantially damaged the reputation of Hari, Bhāviveka directs his satire at another manifestation of god, referred to as Śiva,124 which some opponent presents as an altogether different display of the one god, an ‘auspicious’ one (the literal meaning of śiva), conferring liberation on mere sight. This other manifestation of Hari, Bhāviveka tells us, is described by its devotees in a manner recalling the Nāgārjunian tetralemma (catuṣkoti): This [form]125 is viewed as not being and not non-being, it is not both being and non-being, therefore, it is beyond being, non-being, and both being and non-being.126 However, Bhāviveka ridicules the idea that such simultaneous possibilities are logical for a god who takes on numerous bodily forms (the avatāra of the fish and so on) and who is said to have at least three different manifestations (Brahmā, Viṣṇu, and Śiva). And since many of these incarnated forms are involved in reprehensible acts, how could this god be pacified? If he were different from these

The beginnings of Mādhyamika doxography  61 other forms, how could his divine form also be Kṛṣṇa (acyutā)127 (v. 76)? With verse 77, the Mādhyamikan dialectician pushes the ridicule to the limit, calling the consort of the god a whore (bandhakī)128 who sleeps with different husbands, depending on which manifestation of god, Hari or Hara, is there to greet her. Respinning the tetralemma in his own logic, Bhāviveka concludes that, in fact, the Vedic god is Neither the cause [of the word], nor different [from the world], neither immortal nor even born.129 In brief: This so-called divine manifestation is nothing at all. Or, rather, it is a pure fabrication (v. 83a: klṛpta) invented to conceal the disturbing plight of Vedic lore. Hence, Bhāviveka rhetorically asks, how can those longing for liberation, having that [form] as mental-object, [ever] be liberated?130 Again, Bhāviveka insinuates that the gods are nothing but deception and that relying upon them is futile. Being fabrications themselves, they surely did not create the universe, nor can they save anyone. An opponent then suggests that devotees obtain liberation by engaging in yogic practices reminiscent of those found in the Bhagavad Gītā (BhG), in the Upaniṣads, and in the Yogasūtra-s (YS): If [the opponent argues that] by withdrawing together the sense-organs [away] from the sense-objects, like a turtle its limbs, by uttering the Oṃ syllable with recollection, his devotee is liberated.131 These practices are worth examining closer. The image of a turtle withdrawing its sense-organs appears in BhG 2.58, where the Sanskrit formulations kūrmo ’ṅgāni-iva and indriyāṇi-indriya-arthebhyas are identical to those of Bhāviveka.132 Therein, the simile comes amidst a series of descriptions concerning the inner composure of the one whose “wisdom is steady” (BhG 2.55: sthita-prajñā). The idea of withdrawing inward likely recalls the cosmogonic act of god withdrawing the universe within himself after its creation, as mentioned in the Śvetāśvatara Upaniṣad (ŚU: 6.3–4).133 In mythology, the turtle (kūrma) is the second avatāra of Viṣṇu, which followed the one of the fish (matsya). Perhaps inspired by these mythological motives, the form of the turtle was later imitated in a yogic posture named kūrma-āsana, mentioned in various texts on yoga, including the famous fourteenth-century Haṭhayogaprādīpika(HYP). However, here, Bhāviveka does not refer to a mere bodily posture, but to the disciplined process of sense mastery, disengaging the mind from external objects, what the YS refers to as pratyāhāra,134 the fifth of the eightfold limbs of the yogic path, often subsumed in later yoga literature within the breath-control techniques of prāṇāyāma. It is to be noted that the YS bhāṣya on pratyāhāra does not use the metaphor of the turtle, but rather refers to bees (makṣikā) following their “king”

62  The beginnings of Mādhyamika doxography (madhukārarājan) wherever he goes. In any case, sense mastery is surely an essential component of the Vedic stream of yoga. As for the Oṃ syllable (oṃ-kāra), a pan-Vedic (and modern Hindu) trope also referred to as the Ṛg-Vedic praṇava, and as the udgitha of the Sama-Veda,135 whence it takes on a solar symbolism, it was first part of Vedic liturgies, within the Saṃhitā-s,136 before being mentioned as the focus of recollection (smṛti) in the Upaniṣads. Such contemplative mentions of Oṃ are also frequent in the Āraṇyaka-s and the Brāhmaṇas. In the ŚU (1.13), for example, it is said that “the Self has to be seized in the body by means of the praṇava (the syllable Om),”137 somewhat implying that recollection depends upon the inner withdrawal of the senses just mentioned. As for the BhG, Finnian McKean Moore Gerety tells us, the Gītā’s treatment of OM is broadly “Upaniṣadic.” This suggests that the composers of the BhG draw on a common stock of lore: the double meaning of akṣara; the correlation of OM and brahman; OM as the essence of the Vedas; and the preeminence of OM in hierarchies of various sorts.138 Thus, Bhāviveka could have picked the reference to the sacred syllable from a wide array of Brahmanical literature. However, in the present theistic context, following sense withdrawal, one can assume that the author has in mind people engaging in yogic practices of the type mentioned in the YS, where the praṇava also appears: “His name is the resounding syllable” (YS 1.27: tasya vācakaḥ praṇavaḥ). Herein, the reference is to the lord (īśvara), towards whom devotion/ attention (YS 1.23: praṇidhāna) can grant meditative achievement (samādhi). Reference to īśvara is as problematic in the YS as in a doxographical chapter on Mīmāṃsā (MHK). Yet, it carries the same contemplative function for the same people. Following these reflections, Bhāviveka asks how a person relying on fraudulent gods could ever be liberated. The opponent answers that liberation occurs through a theistic practice of yoga involving sense withdrawal and contemplation on the divine, such as the practice involving the oṃ-kāra. Expectedly, Bhāviveka is unimpressed. His refutation of yoga prefigures the one attributed to Śaṅkara in the Pātañjalayogaśāstravivaraṇa (PYŚV). It is worth examining: Liberation caused by the sight of Hari is not possible for Hari’s devotees, because it involves conceptualization and memory; for example: it is caused by the sight of Hari. A cognition free of conceptualization is [also] not granted to lead to the liberation of a yoga practitioner.139 Since [already] a cognition derived from the perception of characteristics is wrong, how much more an imaginary [cognition]!140 Herein Bhāviveka argues that because the visualization of Hari is achieved by such mental means as imagination (vikalpa) and recollection (smṛti), both act as an instrumental cause (nimitta). Even if the said visualization were somehow non-conceptual (nirvikalpā), liberation could not occur. The reason is that, for

The beginnings of Mādhyamika doxography  63 Bhāviveka, liberation is not the product of a cause, or else it could not be permanent. It is not arrived at by any causal means (yoga) whatsoever. This is likely why this sudden refutation of ‘Vedic Yoga’ appears within a chapter on Mīmāṃsā. Bhāviveka seems to have assimilated the practice of Vedic Yoga with the ritual performances of the karma-mārga, a path which he sees as being followed by an unintelligent (v. 9.22: adhītva) and submissive (v. 9.58: praṇeya) people who, because they are incompetent in critical reflection (v. 9.58: vicāra-akṣama-buddhibhiḥ), believe in the soteriological efficiency of action (karma). The critic of Bhāviveka precedes the one of Śaṅkara, who was not completely hostile to Yoga, as we shall see when discussing the SDS, and who even advocated the dhyāna-yoga meditation on the Oṃ syllable, for persons of “slower understanding” (!). To that effect, Halbfass observed, Śaṅkara’s treatment of Yoga practice is not so much an extension of his critique of Sāṃkhya, but an expression and application of his general attitude towards “works” (karman), and more specifically, of his rejection of the “work orientation” of the Pūrva-Mīmāṃsā. The “mental acts” (mānasī kriyā) that constitute yogic meditation are not physical activities motivated by personal desires, and they are not rituals in a literal sense. They are nevertheless “works,” and they are oriented towards results. They are a part of that network of means and ends which keeps us in saṃsāra. Bondage itself is of the nature of means and ends: sādhyasādhanalakṣaṇo bandhaḥ.141 Again, it appears that the great Vedāntin was borrowing from the Buddhists, even in his treatment of competing views (darśana). For, involving Yoga in the critic of the Mīmāṃsā is exactly what Bhāviveka did three centuries before him. Yet, while Bhāviveka completely rejected the authority (āptatva) of the Veda, Śaṅkara rather sought to defend and promote it. This discussion also leads me to disagree with Qvarnström who, as I mentioned in the introduction, argued that one common denominator of Indian philosophical doxographies is to be found in their “lack of criticism directed towards yoga.”142 Yoga is undoubtedly discussed within the MHK, but it does not appear where it would normally be expected, in an “independent discussion” (even if Ham would like to see the present discussion as such). Yet, this classificatory ‘anomaly,’ if it is one, is explainable. Moreover, Yoga might not be discussed in Haribhadra’s ṢDS, but the Jaina author dedicated a full doxography to Yoga alone, in the Yogadṛṣṭisamuccaya (YDS).143 One can hardly suggest neglect here. As for the SSS of ‘Śaṅkara,’ it unambiguously lists Yoga among the darśana-s. Hence, one is puzzled by Qvarnström’s claim. At the end of this section, when Bhāviveka claims that the many manifestations of the one god are irrational and helpless, he further accuses such gods of behaving like savages. They are unconcerned with agency (kāraṇa), perhaps the sole consideration which could have led them to a reflection on truth. Causation, according to Bhāviveka, is the main problem of Vedic business (karma). Moving on, the author makes fun of the idea that deplorable gods are the guardians of

64  The beginnings of Mādhyamika doxography dharma and that they altogether are the creators of the universe. If they are all one, he says, then how can even Brahmā be a Brahmin-killer (v. 9.90: brahmā api brahmahā katham)? To ridicule the gods and yogas which rely upon them, Bhāviveka concludes, Because of the deficiency of the claim about that (suffering), [found in] the teachings and so forth of the Veda and of Yoga, these are certainly not able to protect dharma by teaching the true dharma.144 The compound veda-yoga-upadeśa, which I translated as a genitive tatpuruṣa followed by a dvandva, along the lines of Lindtner – “the teachings and so forth of the Veda and of yoga,” could also be rendered as a genitive tatpuruṣa followed by a bahuvṛhī samāsa, a compound meaning “the teachings [and so forth] of Vedic Yoga,” insisting on the provenance of the problematic Yoga. However, the Sanskrit stanza appears to be defective. It lacks a proper subject. The Tibetan commentary provides little help in this regard. In any case, there is no doubt concerning the mention of yoga within Chapter 9. It is clearly rejected as soteriologically sterile, on the same ground as all Vedic rituals. The gods are as impotent and helpless as the yoga based upon their worship (v. 94–119) This discussion is in direct continuity with the previous one. It is introduced by the ironic formula “It is suitable that the Threefold [Veda] be avoided” (yuktaṃ yat tyajyate trayī). This conclusion is derived from the fact that, somewhat like Brahmā’s net, mentioned in the Brahmajālasutta of the Buddhist Dīghanikāya, where sixty-two beliefs clung to by various ascetics are discussed, the three Veda are said by Bhāviveka to be assembled by a net of fictions (kalpanājālakalpita) devoid of rationality. A recurrent theme of this section is that of the “impotent” (anīśvara) lord (īśvara), which is mentioned three times: at verses 99, 100, and 107. My translation of the term anīśvara aims at capturing the ironic context created by Bhāviveka. Whereas the lord is usually said to be a puruṣottama (v. 9.73), the “Highest Self,” Bhāviveka portrays him as totally impotent. Just as Bhāviveka argues that the lord cannot create the universe, logic suggests that he can neither create the self (ātman) nor has any power over karma. He cannot even be said to be able to perform any action, lest he would be liable to roast in hell (v. 9.103). The whole impotency irony culminates in verse 107. There, Bhāviveka ridicules Vedic mythology by observing that, even if the universe was god’s creation, manifested for the sake of his amorous sporting (with bandhakī-s – prostitutes, the same formula as in v. 77), because such pleasures depend on self and others, the lord must be impotent – on his own.145 What lord of the universe would depend upon another for his own pleasure? In other words, the lord is a libidinous slave of lust, and an impotent one. Overall, he has no lordliness and is incapable of helping anyone in alleviating suffering.

The beginnings of Mādhyamika doxography  65 With verses 108–109, Bhāviveka moves from irony to calumny, with a praise to Rudra/Śiva, which suggests that not only can he not save beings from suffering, but also he even takes delight in their plight: Homage to him, Rudra (the Horrible), whose name the meaning is suitable, he who is delighted with joy from meandering beasts frightened of having to feed upon one another; with those in hells, afflicted by the pain of grinding, slicing, burning and so on; with humans tormented by birth, old-age, sickness, fear, sorrow, and exhaustion.146 The previous section had insisted on Hari’s shortcomings, now it is Hara’s turn. With two couplets recalling the lamentations of King Solomon in Qohelet, Bhāviveka points out how unjust this god must be who ‘created’ such an absurd existence: That the miserable are rich, while the virtuous are eating the food of others, and that those of bad conduct are in heaven, [this] is an apparent act of god. That those who are rich in virtues are short-lived while villains live long, and that those of little wealth are donors, [this] is an apparent act of god.147 Surely, for Bhāviveka, this Śiva is an ominous manifestation devoid of piety. To this onslaught against his chosen divinity, the Brahmin opponent retorts that perhaps some Buddhists are happy, even if they don’t venerate god, but why are some not? And why are the bhākta-s following the precepts of god, not wrongdoers (v. 112)? Again, the question begs for a ready answer: because of causality, the engine of karma (v. 113). The same causality which “refuted the creative agency of Brahmā and Kṛṣṇa.” Causation is the main argument against all of those who claim spiritual liberation through ritual performances. The same principle of causality also refutes the yoga invoking Śiva, which Bhāviveka depicts as follows: The doors of the mind controlled, having fixed one’s mind on Shiva, in that manner, one meditates on the Oṃ syllable, holding one’s concentration on to heart. Having first gathered one’s mind, by practising concentration on the earth and so on, when the lord is pleased, one approaches the end of suffering. That too is falsehood.148 The passage and its refutation restage the previous rebuttal of the yoga invoking Hari. This time, Bhāviveka insists on the mental activity which subsumes vikalpa and smṛti, inherent in this śaiva-yoga. So long as mental cognition arises, liberation does not (v. 116). Moreover, the mere sight of the “motionless” (sthāṇu), an epithet of Śiva recalling his austerities and inviting Bhāviveka’s humour, cannot logically grant liberation. Neither can the mere sight of a motionless trunk (sthāṇu). Because it involves perception, which, in turn, involves mental cognition and causation. Mere perception (pratyakṣa), just as mere view (darśana), does not

66  The beginnings of Mādhyamika doxography confer liberation (mokṣa). Thus, no yoga whatsoever, invoking whatever aspect of the divinity, leads to the end of pain. Devotion is no soteriological means (yoga), as stated in the concluding verse of the section: With this, the remaining proponents of Brahma, Viṣṇu and the self, have been refuted. Because it is inappropriate, one should not entertain the idea of directing one’s mind towards God and so on.149 As with the general verdict concerning the liberative efficiency of the karmamārga, the pious petitioning of god through yoga is not only declared useless, but also completely illogical. God is simply impotent. No help whatsoever is to be expected therefrom. Now, to make sure that no one within the Brāhmaṇa is spared, Bhāviveka lists a series of other Vedic activities which fare no better than theistic yoga. Other impotent and helpless practices found in the Veda: Ritual bathing This section begins with the ironic formula “It is suitable that the Threefold [Veda] be avoided” (yuktaṃ yat tyajyate trayī). The verdict comes after having mentioned the dubious practice of purifying misconduct with waters (v. 120a: pāpaprakṣālanaṃ cādbhiḥ), while bathing in holy rivers, for example, an activity which Bhāviveka ridicules. He compares it to trade: as if selling and buying virtue and so on (v. 120a: śubhaādikrayavikrayam). Moving into more rational grounds, Bhāviveka refutes the logic of ‘trading’ merit (puṇya): Certainly, the give-and-take of merit and so on is not proper, because of [merit] being related with the mind, just as with pleasure and pain.150 In other words, because it is mind-based, and because it is formless (v.  125: amūrtatvād), one’s merit or demerit cannot be exchanged by mere physical activity (karma). Moreover, Bhāviveka argues, “It is known that misconduct cannot be transferred to a different stream [of consciousness]” (v. 125a: santānāntarasaṃkrāntaṃ na pāpamiti gṛhyate). One is then led to wonder how Bhāviveka felt towards the Buddhist practice of “transference of merits” (puṇya-pariṇāmanā).151 Based on the present discussion, one can infer that he did not approve, in principle. Perhaps, on this issue, the TJ could provide further information. Ritual fire suicide Having dealt with the practice of bathing, the ironic formula (yuktaṃ yat tyajyate trayī) comes back once more to denounce the practice of those who jump into fire to reach heavenly realms (v. 127a). It is unclear if the criticism refers to the practice of satī, where a “virtuous” (satī) widow immolates herself on the funeral pyre of her husband, or simply to anyone who, taking its symbolism literally, performs the anyeṣṭi (the final

The beginnings of Mādhyamika doxography  67 burial rites) on himself before time, in the hope of being carried away to heaven by Agni. Scriptural justifications for the practice of satī appear late. It is ambiguously mentioned in the Mahābhārata (MhB) if at all, and only formally appears in the last smṛti texts, such as the sectarian Viṣṇu Smṛti (700–1000 CE).152 On the other hand, the practice of agnipraveṣa (entering fire), probably related to ancient trials, where an accused walks into fire to prove his/her innocence or loyalty, is mentioned in the Rāmāyaṇa.153 Therein, it is not performed in the hope of attaining heavenly realms. Whatever selfimmolation practice Bhāviveka has in mind, he derides it as wicked (durvihita) and soteriologically ridiculous. It cannot lead to heaven but only to self-destruction. It fares no better than the suicidal flight of a moth (chalabha-ādi-prapātavat). Ritual fasting to death A similar reasoning is then applied unto the practice of fasting until death, an extreme asceticism particularly promoted in Jainism,154 under the name of sallekhanā, but not limited to this tradition. The Buddha himself, before enlightenment, is said to have severely damaged his health through fasting, as dramatically depicted in Buddhist canonical literature and Gandhāran art.155 To those who argue that this form of ritual suicide156 conveys merit, because it cultivates renunciation, Bhāviveka retorts that renunciation alone is no source of merit. For, one can also renounce truth (v. 130: satya-tyāga), and it is certainly not meritorious. Hence, on strictly logical grounds, being too broad, the reason is fallacious (vyabhicāritā). Bhāviveka’s reasoning on renunciation recalls a famous verse from the BhG: A man does not attain freedom from the results of action by abstaining from actions, and he does not approach perfection simply by renunciation.157 The verse serves to introduce the discussion on the “yoga of action” (karma-yoga) in the BhG. It is as if Bhāviveka was subtly reminding his Brahmin interlocutor of his own scriptures, though investing it with a purely logical outlook. The logic here, basically, is that eating, in itself, is no offence. Thus, to renounce it is not meritorious, just as abandoning life by jumping into a fire is not particularly virtuous, for being alive is no offence. Ritual diet (vegetarianism) The discussion on ‘not-eating’ moves on to include the practice of vegetarianism. Bhāviveka argues, “When pure from the three perspectives, the flesh which is eaten incurs no offence” (v. 132a). The notion of ‘threefold purity’ (trikoṭiśuddha) is a concept which appears both in the Buddhist sūtra-s and in the Vinaya literature. James Stewart explains, Meat eating is acceptable but only if that meat satisfies three conditions: that it is not seen, heard, or suspected that the animal was killed for the monks who were to receive it.158

68  The beginnings of Mādhyamika doxography Hence, in Bhāviveka’s logic, so long as the meat respects these three conditions, there is no merit in renouncing it or any blame in consuming it. Meat only serves the function of nourishing the body, because of hunger, which implies no immorality (v. 133). If meat is rejected because it is considered impure, so should one’s body be renounced (v. 134), ironically retorts Bhāviveka. Further arguments are given to alleviate doubts about meat eating. Ritual worship of inanimate objects Finally, the last ironic formula (yuktaṃ yat tyajyate trayī) is mentioned, in relation to people who imagine inanimate objects to be imbued with consciousness, like trees, and who venerate them as supernatural beings. Tree-worship is one of the most widespread and ancient forms of folk religion in India.159 The MhB story of Śakuntalā, famously dramatized by Kālidāsa, vividly describes the affection between the plants of the heroine’s ashram and herself, a theme which recurs in Kālidāsa’s Meghadūta as well. The same affection is somewhat recalled by Bhāviveka’s opponent in verse 144. He argues that, like other sentient beings, plants grow and long to be touched (dohada) while budding.160 Interestingly, Bhāviveka concludes his list of refutations by observing that plants do not reproduce nor are they born like other sentient beings. For: Due to the karma of sentient beings trees and so on are produced by time, as in hell and in the heavenly realm the sword and the jewel trees.161 The dialectician thus suggests that while karma regulates the development of sentient beings, plants and so on, because they are not sentient beings,162 simply appear in various forms according to sentient beings’ merit, depending on their realm of existence. In other words, the appearance of green plants, trees with jewel-like leaves, or trees with blade-like leaves163 depends on one’s karma; as if the mineral and vegetal realms were but a mere feature of experience (of the mind?), insentient objects born of time (kālaja), that is to say that they are momentary phenomena dependent on causes and conditions, like everything within saṃsāra. The theory, though consistent with the Madhyamaka notion of nisvabhāva, is unlikely to have gathered consensus within the whole of Buddhism, and even less to convince moderns.164 The case serves to illustrate that, while Bhāviveka’s logic is well armed to deal blows to the assertions of his opponents, the moment the Mādhyamikan begins to make assertions of his own (svatantra-anumāna), based on strictly Buddhist doctrines, as with the trikoṭiśuddha invoked to defend meat eating, he unavoidably exposes himself to counter-attacks. In any case, the crux of the discussion, in the present context, is meant to indicate that worshipping insentient objects is pointless. As mentioned, the reason for listing these practices – bathing, jumping into fire, fasting to death, vegetarianism, and worshipping inanimate objects – is, as with Vedic Yoga, to insist on the irrationality and impotency of such activities concerning liberation. Hence, these are not independent discussions but the logical

The beginnings of Mādhyamika doxography  69 continuity of the refutation of the karma-mārga, professed by the Mīṃāṃsā, extending it to every kind of Vedic-inspired ritualism. It is also conceivable that Bhāviveka had in mind some of his Buddhist brethren when composing his refutations. The lure of purity and sanctity, exhibited in extreme asceticism, vegetarianism, and the worship of inanimate objects, continuously tempted Buddhism, if not the Buddha himself, from its beginnings until today. Even the words of so-called gods and seers must be analyzed based on their arguments This last discussion concludes the long reply to verse 11. It contains the last preserved Sanskrit verse of the MHK. The remaining text is only available in Tibetan. Coming back to the initial claim of verse 11, that the Vedic path is “desired by the learned and practiced by ṛṣi-s and gods,” Bhāviveka replies that, even so, they must be assessed based on their logical coherence. Pūrvapakṣa v

Anything worthy of interest belongs to these men (verse 12)

Verses summary: (12) Everything comes from the Veda, even the four goals of life Analysis of the Uttarapakṣa (152–154) Based on the translation from Tibetan produced by Lindtner, one grasps that the rejoinder to verse 12 answers the pretentiousness of the Brahmin with a corresponding insult, before turning the assertion around, like a mirror. Like words carved by worms, the sayings of the Veda can appear surprising, Bhāviveka says. But even if some respectable doctrinal element might fortuitously be found in the Veda, it is like finding a jewel in a heap of shit (v. 153–154). It does not belong there. The image of a jewel in the mud, or of a lotus growing out of the mud, is a common Buddhist trope for the awakened mind, or “Buddha nature” (tathāgatagarbha/buddha-dhātu), said to be covered by defilements which need be removed. It is a central theme of the Saddharmapuṇḍarīkasūtra (Lotus Sūtra). In the present case, the mud refers to the Veda, while anything worthy in it, its possible jewels, able to reveal the awakened mind, already belongs to the Buddha’s teachings. In brief, Bhāviveka turns the assertion of the Brahmin against him. Yet, even such a Vedic ‘fool’ has the potential to become a Buddha, as explained in the Lotus Sūtra. Bhāviveka’s task, as a Bodhisattva, is to skillfully extract him from the stinky situation he finds himself in, the inhuman stench of the Veda. Pūrvapakṣa vi Reasoning is a dangerous abuse of knowledge spoiling the Veda (verses 13–14).

70  The beginnings of Mādhyamika doxography Verses summary: (13) The inferential reasoning of debaters spoils the Vedic path (14) Inference dangerously leads the blind in unwarranted directions Analysis of the Uttarapakṣa (v. 155–158) This second to last discussion, however short, is of great significance. Therein, Bhāviveka reasserts the relevance of reasoning and, thus, of his whole endeavour within the MHK. It closes the circle. If one is to seek jewels in the mud, one must first learn how to recognize them. Bhāviveka explains: If one does not investigate what is logical and what is not logical by means of anumāna free from faults, then one’s understanding will be formed by other [traditions], and therefore one will be in doubt about other traditions (āgama). (v. 155)165 In other words, as in the Biblical ‘bad seed parable’ (Matthew 13:1–58), anumāna, by cultivating discrimination (viveka), teaches the sower how to recognize good seeds from bad ones, good soil from sterile grounds. He who has ears, let him hear. It removes doubts brought about by competing claims of alien scriptures. Anumāna serves to acquire certainty in knowledge. Even if the knowledge gained therefrom is not ultimate yet, since anumāna is no ultimate means, it is a skillful tool nonetheless: It is like a man wanting to cross a large river who gets hold of a boat. Likewise, one must first hold on to anumāna, even if one has to abandon it [later on]. (v. 156) Anumāna is like the rescue vessel of the ‘raft parable’ in the Majjhimanikāya, standing for the whole bauddhadharma. Though extremely useful when drowning, it must not be clung to once the river has been crossed. Without relying on reason, however, one is easily carried away by all sorts of claims and traditions: Following mere words (śabdamātra), and going along with the past like blind men, here in this circuit, fools revolve in saṃsāra.166 The ocean of doctrinal literature is endless and treacherous. It is like saṃsāra. Like the mythical ocean in heaven, this scriptural ‘chaos’ must be churned by the pillar of reasoning to produce the nectar of immortality (amṛta): right view. With the MHK, what Bhāviveka offers to the Brāhmaṇa is a safety vest, an eye vaccine against false views. But the patient must first learn to trust in his doctor’s judgment. Pūrvapakṣa vii Anything which objects to the Veda is spurious and devoid of authority (verses 15–17)

The beginnings of Mādhyamika doxography  71 Verses summary: (15) No one is omniscient, the Buddha’s omniscience is a deception (16) The Buddha is human hence his discourse is no pramāṇa and he is not omniscient (17) Any discourse objecting the Veda is no pramāṇa, be it from the Buddha or the Jainas Analysis of the Uttarapakṣa (v. 159–167) The last discussion of Chapter 9 revolves around the authority of the Buddha. Bhāviveka argues that Buddha’s omniscience is only mentioned figuratively (v.  159). However, the Buddha does know all there is to be known about the way (v. 164). Moreover, referring to the Mahāyāna doctrine of the three bodies (trikāya), even if the Buddha had a human appearance (nirmāṇakāya), his dharma and sambhoga bodies do not exactly belong to a human being. Hence, Buddha’s teachings cannot be said to be like that of any other human speech (as the opponent argued in verse 16). If the Brahmin then argues that his gods are similarly omniscient, Bhāviveka asks him how one is to explain their shortcomings (v. 162). Verse 166 presents an interesting reasoning: Moreover, the standpoint that [our] sacred texts, ideas and words [are not valid, because they are created, cf. 16] should be answered in the same way: [The mind] that analyses is uncertain, since it is created; therefore [the argument is] fallacious. (v. 166) Even if it was true that the Brahmin’s scriptures are non-human (apauruṣeya) and uncreated (akartṛ), in order to understand them, the Brahmin necessarily needs to rely on his intelligence. Any understanding of scriptures entertained by the Brahmin is a product, in the sense that it is informed by the conventions which invariably structure its intellection. Scriptural understanding belongs to a human (puruṣa) agent (kartṛ). The logical result is the same: Scriptural knowledge invariably conveys conventional knowledge to a conventional mind and agent. Therefore, conventional reasoning, like anumāna, is required to analyze scriptural conventions. Only then can scriptures convey the soteriological knowledge which they might conceal, as in a heap of shit, to reveal its jewel-like lustre.167 To conclude, as if to introduce the last chapter on omniscience, Bhāviveka observes that even if he criticizes the Veda like the Jainas do, one should not confuse him as one of them.

1.7  Irony in Chapter 9: peace by polemics After having examined the main discussions of Chapter 9, is it possible to answer the question raised while introducing its dialectical strategy: Was the dialectical exercise performed in Chapter 9 soteriologically relevant in any way or was it an

72  The beginnings of Mādhyamika doxography artificial debate carried on only for the sake of quieting an imagined Vedic opponent? In other words, what is the significance of MHK Chapter 9 for the religious practice of a Buddhist recluse studying it? At the outset, there should be no doubt remaining about the polemical intent of Bhāviveka. As discussed, the dialogue between the author and his Brahmin puppet is a staged one. Bhāviveka’s rhetoric is far from the ecumenical interreligious dialogue promoted by modern proponents of religious pluralism.168 However, though political correctness is not part of the author’s method, he nonetheless entertains a philosophical dialogue – with a pinch of humour and drama, of course – where reason, not politics, is said to set the rules of the game: establishing a coherent hermeneutical and dialectical apparatus for dialogue, in the service of pedagogy and soteriology. One wonders, had he known about them, if and how Bhāviveka would have scorned Abrahamic prophecies. Would he have mocked the Holy Trinity as irrational in the same way as the Vedic Trimūrti? Would he have ridiculed the childlike idea that god washes sins away with snow, water, and hail, as requested in prayer between the opening takbir and the recitation of the Qur’an?169 Would he have introduced as ‘pretentious’ the claim of Israel to be God’s chosen people? Surely, Bhāviveka’s mockery is as controversial today as it was in his time. In the end, the purpose of this kind of dialogue is not to make friends by finding common ground in the views of others, but only to acquire certainty (and power!) by critically assessing and defeating competing doctrinal claims. Consequently, those claiming an inclusivist tendency170 in Bhāviveka’s MHK are misreading his rhetoric. One attached to Vedic lore and rituals is unlikely to find any scope for friendly bonding in the MHK. Indeed, Ham fittingly noticed how the depiction of Brahmins found in the MHK matches Dharmakīrti’s “five signs of complete stupidity” (pañca liṅgāni jāḍye) mentioned in the Pramāṇa-vārttika (PV), themselves a summary of the Buddhist charge against Brahmins: 1. [Believing in the] authority of the Veda, 2. Claiming something [permanent] to be agent, 3. Seeking merit in ablutions, 4. Taking pride in one’s caste, and 5. Undertaking penance to remove sin, these are the five signs of complete stupidity devoid of any discrimination.171 Showcasing the ‘stupidity’ of Brahmins is clearly a negative apologetic strategy on the part of Bhāviveka, deconstructing the Vedic worldview to reveal its ridiculous consequences (reductio ad absurdum). As discussed in the introduction, the apologetic dimension of the MHK reflects the political tensions and ideological struggles of its days. Within this sociohistorical context, interreligious debates played a significant regulating function. Their rules and dynamics dictated a dialectical methodology and rhetorical attitude which influenced the MHK. In the event of a public debate, where judges were no passive readers and likely not impartial, the rhetor had to seduce his auditors by all means. Logical demonstrations could surely bring about the defeat of an opponent, but the sympathy of the public also had to be won. This

The beginnings of Mādhyamika doxography  73 naturally called for other rhetorical means, perhaps more concerned with human nature than with mere logic. Hence, within the MHK, Bhāviveka also showed his student how to flatter the ethos of a broad public by skillfully drawing from a wide array of materials, especially non-Buddhist materials, to convey images entertaining several layers of meaning which could be appreciated by a wider audience. Surely, a good way to win the sympathy of others also involves humour. Boredom is the Damocles sword of the orator. A dreadful reminder that “a yawn is a silent shout.”172 A reader should thus keep this context in mind when approaching a text like the MHK, and let himself be amused by arguments which are not only designed to convince his rationality, but also to win over his intellect by triggering amusing emotional responses. “Responding to humour is part of human behaviour, ability, or competence,”173 seriously noted Victor Raskin in his well-documented study on the semantics of humour. By being too serious, one risks missing the irony of a potentially insightful joke. Being comical does not necessarily imply being trivial. On the contrary, one can argue that potent humour displays an acute intellectual aptitude: the ability to take certain liberties with reality in order to highlight problematic elements otherwise kept hidden by the monotony of routine and conventions. In this sense, humour has a corrective function. What is being laughed at is a behaviour or an idea worth reviewing. Accordingly, irony can be said to be a particular form of philosophical discourse. One doubting the place of irony in philosophy would be well advised to consider its function in Socratic dialogues. Søren Kierkegaard’s doctoral thesis On the Concept of Irony With Continual Reference to Socrates (Danish: Om Begrebet Ironi med stadigt Hensyn til Socrates) extensively discussed the topic. Therein, the famous Danish philosopher observed how, among other properties, irony carries a certain air of superiority: The ironic figure of speech has still another property that characterizes all irony, a certain superiority deriving from its not wanting to be understood immediately, even though it wants to be understood, with the result that this figure looks down, as it were, on plain and simple talk that everyone can promptly understand; it travels around, so to speak, in an exclusive incognito and looks down pitying from this high position on ordinary, prosaic talk. . . . Just as kings and princes speak French, the higher circles (this, of course, must be understood according to an intellectual ordering of rank) speak ironically so that lay people will not be able to understand them, and to that extent irony is in the process of isolating itself; it does not wish to be generally understood.174 Naturally, by ironizing over his opponent, Bhāviveka dialectically places himself, and his brethren, on high grounds. Ironically, though isolating, irony is also bonding. It creates a community, perhaps slightly insular. Obviously, Bhāviveka does not write for the ordinary ritual-oriented Brahmin, making a living by selling prasāda, whose intellectual insight would unlikely fathom the author’s discourse,

74  The beginnings of Mādhyamika doxography but for the initiated Mādhyamikans, or at least for the learned neophytes, if not for the educated dialecticians of other creeds who might share his anti-Vedic and antiritualist stance. For, nothing could be further remote from the gnostic pursuits of Bhāviveka than the ritual technician boasting over his knowledge of set formulas and performances. Indeed, whereas, having the sole ownership of the Veda, the Brāhmaṇa claims that he knows much and is eager to capitalize on his sapiential asset, the Veda, Bhāviveka, on the other shore, examining this dubious knowledge, benevolently admits being aware that he knows nothing (śūnya). When he mimics the position (pakṣa) of the Mīmāṃsā, by dialectically presenting a caricature of its theses and assessing them as though they were worth discussing, Bhāviveka fully embraces the most theatrical aspect of irony. Although the whole discussion is staged from the beginning, so as to lead to the logical conclusion that the Mīmāṃsā’s views are ultimately inconsistent, it convincingly appears as a genuine ‘historical’ debate. The irony even fooled modern commentators. This is where the dialectical trickster shows his mastery. Accordingly, Kierkegaard noticed two modes of ironic expression: Either the ironist identifies himself with the odious practice he wants to attack, or he takes a hostile stance to it, but always, of course, in such a way that he himself is aware that his appearance is in contrast to what he himself embraces and that he thoroughly enjoys this discrepancy.175 Thus, Bhāviveka’s Brahmin fool is made even funnier by exhibiting somewhat recognizable ‘mimics’ – his theses – which recall those of an identifiable community, while leaving some ambiguity as to its exact identity. Raskin pointed out that deliberate ambiguity underlies much, if not all, of verbal humour. Yet, Bhāviveka’s humour is scathing. It is not solely intended to bring laughter, but to judge and thus to incite a change. In theorizing the sociological aspect of humour, Henri Bergson argued, [L]aughter is, above all, a corrective. Being intended to humiliate, it must take a painful impression on the person against whom it is directed. By laughter, society avenges itself for liberties taken with it. It would fail in its object if it bore the stamp of sympathy or kindness.176 Surely, Bhāviveka is not too kind with his Brahmin puppet. At least not in the sense of being flattering. Yet, from a religious or ethical perspective, the author can be said to be compassionate for pointing out flaws in conducts and views which, if left unexamined, are seriously detrimental to one’s development – at least from the standpoint of the author. So long as the potential danger has been assessed with reason, and ideally with impartiality, it is the moral duty of a Bodhisattva, if not of any good friend, to inform the party concerned, just as a doctor would inform his patient of the dangers of potentially unhealthy habits, letting him with the freedom to decide on his own what to do with the assessment afterwards. For, though the outcome of a public debate could have serious consequences, the judgment of a

The beginnings of Mādhyamika doxography  75 philosopher, in itself, is not like one of the Inquisition. It is meant to ‘in-form,’ to ‘trans-form,’ a problematic view, and not to violently eradicate its human host. According to Kierkegaard, irony is perfectly suited for the task: In irony, however, since everything is shown to be vanity, the subject becomes free. The more vain everything becomes, all the lighter, emptier, and volatilized the subject becomes. And while everything is in the process of becoming vanity, the ironic subject does not become vain in his own eyes but rescues his own vanity.177 To empty out the Mīmāṃsā’s view by displaying its vanity is Bhāviveka’s strategy in Chapter 9. One should bear in mind that Brahmins were not always outsiders to Buddhism, but that many members of the saṃgha notoriously came from such ‘distinguished’ backgrounds. As any individual joining a new community, they would naturally carry their previous habits with them. Thus, the outsiders portrayed in the MHK might not be so foreign after all. This becomes especially meaningful after having noticed how some practices criticized in Chapter 9 are fairly close to Buddhist ones. Hence, for some members of the Buddhist community, Bhāviveka’s Brahmin puppet might have served as a skillful mirror, reflecting precise ‘defects’ to be debated, without the social stigma ensuing from a too obvious identification. For, among all Indian philosophical communities, the MHK would have been mainly, if not only, read among Buddhists.178 It was composed for them, keeping their sole concern in mind. Accordingly, the dense dialectic of the MHK, permeated by the hetu-vidyā, did not only serve apologetic purposes. But, by encouraging selfexamination, it establishes the foundations for the reflective and meditative praxis of the Buddhist way, a path at the antipode of the Brahmanical karma-mārga. The MHK, as a doxography, is a literary technology befitting the training of Bodhisattvas. Learning about the various doṣa-s affecting the mind, and about their appropriate ‘therapeutic,’ prepares them to fulfill both their healing and predicating duties – having first critically examined their own mind. Eltschinger observed a similar therapeutic dimension in the doxographical work of Śāntarakṣita: As Kamalaśīla insists, the doctrines Śāntarakṣita is concerned to refute all involve “mistaken views of the self” (vitathātmadṛṣṭi). As such, the importance of critical reflection upon them lies precisely in the fact that they are not just others’ views of themselves, but that potentially at least, they are views that any of us may harbor, either explicitly or not, with respect to ourselves. Śantarakṣita’s critical journey through the byways of Indian philosophy is therefore no mere exercise in doxography; rather, it is a therapy whereby one must challenge one’s own self-understanding so as to disclose and finally uproot the misunderstandings that are concealed therein.179 Learning about competing views, especially in the monastic context for which the MHK was composed, is no mere intellectual pursuit. Rather, to train oneself in

76  The beginnings of Mādhyamika doxography identifying the deficiencies in an opponent’s reasoning, through dialogue, is akin to a kind of mental ‘vaccination,’180 making oneself immune to infectious cognitive impurities (doṣa-s) by inoculating in one’s mind stream a diminished (parodic?181) version of them. Within the doxographical content of the MHK, polemics and therapy are intertwined in a dramatic dialogue, at times humorous, or at least ironic, engaging the student in a methodological propaedeutic designed to cast away misleading views and to firmly establish oneself in the non-view of voidness (śūnyatā), the ironic vanity of all view (darśana).

Notes 1 Burnouf 1844: p. 499. 2 Although not exhaustive, the following list covers the most important contributions to date in relation to Bhāviveka and the MHK: Saitō 2005; Bahulkar 1994; Eckel 1992, 2008; Ejima 1980b, 1980a; Eltschinger 1998; Gokhale 1958, 1972, 1985; Ham 2016; He 2014; Heitmann 1995, 2004; Hirabayashi and Iida 1978; Honda 1967; Hoornaert 1999, 2001, 2003, 2004; Hsu 2011, 2013; Iida 1966, 1980; Jones 2011; Kawasaki 1973, 1976; Krasser 2011, 2012; La Vallée Poussin 1932–1933, 1933; Lindtner 1982, 1986, 1995, 2001b, 2001a; Miyasaka 1954b, 1954a, 1958; Nakada 1972; Nakamura 1950, 1955, 1958, 1968; Potter 2003; Qvarnström 1989, 1999, 2012, 2015; Ruegg 1990; Warder 1970b; Watanabe 1994; Yamaguchi 1941 & 1964. 3 See Hsu 2013: p. 13–14. 4 See Verardi 2011: p. 74. 5 Shotaro Iida presented a table linking the various names with their textual reference. See Iida 1980: p. 5. 6 Lindtner indicates that the Mahāvyutpatti has ‘Bhavya’ (Tib. sKal ldan, Mong. Tegüs qubitu – clear distinction), see Lindtner 2001b: p. ix–x. 7 On Bhāviveka’s name, see Ejima 1980a and Hsu 2013: p. 10–12. 8 See Iida 1980: p. 12–19 and Watanabe 1994: p. 9–15. 9 The Tibetan translation of the Prajñāpradīpamūlamadhyamakavṛtti, with the title Śes rab sgron ma, was made at the beginning of the ninth century by Klu’i rgyal mtshan and Jñānagarbha, whereas the Chinese translation, Bānruò dēng lùn shì (般若燈論釋), was made by the Indian monk Prabhākaramitra (565−633 CE) between 630 and 632. 10 For more information on the two other texts of Bhāviveka, see Hsu 2013: 40–43. Hsu’s doctoral thesis is dedicated to the Karatalaratna. 11 On how the MHK was discovered and hand-copied by Rāhula Sāṃkṛtyāyana at Shalu (zhwa lu) monastery in Tibet in 1936 and passed on to V.V. Gokhale, who later on found photographs of it in G. Tucci’s collection, see Watanabe 1994: p. 14–15. 12 Ames 1985: p. 36. 13 Hsu (2013: p. 18) indicates that this is the case of some modern Japanese scholars, perhaps referring to Yamaguchi, but she mysteriously refers to Hirakawa’s History Hirakawa 1990 (?) where no such mention is made. On the other hand, Warder, along the line of Iida, mentions that the impulse of Bhāviveka was continued beyond the eighth century and up to Tibet, by Avalokitavrata, Śāntarakṣita, Kamalaśīla, and Haribhadra. See Warder 1970b: p. 451–456. To that list, Qvarnström adds Jñānagarbha, in Qvarnström 1988: p. 3. 14 Brackets are mine. See Hsu 2013: p. 19. 15 See Warder 1970b: p. 368–369. 16 See Qvarnström 1988: p. 6. 17 See Bouthillette 2017a.

The beginnings of Mādhyamika doxography  77 18 Quoting Kendall W. Folker, Eckel explains that the mysterious number 363 is meant to convey the “bewildering complexity” of the variety of philosophical positions. On the number 363 and its references in the MHK, see Eckel 2008: p. 28–35. 19 As quoted in a bottom-page note by Eltschinger 2014: p. 13. 20 For Eltschinger’s provisional definition of parīkṣā, see Eltschinger 2014: p. 18–19. Moreover, one should note that the term parīkṣā already appeared in the third- or fourth-century Buddhist Saddharmasmṛtyupasthānasūtra, in what is arguably both a contemplative and a scholastic context. The term can be found in the bahuvrīhi compound dharmādharmaparīkṣātattvajñaḥ. See Stuart 2015: [vol. I]: p. 482 [487.488]. 21 Eltschinger 2014: p. 71. 22 Bronkhorst 2016a: p. 182. 23 Clayton 2006: p. 71–74. 24 Eltschinger 2014: p. 72. 25 Eltschinger 2014: p. 72. 26 Verardi 2011: p. 205. 27 The term, which came to designate those who do not recognize the authority of the Veda, will be discussed further in section 2.2.1 From the Ṣaṭ-Tarkī to the Ṣaḍdarśanasamuccaya: The Rise of the Mīmāṃsā. 28 In the case of Śaṅkara’s SSS, the present study will also capture an intra-Brahmanical episode of that struggle, focusing on the Advaitin’s reevaluation of the Mīmāṃsā’s doctrinal boundaries. 29 Ghosh 2002: p. 96–97. 30 See Iida 1980: p. 52. 31 Collins 2000: p. 177. 32 Ronald Davidson (2002) was the first to fully theorize such ideas in relation to Buddhist esotericism. 33 Eltschinger 2014: p. 174. 34 See Verardi 2011: p. 106. Also, Jonardon Ganeri (2001b) made a similar argument in the context of the scholastic tradition. 35 Eckel 2008: p. 53. 36 I am borrowing the expression ‘literary technology’ from the social studies of scientific knowledge. It was first coined by Steven Shapin and Simon Schaffer, two historians of science, in their work Leviathan and the Air Pump (Shapin and Schaffer 1985). In a separate publication, Schaffer explains, “The term ‘literary technology’ indicates how scientists use texts as knowledge-producing tools and connects literary with other scientific work. . . . These technologies help fashion the bearer of knowledge as authoritative and competent, and the item of knowledge as independent from the contingencies of human judgment. They make authors and they make facts. In this sense, the array of technologies helps order nature and society” (Schaffer 1998. In these pages I argue that doxographies use the epistemological framework of pramāṇa-s, in tune with the practice of the hetu-vidyā (logical science) to teleologically organize, according to a given propaedeutic, and at times even to examine and secure, claims about reality, so as to both convince and inform the views of their intended audience. In this sense, doxography is a contemplative philosophical literary technology. 37 In 1929, Pandit Sāṅkṛtyāna courageously began his journeys from Patna (India) to Tibet in search of manuscripts. He undertook further expeditions in 1934, 1936, and 1938. Besides the MHK, found in 1934, he also brought back other important Sanskrit texts from the same Shalu’s archives, notably a series of Mahāsāṃghika Vinaya texts on palm leaves (see Prebish 1996: p. 34). As mentioned by Steinkellner 1980, his several journeys led to the rediscovery of such important works as Dharmakīrti’s Pramāṇavārttika (ed. 1938) and Vādanyāya(ed. 1935–36), Karṇagomin’s subcommentary on the Pramāṇavārttikavṛtti(ed. 1943), Prajñākaragupta’s Pramāṇavārttikabhāṣya (ed. 1953), Śāntarakṣita’s Vādanyāyaṭīkā (ed. 1935–36), and Manorathanandin’s

78  The beginnings of Mādhyamika doxography Pramāṇavārttikavṛtti(ed. 1938–40). The photographic prints made from Shalu’s archives were entrusted to the Bihar Research Society in Patna for editing. Today, the site of Shalu Monastery can be more easily accessed on Wikipedia (https:// en.wikipedia.org/wiki/Shalu_Monastery), where some photos are generously provided (the same content is more or less copied, spelling mistakes included, in the online Chinese Buddhist Encyclopedia). Therein, among other general information, one can read that the small monastery, located about 22 kilometres south of Shigatse, in Tibet, was founded in 1040 by Chetsun Sherab Jungnay. Important for the Sakya tradition, the monastic institution used to be a renowned centre of scholarly learning and was famous for its mural paintings. 38 For more details on the MHK manuscript, see Iida 1980: p. 52–53; and, more recently, Watanabe 1994: p. 14–15. Concerning the Rañjanā script, Jens-Uwe Hartmann informs us, “The script called Rañja or Rañjanā (‘pleasant, delightful’) originated in Northern India relatively late and therefore presupposes a long prior development of writing. . . . The script originated in Bengal towards the end of the first millennium A.D. It was used during the Pala dynasty, under whose reign Buddhism in North-East India flourished for the last time, and spread from there to Magadha and Nepal. Since Rañjanā was one of the scripts used for Buddhist manuscripts, it accompanied both Mahayana and Tantric Buddhism not only to Tibet but also to Central Asia. Under the influence of Tibetan Buddhism, it was used for Buddhist inscriptions even in China and Korea as early as the Mongolian Yuan dynasty (1280–1367).” For more information on the script, see Hartmann 1998: p. 37. 39 As in verse 77 of MHK Chapter 9, where Bhāviveka makes a prostitute (bandhakī) of the consort of the gods who are said by the opponent to have multiple manifestations. See also note 128. 40 Concerning my translations of the verses of the MHK, as a general rule, I follow Lindtner’s (Lindtner 2001a, 2001b) editions for the Sanskrit text. My translation relies on the Sanskrit alone, unless the verse proves too problematic. In this case, with the help of Philipp Maas, the Tibetan commentary has been consulted. Some verses remain problematic, nonetheless. Comments on the translation are given whenever necessary. As noted already, the purpose is not to present a critical edition of the text, but to reflect on its doxographical nature based on what we know thus far. 41 These verses are also translated by Gokhale in Potter 2003: p. 400. Surely, the black antelope skin symbolizes the yajña, the sacrifice itself, but it also became part of the iconography of Śiva (see Kramrisch 1988: p. 336–340). The passage is replete with subtle allusions to Brahmanical devotional practices, like carrying a water jug, sitting on a reed rug, feeding on fallen fruits, muttering the Sāvitrī, worshipping the sun, and making offering to fire. As will be shown, Bhāviveka does not clearly distinguish the Brahmanical followers of the Veda and the worshippers of the Trimūrti awaiting divine grace. Hence, Bhāviveka seems to suggest that, in order to be truly blessed by grace, god-believers, or Brahmins, if not the gods themselves, should embrace the Bauddhadharma and practice their devotion accordingly. 42 MHK 2.8–12 kṛpākṛṣṇājinadharaḥ śraddhāmalakamaṇḍaluḥ | smṛtiguptendriyadvāro dhṛtivetrāsanāsanaḥ || mahāyānamahāramyatapovanasamāśrayaḥ | dhyānaprītiphalāhāraḥ smṛtyupasthānagocaraḥ || gambhīrodārasūtrāntasvādhyāyahatakilbiṣaḥ | pratītyotpādasāvitrīṃ japan satyadvayāśrayāt || varṇāḍhyaiḥ pratipatpuṣpaiḥ sarvadiggandhavāhibhiḥ | upāsīnas tv aharahaḥ samyaksaṃbuddhabhāskaram || hutākuśalasaṃkalpaḥ pratisaṃkhyānapāvake | munivrataṃ cared evam anuttarapadāptaye ||

The beginnings of Mādhyamika doxography  79 43 The Sāvitrī is a prayer addressed to savitṛ, a solar deity, found in the Ṛg-veda. It is made famous by the celebrated verses RV. iii, 62, 10, also called gāyatrī, particularly used during the initiations of the twice-born. The reference to famous figures of Brahmanical devotion would immediately be recognized by Bhāviveka’s audience. 44 Similarly, Eltschinger observed, “according to Sthiramati, the yukti/hetuvidyāenterprise was expected to shape arguments in order (1) to found and support key Buddhist doctrines, (2) to defeat the non-Buddhist intellectuals’ hostility towards Buddhism; (3) to convert them to the Buddhist sad-dharma or śāsana, (4) to strengthen the coreligionists’ adherence to Buddhism” (Eltschinger 2010: p. 562). 45 MHK 3.10–11, translated in Eckel 2008: p. 42. aśeṣakalpanājālapratiṣedhavidhāyinī | śāntapratyātmasaṃvedyanirvikalpanirakṣare || vigataikatvanānātve tattve gagananirmale | apracārapracārā ca prajñā syāt pāramārthikī || 46 This idea is expressed in Nāgārjuna’s MMK, XXIV. 9–10: Those who do not understand the distinction between these two realities do not realize the profound truth embodied in the Buddha’s doctrine. Without relying on conventional [reality], an ultimate [reality] cannot be taught. Without understanding ultimate reality, nirvāṇa is not realized.

This translation is from La Vallée Poussin 1903–1913: p. 494. 47 See Hsu 2013: p. ii–iii, 61, 143–145. 48 The expression ‘Weltanschauung’ was first coined by the Prussian philosopher Immanuel Kant in his Critique of Judgment, published in 1790, wherein he introduced the notion of Weltanschauungin a paragraph stressing the power of perception of the human mind: “If the human mind is nonetheless to be able even to think the given infinite without contradiction, it must have within itself a power that is supersensible, whose idea of the noumenon cannot be intuited but can yet be regarded as the substrate underlying what is mere appearance, namely, our intuition of the world [Weltanschauung]. For only by means of this power and its idea do we, in a pure estimation of magnitude, comprehend the infinite in the world of sense entirely under a concept, even though in a mathematical estimation of magnitude by means of numerical concepts we can never think it in its entirety” (Kant 1987: p. 111–112). If, within that passage, of minimal importance within the Kantian project, the concept of Weltanschauung primarily translates as a simple sense perception of the world, a mundus sensibilis, a “world-intuition in the sense of contemplation of the world given to the senses” (see Heidegger 1982: p. 4), it was soon adopted by Fichte, Schelling, Goethe, Alexander von Humboldt, Novalis, and Hegel, among others, to become a key word in German idealism and romanticism. It was further developed by thinkers such as Kierkegaard, Engels, and Dilthey who reflected on Western culture in more general terms. Recent research in the field has shown that already by 1840 the concept had become a standard German word “denoting a global outlook on life and the world – akin to philosophy but without its rational pretensions” (Wolters 1983). It was not long before the idea caught the attention of the English-speaking world.   In a book dedicated to the history of the concept, David K. Naugle mentions, “According to the textual apparatus, Weltanschauung first appeared in an English context in 1868 in a letter written by William James and quoted by R.B. Perry in his book The Thought and Character of William James (1935): ‘I remember you saying . . . that the characteristic of the Greek ʻWeltanschauungʼ was its optimism.’ Other documentation of the English usage of Weltanschauungare carried up through 1978. Of particular interest is the 1934 citation in M. Bodkin’s Archetypal Patterns in Poetry; she wrote that ‘a man’s philosophy . . . is his Weltanschauung – the individual vision, or perspective of reality’” (Naugle 2002: p. 64). Adopting a similar meaning to the one entertained by Bodkin, “a particular philosophy or view of life; a concept of the world held by an individual or a

80  The beginnings of Mādhyamika doxography group,” the Oxford English Dictionary (2nd ed. (1989), s.v. “Weltanschauung”) suggests rendering the German term in English as “world-view.”   In Buddhist context, the ultimate truth cannot be a ‘view,’ or a ‘worldview,’ because it is the opposite of conceptuality. On the other hand, the conventional realm is nothing but conceptual. It is made of ‘views.’ Hence, the Sanskrit term darśana aptly denotes a philosophical system making sense of the world in conventional terms. To translate darśana by ‘worldview’ is thus suitable. The Madhyamaka system of Buddhism, however, by refusing to present a counter-thesis to the views it refutes, makes the philosophical gambit of being no view itself. This is the traditional taboo which Bhāviveka is breaking by presenting, on the tip of the tongue, through negations, a definition of truth which he can defend in debates using syllogisms. 49 B. J. Dreyfus suggests that ‘acumen’ better translates the term prajñā in this context. He also noted that this progressive model of insight continues to inform the way in which Tibetans explain the soteriological value of their scholastic training. It represents a workable approach to contemplative practice, moving from grosser to ever subtler conceptual mental activities, acquiring certainty through logical analysis and inferential reasoning, until concepts are completely transcended. See Dreyfus 2003: p. 164–165. However, to remain consistent, and because ‘wisdom’ is a general term which can easily be inflected to suit the various doctrinal contexts of the present study, I retain the term for translation. 50 The first appearances of the three wisdoms are noticed as early as the Saṅgītisutta (Dīghanikāya III 219) and the Vibhaṅga (324). 51 On cintā-mayī-prajñā in Dharmakīrti, see Eltschinger 2010. 52 Eltschinger 2014: p. 72. 53 MHK 9.48–49 śravaṇo yadi śabdas te dhvanivyaṅgyaḥ kathaṃ mataḥ | pratipattis tu saṃketād asau śabdaḥ prasajyate || saṃketāsaṃbhavād ādau pratipattir na yujyate | saṃsāravad anāditvāt saṃketasyānuvādataḥ ||

The exact rendering of these two verses is tortuous and difficult, even when consulting the Tibetan commentary (TJ). Philipp Maas has kindly pointed out to me that the commentary seems to suggest a different reading than ‘anuvādata,’ perhaps ‘anuvartatā’ (following). Also, he observed that the Tibetan reads ‘śrāvaṇo’ (audible) instead of ‘śravaṇo.’ As a general rule, my translation follows the edition of Lindtner. Moreover, the notion of anuvāda seems legitimate in a context where words are used, repeatedly, to understand the meaning of conventions. 54 I am referring here to Nāgārjuna’s famous verses, in MMK 25.19 na saṃsārasya nirvāṇāt kiṃ cid asti viśeṣaṇam | na nirvāṇasya saṃsārāt kiṃ cid asti viśeṣaṇam || Saṃsāra (i.e., the empirical life-death cycle) is nothing essentially different from nirvāṇa. Nirvāṇa is nothing essentially different from saṃsāra.



Translation is from Inada 1993: p. 158. 55 MHK 9.50a anumāne pramāṇatvaṃ bhinnaṃ ca na tadātmanaḥ | 50a 56 MHK 9.19–20 saṃpradāyānupacchedād āgamasya āgamatvataḥ | sarvasyāgamatāsiddheḥ kiṃ tattvam iti dhāryatām || yat parīkṣākṣamaṃ yuktyā vacanaṃ cet tad āgamaḥ | tad eva tāvan mīmāṃsyaṃ paścāt tenoditaṃ hi yat ||

The beginnings of Mādhyamika doxography  81

Eckel translates verse 9.19 as “If tradition has the status of tradition because it has an unbroken transmission, then everything is tradition, and it is necessary to determine which is true” (Eckel 2008: p. 31). He further notes that the Tibetan commentary (TJ) explains the word ‘everything’ as referring to the ‘363 views’ (Tibetan: lta ba translates either dṛṣṭi or darśana).   Lindtner inserted the Tibetan commentary directly in his translation of the verse, giving it the following heavy format: “If āgama has the status of āgama because it has been handed down without interruptions [cf. 4], then it is established that all [the 363 doctrines] are āgama. [But this is absurd]! One should hold on to what is true!”   Ham offers a different interpretation of verse 9.20, likely influenced by the commentary: “If it is argued that [only] those words that survive the inquiry of rationality are [qualified to be called] a scripture, first of all, such words must be examined. Then [other insignificant] sayings of it [are to be examined] later” (Ham 2016: p. 203). Ham’s discussion extends further, covering verse 21, where he sees a veil reference to the “action-part” (kriyā-kaṇḍa) and “knowledge-part” (jñāna-kaṇḍa) of the Veda, though no such explanation of the verse is found in TJ. 57 As will be discussed further in the following section, Bhāviveka distinguishes the different philosophical systems according to their ‘truth-principles’ (tattva-s), and does not refer to the typology of philosophical ‘view’ (darśana), a practice which was slowly emerging in his days. 58 MHK 9.21 tatra tatpratipakṣatvāj jñānān muktir itīṣyatām | āmayapratipakṣatvād auṣadhād vyādhimuktivat || 59 Bouthillette 2017b: p. 109. 60 On the doctrinal nexus found in the broader realm of Indian philosophy between therapeutics, valid knowledge (obtained through discrimination), and liberation, see Wezler 1984. 61 Gokhale 1985: p. 78; Potter 2003: p. 398. 62 MHK 1.4 mahābodhau kṛtadhiyāṃ parārthodayadīkṣayā | tattvāmṛtāvatārāya śaktitaḥ kiñcid ucyate ||

A German translation, based on the Tibetan, is available in Heitmann 2004: p. 12–13. 63 MHK 1.13–14 kiṃ punas cakravartīndrabrahmaṇām api durlabham | atyantatṛṣṇāvicchedi sādhāraṇam upāyataḥ || vigrahakṣayaparyantaduḥkhādyanabhibhāvitam | niḥśeṣaduḥkhaśamaṃ tattvārthādhiga­ māmṛtam ||



Another translation, in German, is also available in Heitmann 2004: p. 24–27. 64 MHK 3.6 prajñāmṛtaṃ tṛptikaraṃ dīpo’pratihataprabhaḥ | mokṣaprāsādasopānaṃ kleśendhanahutāśanaḥ ||



Another translation, in German, is also available in Heitmann 2004: p. 90–91. 65 Halbfass 1988: p. 276–277. 66 Gerschheimer 2000–2001: p. 183–184. 67 On the development and place of inference in Buddhist logic, see Stcherbatsky 1970: p. 231–274. 68 I have made this argument in Bouthillette 2017b: p. 109. 69 For further discussion on the link between therapy and dialectic, see my article in Bouthillette 2015.

82  The beginnings of Mādhyamika doxography 70 The same aspiration is formulated in the concluding verse of Śāntideva’s (late seventh to mid-eighth century CE) Bodhicaryāvatāra (BCA): BCA 9.168 kadopalambhadṛṣṭibhyo deśayiṣyāmi śūnyatām | saṃvṛtyānupalambhena puṇyasaṃbhāram ādarāt || and when I may reverently declare to the souls who imagine a real world that all is void, and righteousness is gathered by looking beyond the Veiled Truth?

The translation is from Bibliotheca Polyglotta, from the University of Oslo, as last seen on 23.03.2019: https://www2.hf.uio.no/polyglotta/index.php?page=record&vid= 24&mid=119443. The English rendering, referring to ‘souls,’ seems a bit of a stretch. However, the offered French reading, from Louis de La Vallée Poussin (462, 13–17), is more in tune with the verse: Et béni soit le jour où je pourrai, en m’appuyant sur la vérité d’apparence, enseigner la vacuité à ceux qui croient à la réalité des choses, où je pourrai, avec respect, enseigner les vertus de mérite dégagées de toute notion de réalité! 71 The semantic of ‘seeing,’ revolving around the ‘eye,’ in relation to ‘knowing,’ is a constant in Bhāviveka’s work. It is in tune with the practice of designating philosophical systems by the term darśana, a practice only slowly emerging in the author’s days. The introductory verses to Chapter 3 of the MHK, the chapter where Bhāviveka lays out his own Madhyamaka interpretation of Buddhist doctrines through a negation of Abhidharma categories, revolve around this metaphor of ‘sight’: MHK 3.1–3 yasya jñānam ayaṃ cakṣuś cakṣus tasyāsti netarat | yatas tasmād bhaved dhīmāṃs tattvajñānaiṣaṇāparaḥ || paśyaty andho’pi matimān didṛkṣur viprakṛṣṭakān | sūkṣmavyavahitān arthāṃs trailokyāhatadarśanaḥ || sahastreṇāpi netrāṇām anetro buddhivarjitaḥ | svargāpavargasadbhūtamārgāmārg āsamīkṣaṇāt || He whose eyes are for knowledge only, for whom there is nothing else, hence desiring above all the knowledge of reality, shall become learned. (1) Even the blind who wishes to examine remote [objects], when intelligent, can see subtle and concealed objects, as his vision is undisturbed in the three worlds. (2) [But] even with a thousand eyes, one devoid of intelligence is eyeless because he does not investigate what is the real path and what is not a path at all towards heaven and final emancipation. (3)

Another English translation is also provided by Jones 2011: p. 146–147.   In other words, true seeing, the sight of truth, is acquired through knowledge alone. To wisely discriminate among the many possible paths claiming to lead one to peace, one needs to cultivate one’s intelligence. Hence, the task of a Bodhisattva, desirous to lead beings ashore, is to operate on the ‘eyes’ of sentient beings, to cure their blindness, brought about by misconceptions. A clear vision is what an established (siddhānta) darśana should convey. This is what the MHK aims to clarify.   On another note, more anecdotal, it might be no coincidence if, in one of the countless legends about Nāgārjuna, he is said to have been an eye doctor (see White 1996: p. 67). An ancient ophthalmological treatise discussing the treatment of the cataract is also attributed to the Bodhisattva, see Triplett 2017. 72 Singh 2017: p. 9. 73 As mentioned by Westerhoff 2014: p. 123.

The beginnings of Mādhyamika doxography  83 74 Lindtner (2001b: p. xxviii), refers to the following verse, listing the titles of the first three chapters: MHK 1.5 bodhicittāparityāgo munivratasamāśrayaḥ | tattvajñānaiṣaṇā ca iti caryā sarvārthasiddhaye || Not to relinquish the aspiration towards enlightenment, to take refuge in the monastic vows, and to seek to know the truth, this is the conduct for the accomplishment of all aims. 75 The title given in Gokhale 1985: p. 78 has a typographic mistake. The ‘v’ from tattva is missing, and it is thus mistakenly rendered as Tattāmṛtāvatāra. 76 See Bouthillette 2017a: p. 78. 77 Eltschinger 2014: p. 11–12. Brackets are mine. 78 See Watanabe 1994: p. 11. 79 In the next chapter, on Jaina doxography, I will criticize Halbfass’s remark, expressed in Halbfass 1988: p. 351. 80 Westerhoff 2014: p. 123. 81 Thus, I disagree with Nicholson’s (2010: p.  152) remarks to the effect that, within Bhāviveka’s MHK, “there is no apparent order in the sequence of doctrines he refutes.” I rather argue that there is indeed a clear order. But, in order to see it, one must be familiar with the dialectical strategies of Madhyamaka Buddhism. In brief, all views are negated from the beginning, based on the ultimate perspective, but, relatively speaking, the best view stems from the analysis of Abhidharma categories, followed by other Buddhist systems. 82 Westerhoff 2014: p. 131. 83 See Qvarnström 2012: p. 398. 84 For more information on this chapter, including a critical evaluation, see He 2011. 85 On the Vedāntatattvaviniścaya chapter of the MHK, see Qvarnström 2015. 86 Unfortunately, there does not seem to have been much scholarly interest on that part of the text so far. 87 The devatā (deity) of each view, mentioned by Haribhadra, mainly serves to distinguish the locus of authority behind the theoretical principles enunciated. 88 As mentioned in Gokhale 1985: p. 78. 89 See Gerschheimer 2000–2001, 2007. His findings will be discussed more thoroughly in Chapter 2, on Jaina doxography, where the number had more consequences. 90 Eckel 2008: p. 32. 91 As mentioned in Gerschheimer 2000–2001: p. 173. 92 For a fascinating account of the revolutionary journey of the famous French Encyclopédie, presented as “the triumph of reason in an unreasonable age,” see Blom 2004, 2005. 93 Various parts of Chapters 1–3 have been translated and published by different scholars: Gokhale 1972; Heitmann 1995, 2004; Iida 1980; Jones 2011; Potter 2003; Watanabe 1994, 2012, 1998. For the subdivisions of Chapter 3, I rely on Watanabe. 94 Chapters 4–5 are fully translated with the TJ commentary in Eckel 2008. However, since no detailed table of contents of the text is provided by the author, I divided the chapters according to the main divisions found in his translation. 95 This chapter has also been discussed by Hoornaert 1999, 2000, 2001, 2002, 2003. 96 Chapters 6 and 8 are fully translated with the TJ commentary in Qvarnström 2015. The chapter’s divisions are based upon Qvarnström’s translation. Another discussion in Saitō 2011. 97 A translation from Tibetan is available in He 2011. The chapter’s divisions are taken from the same article. Other discussions are found in Miyasaka 1954b, 1954a, 1958. 98 Further discussions are found in Gokhale 1958; Nakamura 1950, 1955, 1968; Qvarnström 2015. 99 Besides my own translation to follow within these pages, another translation was given by Lindtner 2001a. The chapter was also discussed by Ham 2016; Iida 1966; Kawasaki 1976.

84  The beginnings of Mādhyamika doxography 100 I am using Roman numerals to separate the seven sections of the pūrvapakṣa which I regrouped myself based on their thematic. I will use the same Roman numerals when commenting on each section. The verses regrouped by each section are given in parentheses. 101 A summary of Chapters 10–11 is given in Potter 2003. 102 The relation of this fivefold division of a text with the Mīmāṃsā is discussed in Nicholson 2015: p. 158. 103 MHK 9.18 tad atrāpi parīkṣante yathābhūtagaveṣiṇaḥ | pakṣapātaviṣaṃ hitvā śabdārthanyāyakovidāḥ || 104 Translation is mine. NS 1.2.1 pramāṇatarkasādhanopālambhaḥ siddhāntāviruddhaḥ pañcāvayavopapannaḥ pakṣapratipakṣaparigraho vādaḥ | 105 One who would argue against this hypothesis by claiming that doxographers and śāstra authors were strictly aiming at presenting an objective description of their opponents based on the state of their knowledge, in a way somewhat resembling modern scientific inquiry or historical dissertations, as if these thinkers were interested in cataloguing Indian philosophy as readers’ digests, would in fact be arguing against the dialectical acuity of Indian śāstric masters – the burden of the proof resting upon the ­claimant – and, in most cases, would be left with no other choice but to state the obvious mediocrity of their historical method. 106 The notion of avatāra, used by Bhāviveka to introduce his chapters on the various worldviews, may also carry a theatrical connotation. André Couture discussed this semantic as it appears in the Mahābhārata, the Harivaṃśa, and the Purāṇa-s: “When referring to the ‘descent’ of the gods upon the earth . . . these terms [avatāra, avataraṇa] also evoke the idea of a performance on a stage (raṅgāvataraṇa)” (Couture 2001: p. 313). What is more, besides a few main avatāra-s of major deities, like Kṛṣṇa for Viṣṇu, the avatāra is generally a diminished version of the deity, a particular aspect of it (as discussed in Biardeau 1976). By extrapolation, one can suggest that the MHK’s avatāra of the Mīmāṃsā opponent is a diminished theatrical version of the genuine system. 107 Kawasaki has examined the content of the MHK in a series of works: Kawasaki 1973, 1974, 1976, 1985, 1992a, 1992b. His summary and opinions on the pūrvapakṣa, in Kawasaki 1973, more or less reiterated in his various works, are also summarized in Ham 2016: p. 85. 108 A common expression in Québec, sometimes translated as “Desperate times call for desperate measures!” However, a more literal meaning would be, “For great evils, great means!” 109 MHK 9. 22 kriyātvān na kriyābhīṣṭā kṛṣivan muktyavāptaye | adhītve sati vācyatvān mitakālatvato ‘pi vā ||

The beginning of the verse seems to be corrupted. Ham’s reading does not offer a solution to the problem. His reading of 9.22b is politer: “Or, it is because they, being non-cognitive acts, are verbally expressible. Or, it is because they last [only for] a limited time.” He admits finding the translation difficult, and notes that the Tibetan reading differs from the Sanskrit: “It is because they are non-cognitive acts and because they are verbally expressible” (blo ma yin phyir brjod bya’i phyir). The TJ provides no explanation on this part of the verse. See Ham 2016: p. 203. 110 Ham 2016: p. 52–70.

The beginnings of Mādhyamika doxography  85 111 Lindtner suggests equating the meaning of apūrva with that of dharma. In this reading, apūrva is the revealed content of the Veda, its dharma, which cannot be perceived by any other means of knowledge. 112 In his introductory chapter to the MHK, Bhāviveka singles out vitarka as the cause of ignorance which binds beings to saṃsāra: MHK 1.9–11 kā vā śaktimataḥ śaktir yad anāthāṃs tapasvinaḥ | rāgādinigaḍair baddhān ghore saṃsāracārake || pramādam adirāṃ pītvā prasuptān mohanidrayā | vitarkataskarāśeṣaviluptaśubhas aṃcayān || prajñāniśitanistriṃśacchinnaniḥśeṣabandhanaḥ | mukto na mocayed enān yad ayaṃ karuṇātmakaḥ ||11|| What could the power of the powerful be [for], [if] he, the liberated one whose bondage is completely cut with the sharp sword of wisdom, does not liberate those who are not freed; the unprotected, the afflicted, bound by the shackles of passion and so on, in the ghastly prison of the continuous flow of existence, falling asleep with the slumber of delusion* after having drunk maddening liquors, whose collection of merit is ruined to naught by the thief of dubious reasoning. *Note that Annette L. Heitmann (2009: p. 46) gives ghoranidrayā (dreadful sleep) instead of mohanidrayā, at 1.10a. She also offers her own German translation in Heitmann 2004: p. 20–24. 113 On the various types of fallacious reasoning discussed within the Indian philosophical tradition, see Matilal 1977: p. 79, 1990: p. 86; Matilal, Ganeri, and Tiwari 1998: p. 43; Vidhyabhusana 1913: p. 15, 1920. 114 MHK 9.24

prāmāṇyam atha vedasya sādhyate ‘kartṛkatvataḥ | asādhāraṇatā hetoḥ syād asiddhārthatāpi ca || 115 On the development and methodology of Indian logic, one can be referred to Ganeri 2001a; Ingalls 2001; Matilal 1990; Matilal, Ganeri, and Tiwari 1998; Stcherbatsky 1970; Vidhyabhusana 1920. 116 MHK 9.40a yādṛk phalam adhiṣṭhāne dṛṣṭe hi kurute kriyā |

117 118 119 120

I decided to insert the qualifier ‘ritual’ in parentheses because, although the verse could be directly aiming at ritual activity, it equally applies to all action in general. Also, I translate adhiṣṭhāna as ‘location’ in keeping with the semantic of the word. However, the term seems to refer to the original ‘basis’ of action, the agent. Lindtner translated it as ‘authority.’ Further discussion on mantras in Bhāviveka’s thought, in defence of their usage within Mahāyāna Buddhism, can be found in Braarvig 1997. See section 1.2 The Doxographical Content and Discourse of the MHK. The “Three-fold” (trayī) Veda refers to the Ṛg, Sāma, and Yajur Veda. MHK 9.11 devarṣijuṣṭaśiṣṭeṣṭaṃ purāṇaṃ vartma śobhanam | vedārthabāhyaiḥ strīśūdrair yuktaṃ yat tyajyate trayī ||



Lindtner took some liberties with 9.11b: “That the three Veda are rejected by women and śūdra-s who have nothing to do with the Veda is, of course, quite logical.”

86  The beginnings of Mādhyamika doxography 121 MHK 1.10 (emphasis added) pramādam adirāṃ pītvā prasuptān mohanidrayā | vitarkataskarāśeṣaviluptaśubhas aṃcayān || 122 MHK 9.71 tṛṣṇayā pāti lokaṃ vā tṛṣṇādāsaḥ kathaṃ kṛtī | kāruṇyāc cet kathaṃ lokaṃ māyayā samamūmuhat || 123 MHK 9.62 iti bruvāṇaiḥ sanmārgān naṣṭair anye ‘pi nāśitāḥ || 124 Lindtner makes no mention of the possible pun on the name of Śiva in 9.74, and simply translates it as ‘blissful.’ anya eva asau harer mūrtiḥ śivā yadi vikalpyate | dṛṣṭvā hi yatayo yāṃ na punar yānti punarbhavam || If an altogether different manifestation of Hari is fancied, it is auspicious (śiva). One which, upon having seen it, of course, the ascetics undergo no longer the ever-new becoming.

Yet, the other manifestation of Hari, here, is said to be ‘blissful’ indeed, but more importantly, it is Śiva, the chosen deity of ascetics. 125 The insertion in brackets refers to mūrti, in feminine, mentioned in the previous verse (9.74). 126 MHK 9.75 na satī na asatī ca asau na asau sadasatī matā | tasmāt sattvād asattvāc ca sadasattvāc ca sā parā ||75|| 127 Lindtner does not associate the form of acyutā to Viṣṇu-Kṛṣṇa, an epithet which is often used in the BhG, but he literally translates the term as “permanent.” However, if the meaning was merely to be ‘permanent,’ the logic would be altered. The reason, ‘because of being different,’ has little to do with permanence, but rather refers to the hypothesis that Śiva be different from the form of Acyutā (Viṣṇu-Kṛṣṇa/the Immovable One). The term acyutā comes back again in verses 79, 80, 81, and 82. 128 MHK 9.77 atha apy aśāntā tasya ekā śāntā anyaikātmanaḥ sataḥ | bandhakī nāma sādhvī syāc charīrārdhena saṃyatā || Moreover, if his nature is such that one [part] is troubled while the other one is at peace, [then,] by [intimately] bonding with one half of his body, the chaste lady should be called a whore.

The translation of MHK Chapter 9 presented by Lindtner tends to mellow down the tone of Bhāviveka. Yet, the remark is meant to be offensive. In Mahābhārata 2.61.35, the pejorative term bandhakīis applied to Draupadī by Karṇa who insinuates, since it has been ordained by the gods (devair vihitaḥ), that women should only have one husband, and that she, since she submits herself to more than one husband, surely is a prostitute. In the present passage, Bhāviveka seems to redirect Karṇa’s attack towards the divine consort, for the same reasons.



The Sanskrit verses of the MhB read as follows: 2.61.35a–35c: eko bhartā striyā devair vihitaḥ kurunandana | iyaṃ tv anekavaśagā bandhakī iti viniścitā |

The beginnings of Mādhyamika doxography  87 (as per Mahābhārata: Sabhaparvan; Electronic text (C) Bhandarkar Oriental Research Institute, Pune, India, 1999) 129 MHK 9.79b

na kāraṇaṃ na cānyāsau nācyutaś ca apy ajātitaḥ || 130 MHK 9.83b katham ālambhamānās tāṃ mucyante nirmumukṣavaḥ || 131 MHK 9.84 indriyāṇīndriyārthebhyaḥ kūrmo ‘ṅgānīva saṃharan | oṃkāraṃ vyāharan smṛtyā tadbhakto mucyate yadi || 132 BhG 2.58 yadā saṃharate cāyaṃ kūrmo ’ṅgānīva sarvaśaḥ | indriyāṇīndriyārthebhyas tasya prajñā pratiṣṭhitā || To recall Émile-Louis Burnouf’s translation: “Si, comme la tortue retire à elle tous ses membres, il soustrait ses sens aux objets sensibles, en lui la sagesse est affermie” (Burnouf 1861: p. 37). 133 In Patrick Olivelle’s comments on this passage of the ŚU, he observes, “The general meaning is that when God has created and then withdrawn the entire creation into himself, one sees that he is distinct from all the created ‘realities,’ which term here refers to the constituents of the world” (Olivelle 1998: p. 627). The process recalls the one of the meditating yogi aspiring to isolate his self (puruṣa) from worldly manifestations (prakṛti). 134 YS 2.54

svaviṣayāsaṃprayoge cittasya svarūpānukāra iva indriyāṇāṃ pratyahāraḥ |

Michel Angot translates the verse as follows: “Le pratyāhāra‘retrait [des sens]’ est comme une imitation de la nature du mental par les sens, lesquels ne sont plus jointifs à leurs domaines respectifs” (Angot 2012: p. 528).



The SSS (8.36b) also mentions pratyāhāra in the context of the Mīmāṃsā: pratyāhārādikaṃ yogam abhyasyan vihitakriyaḥ || [By] the repeated yoga practice consisting in withdrawal and so on, [by] the act enjoined [by the Veda].

135 For examples, see Maitrāyaṇiyabrāhmaṇa Upaniṣad 6.4, Chāndogya Upaniṣad 5.1–5. 136 For an exhaustive account of the symbolic developments and contemplative usages of the Oṃ syllable from ancient to modern times, see Gerety 2015. 137 Müller 1879: p. 458. 138 Gerety 2015: p. 371. 139 The expression yoga-yukta appears in BhG 6.29: sarvabhūtastham ātmānaṃ sarvabhūtāni cātmani | īkṣate yogayuktātmā sarvatra samadarśanaḥ ||.

In Flood and Martin 2015 (emphasis added), the whole verse is rendered as: He whose self is yoked by yoga, and who perceives sameness always, will see the Self in all beings and see all beings in the Self.

140 MHK 9.85–86 muktir na haribhaktānāṃ yujyate haridarśanāt | vikalpasmṛtiyogatvāt tadyathā haridarśanāt ||

88  The beginnings of Mādhyamika doxography



nirvikalpāpi dhīr neṣṭā yogayuktasya muktaye | nimittagrahaṇān mithyā kiṃ punaḥ parikalpitā || Lindtner’s translation of 86b misses the point of the argument. He translates as follow: This is because it falsely holds on to signs. So much the more a mind that is full of images!

141 142 143 144

However, the crux of the matter, as previously discussed and as indicated in the immediately following verse, is because of ‘causality.’ Liberation is not produced or caused by something. Halbfass 1991: p. 226–227. Qvarnström 1999: p. 174. For a translation and commentary, see Chapple and Casey 2003. MHK 9.92 vedayogopadeśādi tad ukter vikalatvataḥ | na tāvat tathyadharmoktyā śaktās te dharmaguptaye ||

145 Verse 9.107, wherein culminates the metaphor of the lord’s impotency, recalls the previous assertion in verse 99b. Although, this time, what is involved is not only “dependence on others” (para-tantratva), but “dependence on self and others” (sva-para-tantratva). Curiously, in translating, Lindtner omitted this distinction and rendered the passage as “Since pleasure only depends on itself [for motivation].” This fascinating conception of ‘pleasure’ might be suitable in a celibate context, yet it fails to convey the full meaning of Bhāviveka’s assertion. In fact, if pleasure was only “self-dependent,” as suggested by Lindtner, the reasoning would be fallacious. For, as in verse 99, Bhāviveka stresses that it is because the lord must rely on others (for his potency as for his pleasure) that he is impotent. In theory, there would be no ‘logical’ problem if the lord could satisfy himself alone. 146 MHK 9.108–109 anyonyabhakṣaṇād bhītais tiryagbhir durlabhotsavaiḥ | niṣpeṣacchedadāhādiduḥkh ārtair nārakair api || nṛbhir janmajarārogabhayaśokaklamārditaiḥ | prīyate yo namas tasmai rudrāya anvarthasaṃjñine || 147 MHK 9.110–111 kṛpaṇā dhanino yad vā parānnādāś ca sāttvikāḥ | svarge cādharmiṇaḥ kecid vyaktam īśvaraceṣṭitam ||110|| alpāyuṣo guṇadhanā durvṛttāś ca cirāyuṣaḥ | dātāraṃś ca alpavibhavā vyaktam īśvaraceṣṭitam ||111||| Compared with the Biblical verses (King James) of the Ecclesiastes: 6:1 There is an evil which I have seen under the sun, and it is common among men: 6:2 A man to whom God hath given riches, wealth, and honour, so that he wanteth nothing for his soul of all that he desireth, yet God giveth him not power to eat thereof, but a stranger eateth it: this is vanity, and it is an evil disease. 7:15 All things have I seen in the days of my vanity: there is a just man that perisheth in his righteousness, and there is a wicked man that prolongeth his life in his wickedness. Surely, Bhāviveka’s grievances against god are an old and common theological trope. The Mādhyamikan’s rhetoric exploits it to exacerbate doubt. 148 MHK 9.114–115

saṃyamitamatidvāraḥ sthāpayitvā śive manaḥ | tathauṃkāram abhidhyāyan dhārayan dhāraṇāṃ hṛdi ||

The beginnings of Mādhyamika doxography  89 kṣityādidhāraṇābhyāsāt prāk samāhitamānasaḥ | īśe prasanne duḥkhāntaṃ gacchati ity etad apy asat || 149 MHK 9.119 etena śeṣāḥ pratyuktā brahmaviṣṇvātmavādinaḥ | prītiś caivam ayuktatvān naiśādau dhīyate dhiyaḥ || 150 MHK 9.126 na dānagrahaṇaṃ yuktaṃ puṇyāder iti niścayaḥ | cittena saṃprayogitvāt tadyathā sukhaduḥkhayoḥ || 151 Transference of merits and its problematic is discussed in Gombrich 1971; Lehtonen 2000; Ruegg 2004. 152 Olivelle 2007. 153 Mohan and Sivaramalingam 2014. 154 The Jain practice of ritual suicide through fasting continues to stir controversies to this day. In August 2015, the Rajasthan High Court had declared the practice of saṃthāra or sallekhanā within the Jain community unlawful. After intense petitioning by the Jaina community, India’s Supreme Court later upheld the practice. The controversy reemerged in 2016 after a 13-year-old Jaina girl died from fasting in Hyderabad. Jainas argue that the practice is not suicide, which they say is an emotional and impulsive act, but rather a conscious act of spiritual purification. Bhāviveka would reject the latter argument, pointing out that renouncing life cannot alleviate former misconduct (pāpa), no more than it can produce any merit (puṇya), because “the give-and-take of merit and so on is not logical” (v. 126). Hence, no purification of anything is produced by fasting until death; it is only self-destruction. In itself, this case serves as a reminder that the dialectician lived in an era when, though politics had its own dynamics, he did not have to bother about offending ‘religious sensitivities’ when composing treatises. Or, more plausibly, this is precisely what he wanted. 155 For a discussion on the fasting Buddha in canonical literature and Gandhāran art, see Rhi 2006 (2008). 156 Jaina texts distinguish between the “death of a wise man” and that of a fool. The former corresponds to sallekhanā and is interpreted as “to thin the passions and the body properly”; see Hotta 2009. On this basis, the label ‘suicide’ is rejected by the Jainas. 157 BhG 3.4 na karmaṇām anārambhān naiṣkarmyaṃ puruṣo ’śnute | na ca saṃnyasanād eva siddhiṃ samadhigacchati || The translation is from Flood and Martin 2015. 158 Stewart 2015: p. 47. 159 On the subject, though not exhaustive: Bhatla, Mukherjee, and Singh 1984; Edwards 1922; Haberman 2013; Jakubczak 2017; Schmithausen 1991, 2009. 160 On the term dohada, literary meaning “sickness of heart” or “nausea,” and its motif in literature, see Jakubczak 2017: p. 85–86. 161 MHK 9.147 sattvakarmādhipatyena kālajāḥ pādapādayaḥ | narake svargaloke ca śastraratnadrumā yathā || 162 Concerning the sentience of plants, Lambert Schmithausen remarks that, in India, at least on a theoretical level, being a living or animated being (e.g. p(r)āṇa/p(r)āṇin, satt(v)a, jīva, bhūta) is by and large equated with being “sentient” ((sa)cetana, sacittaka, cittamaṃta), capable of perception. In doctrinally developed Buddhism, however, apart from humans and mythological beings, only animals are also considered

90  The beginnings of Mādhyamika doxography

163

164 165 166 167 168

169 170 171

sentient. This Buddhist view breaks away from those of the Vedic and Jaina traditions (Schmithausen 1991: p. 1–2). See also Schmithausen 2009. Many Indian descriptions of hells mention forests of trees having leaves like blades (asi-patra-vana), cutting down the wrongdoers in pain who approach or climb over them, trying in vain to escape from other torments. Indian imagination envisioned all sorts of hellish torturing trees, even cotton trees with long thorns (kūṭa-śālmalika). The Mārkandeya-purāṇa mentions seven hells; the sixth one is the asi-patra-vana. The Buddhist Devadūtasuttaof the Majjhimanikāya is also particularly creative concerning hells. For an overview of the grim subject in South, South East and Central Asia, see Zin 2014. Such hellish ecology continued to inspire other Buddhist cultures, like the Tibetans. The lam rim chen mo of Tsongkhapa, for example, abundantly indulges in their descriptions. A recent and appealing call has been made, in Emanuele Coccia 2016, to rethink the vegetal realm, in order to develop a new “philosophie de la nature,” reasserting the fundamental role of plants in the development of sentience. All following translations of verses numbered beyond 148 are taken from Lindtner 2001a: p. 50–53. I inserted the commas to isolate “here in this circuit,” to prevent a possible misreading. Finding a jewel in heaps of rubbish is a common trop of Buddhism, referring to the birth of the “spirit of awakening” (bodhicitta), as in Bodhi-carya-avatāra 3.27. The political notion of ‘religious pluralism’ has been defined in Religious Pluralism, Globalization, and World Politics as “the interaction of religious actors with one another and with the society and the state around concrete cultural, social, economic, and political agendas. It denotes a politics that joins diverse communities with overlapping but distinctive ethics and interests. Such interaction may involve sharp conflict. But religious pluralism as defined here, ends where violence begins” (Banchoff 2008: p. 5). Sahih Muslim 598 a–b. As can be observed on the Wikipedia entry of Bhāviveka, in English, under ‘Biography,’ as last seen on 30 November 2017. It is also suggested in Bronkhorst 2010b: p. 32. PV 1,340 vedaprāmāṇyaṃ kasya cit kartṛvādaḥ snāne dharmecchā jātivādāvalepaḥ | saṃtāpāraṃbhaḥ pāpahānāya ceti dhvastaprajñāne pañca liṅgāni jāḍye ||

172

173 174 175 176 177 178

179 180 181

The quotation of Ham 2016: p. 96 reproduces the translation of Eltschinger, Krasser, and Taber 2012: p. 77–78. The expression is attributed to G. K. Chesterton. On yawning, Cedric Mims humorously observed, “Yawns have been given all sorts of meanings. G. K. Chesterton said that a yawn was a ‘silent shout,’ and others have gone further, suggesting it signifies aggression, a symbolic attempt to eat a victim” (Mims 2014: p. 324). Raskin 1984: p. 2. Kierkegaard 1855: p. 248–249. Kierkegaard 1855: p. 249. The original version is from Bergson 1899. The mention comes from ‘Comedy’ 1956: p. 187, as quoted in Raskin 1984: p. 17. Kierkegaard 1855: p. 258. Helmut Krasser showed how certain ‘digressions’ or ‘appendices’ in Bhāviveka’s PP refer to discussions found in the MHK and TJ. His paper particularly focuses on a digression related to Chapter 9. “These digressions seem to have been written down by students who were being trained to debate as based on their teacher’s oral instructions” (Krasser 2011: p. 49). Thus, one can safely assume that Bhāviveka’s work was indeed studied by monks, for dialectical purposes, at least from the sixth century onwards. Eltschinger 2014: p. 30. The last sentence of this quotation is equally mentioned in Eltschinger 2008: p. 520. It is a quotation from Kapstein 2003: p. 15. I first made the ‘vaccination’ metaphor in relation to doxography in Bouthillette 2015: p. 63. Richard King (1995: p. 202) observed the irony of Bhāviveka when parodying the inclusivism of Vedānta.

2

The beginnings of Jaina doxography Haribhadra’s ṢDS

2.1  Haribhadra’s doxography in context Among Indian doxographies, the Ṣaḍdarśanasamuccaya (ṢDS) of Haribhadra Sūri captured the lion’s share of attention from the few scholars who have reflected on the literary phenomenon of philosophical compendia. Depending on how they defined the genre, some, like Halbfass, saw the ṢDS as the first of its Indian kind. The late scholar also considered the ṢDS to be among the two most interesting and significant doxographies, along with the Sarvadarśanasaṅgraha (SDS) often attributed to Mādhava Vidhyāraṇya,1 which he estimated to be from the fourteenth century.2 Whether or not one considers the text to be the first Indian systematic philosophical doxography,3 there are few competitors to challenge its pioneering role among Jaina doxographies. The only other possible candidate could be Siddhasena Divākara’s Dvātriṃśikā, which influenced Haribhadra’s work, as mentioned by Qvarnström, but which is lacking in systematic structure.4 Nicholson also noted that the ṢDS is the earliest Indian doxography to be “widely edited and translated in the modern period.”5 However, the broadest discussion to date, on the text and its authorship, date, and content, is from Kendall Wayne Folkert’s doctoral dissertation,6 parts of which are collected along with other papers and articles in a posthumous publication,7 following his untimely death in 1985. Concerning the dates of the ṢDS, Folkert maintained that the earliest possibility should not be much before AD 700,8 and that the eighth century appears to be a safe conclusion. Concerning the debated question of authorship, Folkert finds it “prudent” to follow the Jainas’ lead and to accept the famous Śvetāṃbara scholar as the author.9 It is worth observing that, besides Folkert, none of the previously mentioned scholars seem to question Haribhadra’s authorship. Before discussing the doxographical method and purpose of the ṢDS, it is indispensable to mention the sociohistorical ground in which this philosophical development germinated, as was done with Bhāviveka’s MHK, by exploring what we know of the life and work of Haribhadra, along with a brief survey of the most meaningful ideological precursors to Jaina doxography. The Jaina philosophical context provides invaluable insights into the nature and function of Indian doxography. Fortunately, some work has already been done in that direction. However, although the ṢDS has been discussed by scholars before,

92  The beginnings of Jaina doxography much remains to be said. Its idiosyncrasies make it an unavoidable component of the present research. Perhaps this is mainly because its overall dialectical structure, often reproduced in Jaina circles, presents a middle way between that of the other two main doxographies under study; but it could also be because its apparent absence of refutation, which puzzled its many critics, sets a literary precedent which has not revealed all its secrets. It has led to hasty and long-lasting assumptions. Hence, the work of previous scholars, valuable in many respects, will allow me to focus on the themes relevant to doxography without being carried away by philological details which, however interesting, are not of immediate concern here. In the end, the translation of Chapter 6, on the Jaiminīya-s, though remarkably short, will serve to illustrate the content of the ṢDS, so as to contrast it with the longer chapters on the same philosophical system within the other doxographies presented within this study. Again, the purpose is not primarily to obtain further knowledge on the Mīmāṃsā, though it does come as a side effect, but to better understand Indian doxography based on its early manifestations. From now on, let’s turn our eyes to the author of the ṢDS, the famous Haribhadra Sūri. 2.1.1  Life and work of a grieving ‘doxographer’10 Concerning Haribhadra’s life, one can find it mentioned in several Jaina sources, including Rājaśekhara’s Prabandhakośa, Bhadreśvara’s Kahāvalī, and Prabhācandra Sūri’s Prabhāvakacaritra, which have been summarized by Hermann Georg Jacobi.11 More recently, Ram Sajiwan Shukla conveniently compiled the data gathered from these many stories, along with information from Haribhadra’s own writings, to convey a picture of India as it was known to the celebrated Śvetāṃbara Jaina.12 Christopher Key Chapple also contributed a chapter on Haribhadra’s life.13 Some elements of this legend are worth mentioning, since we rarely have the privilege of knowing much about an author’s life and can thus gain a glimpse of his alleged character and erudition according to ancient narratives. At the outset, the suggestion that Haribhadra was born a Brāhmaṇa, son of Śaṅkarabhaṭṭa and Gaṅgā, and lived in Citrakūṭa, identified with Chittor, the capital of Mewar in Rajasthan, where he is said to have performed the duties of a royal priest, before converting to Jainism, is not trivial.14 Brahmanical education likely informed Haribhadra’s erudition and concern for non-Jaina theorization. The account of his conversion gathered from the Prabhāvakacaritra suggests that he had taken pride in his knowledge and enjoyed the fame brought to him by victorious public debates. He is said to have declared in his youth that he would only accept tutelage under one whose teachings he would fail to fathom. As the legend goes, in Citrakūṭa, he came to hear a Prākrit verse15 from a Jaina nun called Yākinī Mahattarā, which he could not comprehend. Upon his request for an explanation, she directed Haribhadra to her preceptor, Jinadatta Sūri, an eminent Śvetāṃbara ācārya. Yākinī became Haribhadra’s spiritual mother, from whom he got the name Yākinīputra (Son of the Yākinī). Haribhadra is also said to have studied under Jinabhaṭṭa, the head of the Vidyādharagaccha.16 After becoming a yati, or Jain monk, having received his dīkṣā from Jinadatta, he is said to never have stayed

The beginnings of Jaina doxography  93 permanently at one place but constantly wandered from one place to the next. This first determining life event, although legendary, allows us to appreciate the pivotal role which the Jaina tradition attributes to debate and religious learning in Haribhadra’s early life. At this stage, however, this high-visibility pursuit betrays a juvenile character. It recalls the classic tale of every bright youthful spirit fully given to worldly pursuits, carelessly boasting of success. One is tempted to label this period the ‘Wolf of Citrakūṭa.’ However, in another account of Haribhadra’s life story, the wind seems to turn, and the debate scene of medieval India dramatically reveals its wrathful and uncompromising face. Two of the Jaina monk’s nephews, who had studied logic under him, took it upon themselves, against Haribhadra’s advice, to infiltrate a Buddhist stronghold in order to learn the doctrinal points which they were successfully using against the Jainas. When their true identity was finally discovered, after many legendary feats, one was killed, while the other is said to have died later on, out of grief. Overwhelmed by his loss, Haribhadra would have brought vengeance upon the Buddhists in the form of a public debate at the court of King Śūrapāla. The penalty for defeat entailed being boiled alive in an oil cauldron. The Buddhist debaters suggested the topics of disputation but were ultimately defeated by Haribhadra, who oversaw their cruel punishment. Haribhadra’s teacher, it is said, ordered him to undertake severe penance for having displayed such hatred, a remorseful episode which supposedly led to the redaction of the Samarādityakathā (Samarāicca Kahā17), reflecting on the consequences of hatred. It is said that grief from the loss of his beloved nephews never left the author. He expressed this feeling by signing several of his works as virahāṃka, suggesting ‘separation’ or ‘absence.’ Needless to say, these various names contribute to the challenge of assessing Haribhadra’s authorship. But there is another major reason why modern scholarship is struggling with the Jaina account of Haribhadra’s overall authorship. According to such authorities as Abhayadeva Sūri, Municandra Sūri, and Vādinadeva Sūri, Haribhadra would have composed nothing less than 1400 or 1444 prakaraṇa-s, a number which recalls Herodotus’s accounts of Xerxes’s invasion force. Even if the term prakaraṇa is meant to denote a short treatise or any section of a book, these figures remain unbelievable and are to be taken symbolically. According to Lal Sanghavi, the total number of original titles composed by Haribhadra is rather forty-seven, while twenty-six are dubious. Nemicandra Shastra lists ninety-three titles, forty-three of which are dubious. As for Kalyanavijaya, he reckoned eightyseven or eighty-eight works.18 Even while contemplating these more conservative accounts, the literary activity of the famous Śvetāmbara remains impressive. It is believed to consist of fourteen philosophical treatises, five yoga texts, two ballads (kathā-s), one astronomical treatise, two eulogies (stuti-s), fourteen discussions on topics gathered from the scriptures (āgama-prakaraṇa-s), and nine commentaries on various scriptures (vṛtti-s and ṭīkā-s). Haribhadra wrote in Prākrit as well as in Sanskrit, in verse and in prose. The thinker is known to have pioneered the use of Sanskrit within the various branches of Śvetāṃbara literature.19

94  The beginnings of Jaina doxography As a staunch logician and debater, besides Jainism, he was well versed in the many philosophical views of his time, as suggested by the content of the ṢDS, among his numerous treatises where opposite views are discussed. His general work refers to well-known earlier scholars, among which are Bhartṛhari, Dharmakīrti, Kumārila, and Śubhagupta. Praising the phenomenal erudition of Haribhadra, Shukla noted, Judged both from the volume and variety of his contributions, Haribhadra surpassed not only his contemporary Jaina writers, but almost all of his predecessors and successors. He revolutionized the age in which he lived and was very aptly called Yugapradhāna.20 In other words, Haribhadra was no common man. His literary activity bears witness to his intellectual erudition, grasp of philosophy, and devotion in religious life. Unlike modern-day armchair philosophers and scholars, his constant travels unmediatedly informed him about the various people of the land, their customs and beliefs. If legends about him are of any value, they reveal a man who experienced firsthand the ups and downs of religious controversies, as channelled through public debates; a man whose perspicacity could lead him to victory over his opponents, and perhaps even to see beyond such mundane verbal skirmishes. It is no wonder that he came to systematize a meta-philosophical approach to heterodoxy, which I seek to highlight within these pages, by decoding his doxography, the first of the kind in Jaina circles. Before moving on to discuss the ṢDS itself, more needs to be said about the ideological environment in which Haribhadra would have contemplated and philosophized. 2.1.2  Ideological precursors of Jaina doxography The contributions of Jaina thinkers to Indian intellectual life, though understudied, should not be underestimated. Melanie Barbato aptly noted how Its realist outlook distinguishes it from the better known idealist forms of Indian thought. Jainism’s radical dualism between soul and matter and its elaborate karma theory also make it a school worth of study in its own right. The most remarkable aspect of Jaina philosophy is, however, that through anekāntavāda it functions as a bridge which connects the opposing views of Indian thought.21 The doctrine of anekānta-vāda, a theory about the many-sidedness of reality, slowly emerged as a crucial feature of Jaina’s systematic philosophy. Not only did it shape ontological and epistemological Jaina theorizations, but also it was designed to play a significant soteriological function within the Jaina religious path. Among Indian philosophical positions, it appeared as a middle way between the views of the Buddhists and those of the Vedāntins. Its understanding

The beginnings of Jaina doxography  95 is essential in decoding Haribhadra’s doxography. He himself composed three treatises on the topic and another one on the related syād-vāda, the quodammodo22 method of conditional predications: Anekāntajayapatākā, Anekāntavādapraveśa, Anekāntasiddhi, Syādvādakucodyaparihāra. Thus, the present discussion will briefly survey the topic of anekānta-vāda from three different angles. First, I will sketch the historical setting of ‘dialogue and heterodoxy’ which contributed to the development of anekānta-vāda. Then, I will briefly introduce the dialectical strategies of anekānta-vāda, before concluding on their soteriological function within the Jaina path. Then only can the question of Jaina doxography be reviewed. 2.1.2.1 Dialogue and heterodoxy: the dialectical soil of anekānta-vāda The earliest historical seeds of anekānta-vāda were sowed in the fertile ideological soil of the dialogical world known to both Mahāvīra and the Buddha. The effervescent socioreligious atmosphere of the days raised high expectations towards the many leaders of thriving mendicant orders. As witnessed in both canons, distinct knowledge claims were already competing in the arena of ideological dispute, where various contenders wrestled to assert their spiritual authority. These knowledge games fostered varying attitudes towards philosophical questioning. The purpose of such dialectical engagement, in the end, was no mere play, but constituted the very touchstone of a teacher’s awakened mind. For this reason, the dialectical strategies devised by the leading contenders were not only meant as rhetorical devices, but also as a display of the very path which they were claiming to master and initiate others into. In other words, the way sect leaders go about facing difficult questions already indicates the way their followers later approach religious life. For example, in contrast to those who held extreme views (ekānta-vādin-s), the Buddha is known to have told Māṇavaka that he was a vibhajya-vādin.23 He also told Poṭṭhapāda24 that, contrary to his critics, he had taught both doctrines (dharma) which could be asserted categorically (ekaṃsika pi) and others which could not be asserted so (anekaṃsika pi). K. N. Jayatilleke and B. K. Matilal suggested that the latter kind could be taken as an early precursor to the Jaina anekānta.25 In fact, the Jaina Sūtrakṛtāṅga (1.14.22) also claims that Mahāvīra was a vibhajya-vādin. Thus, Matilal argued that both the anekaṃsika and the vibhajyavāda attitudes were forerunners of the anekānta-vāda. He explained vibhajya-vāda as a method of analysis and differentiation applied to questions which cannot be answered directly by a definite ‘yes’ or ‘no,’ hence questions which are not ekaṃsa (not one-sided: an-ekaṃsa) but which are answerable through analysis, or ‘breaking-up’ (vibhajya). The Buddha, who did not want to adhere categorically to extreme views, adopted the vibhajya method when faced with metaphysical questions to which no determinate answer could be given. He analyzed their presuppositions and broke down their arguments, either to show that their formulation was defective or that they were of no help to religious life. Matilal explains,

96  The beginnings of Jaina doxography Sometimes such a question may be resolved into a number of separate questions answers to which should be sought separately. (That explains why the Buddha remained silent when a ‘compounded’ question was put to him directly.) Sometimes, the question may dissolve itself in the face of an ‘analysis’ to which it would be subjected. In the latter case, the question can thus be identified as a pseudo-question. In fact, this latter one was the method the Buddha seemed to have followed in most of his dialogues.26 This last method, if need be recalled, also inspired the dialectic of the Madhyamaka, championed by Bhāviveka, who systematically applied its enhanced version to every system of thought presented within his MHK, where they feature as little more than ‘pseudo-systems.’ The dialogical attitude of the initiating master, the Tathāgata, carried on within his tradition, continuously informing his followers’ spiritual attitude towards views in particular and truth in general. It also percolated through the development of systematic doxography. For, essentially, doxographies are literary technologies built upon dialectical strategies designed to deal with a variety of questions or perspectives. To put it differently, doxography answers compounded questions. Questions raised in dialogue. Their answers, like the ones of the ancient vibhajya-vādin-s, are meant to ‘dissolve’ the plurality of compounded questions. Silence is victory. Coming back to the Jainas, Matilal distinguished two varieties of vibhajya-vāda. The first type operates by dividing the subject class into sub-classes, while the second type proceeds by specifying or relativizing the predicate. According to Matilal, Mahāvīra preferred the second group. Progressively, this dialogical attitude was systematized by Jaina thinkers to become the famous anekānta-vāda, a dialectic which avoids extreme positions by putting each claim into perspective. Matilal suggested that, unlike the Buddha, who tended to withdraw from philosophical disputes by avoiding any commitment to extreme positions, the Jina rather espoused commitment by listening to all sides of the dispute and by seeking a conciliatory way encompassing all perspectives. Thus, he said, The essence of the anekānta-vāda lies in exposing and making explicit the standpoint or presuppositions of different philosophical schools.27 In the same vein of ideas, Matilal acknowledged that both religious leaders, Buddha and Mahāvīra, aimed to set forth a ‘middle way,’ but he remarked that the way of the Buddha was rather based upon rejection and exclusion whereas the one of Mahāvīra involved acceptance and inclusion.28 However appealing, the idea that conciliation or inclusivism motivated Mahāvīra’s strategy and that it was not, rather, a clever dialectical way to distinguish his own position among the many, an attitude which could be expected from a religious leader having to assert his authority, is not evident.29 As will be discussed within these pages, the principles of anekānta-vāda, when applied within a doxographical scheme, tend to suggest a pattern of subordination

The beginnings of Jaina doxography  97 rather than one of conciliation. Thus, if the dialogical attitude of the initiating master carried through within the tradition and percolated through doxographical schemes, as I suggest it did within Buddhism, it is perhaps warranted to repeat the reasoning backward and, by induction, to postulate that the Jina, as with the anekānta-vāda in later Jaina doxography, by resorting to his peculiar use of vibhajya-vāda, was strategically establishing the superiority of his vision, overseeing that of all the others. Though this method seemingly accommodates competing claims, it also trivializes their core position as limited perspectives on reality, turning the opponent’s system into a mere province of the wider realm of knowledge. By comparison, a nation being subjugated by, let’s say, the British Empire, and turned into a mere province or Dominion of the crown, seldom takes pride in being ‘included’ in the Commonwealth. To praise such broad-minded ‘inclusion’ of the new ‘subjects’ is likely to pass as an insult. Whether or not the logical intricacies of their discourse displayed some inclusive or exclusive patterns, the dialogical strategies of the Buddha and Mahāvīra contributed to distinguish their insight from those of competitors, in a way which asserted their authority over others as enlightened leaders. There is little reason to believe that this was not intended in the writings of canonical literature as well. When these strategies were translated into religious paths, their followers, who seek the same knowledge, which transcends mere views, emulated their leader’s dialectical ways. They too had to learn to recognize and overcome ‘mere views.’ For, views shape behaviour. But the way that both groups, Buddhist and Jaina, went about achieving this goal differed in many respects. To better understand their need for doxographical literature, it is necessary to investigate how the Jainas proceeded in their epistemological and soteriological endeavours and continued to evolve once deprived of the immediate tutelage of Mahāvīra. It is impossible, within this limited overview, to retrace the long evolutive process of the Jaina’s philosophical jewel. Hence, I will limit my observations to the determining influence of heterodoxy over the formulation of anekānta. At the outset, Matilal noted that the development of anekānta-vāda happened within a historical window wherein Jaina writers began to adopt the linguistic medium of Sanskrit; which by then, roughly around the fifth/sixth century CE,30 had become the lingua franca of India. In comparison, this linguistic shift had already started to affect the north-western Buddhists under the Kuṣānas in the early first century.31 As mentioned, Haribhadra was one of such Sanskrit literati, active within the Śvetāṃbara community. This new linguistic trend within the Jaina communities indicates a renewed impetus to relate with outsiders (pavayaṇa-nihṇaga/ tīrthika),32 most likely in order to compete on the same grounds. But also it signals the growing dominance of Brahmanical culture, not only in literature, but also at the court. By the time of Haribhadra, a Brahmin by birth, familiar to court disputes, not only was the brahmanization of India well on its way, but competing religious communities had long been systematizing their doctrinal tenets. The Jainas had been following suit with the formulation of anekānta-vāda. Already

98  The beginnings of Jaina doxography in the second century BCE, in Greater Gandhāra, the systematization of Abhidharma’s thought had given rise to the Buddhist Sarvāstivāda school, perhaps the first systematic philosophy of the subcontinent.33 The emerging system brought forward original doctrinal elements, like the theories of dharma-s and pudgala, which, according to Bronkhorst, were borrowed by the Śvetāṃbara canon. Thus, even in its early stages, non-Jaina ideas influenced the development of Jaina thought, which, however, felt no immediate need to adopt Sanskrit as a vehicle of knowledge.34 Although ‘foreign’ elements can be found within the Jaina canon, it contains no mention of either anekānta-vāda or syād-vāda, often used synonymously. The term anekānta-vāda was coined by the logician Siddhasena Divākara (fifth/sixth century), while the first comprehensive account of the doctrine appeared without its name in the work of his predecessor, which he commentated: Umāsvāti’s Tattvārthasūtra (TAS),35 the first Jaina Sanskrit philosophical treatise. In fact, the development of such a philosophical midway position as the anekānta-vāda requires the existence of well-defined antagonistic doctrines. To that effect, Piotr Balcerowicz explains, Historically speaking, the Jaina theory seems to be a dialectical interplay between two basic standpoints concerning the nature of the world and all phenomena.36 Jaina thinkers paid attention to heterodox ideas in the formulation of their own. In articulating the anekānta-vāda, they surely did. Balcerowicz identifies the two opposite standpoints the anekānta-vāda aims at mediating. On one side, (1) the śāśvata-vāda (eternalism) of the Sāṃkhya school, which postulates that all that truly exists must be eternal, permanent, and unchanging. This position can also be identified with the one of the Vedānta, particularly of the Advaita type, which will be discussed in Chapter 3. At the other extreme end of the dialectical spectrum, stands (2) the uccheda-vāda (annihilationism) of the Lokāyatas (materialists), according to which everything is impermanent and destined to destruction. This last extreme could also be represented by the kṣaṇika-vāda (momentariness) standpoint of the Buddhists, which equally maintains that existence is momentary, constantly changing from one instant to the next in a seemingly continuous flow. It is in between these antagonistic worldviews that the Jaina anekānta-vāda intervenes and reveals its philosophical worth. In order to make sense, the anekānta-vāda theorization requires the existence of some primary doxographical system. The same could likely be said of most philosophical doctrines. Philosophy is born in dialogue. Recognizing valid reasoning within each assessable theory, anekānta-vāda stresses the multiplexity of reality over the determination of a single position. It argues that it is a mistake to pretend that reality can be reduced to a single dimension of its manifold aspects: to a single view. The yogin rather seeks the overview. To become omniscient (sarvajña). To better understand how the anekānta-vāda proceeds as a contemplative philosophical way, one must examine its dialectical strategies.

The beginnings of Jaina doxography  99 2.1.2.2 Language and reality: the dialectic of suspicion in anekānta-vāda The term anekānta-vāda can be analyzed as the following compound: an + eka + anta + vāda. It can be rendered, literally, as the theory (vāda) of not (an) one (eka) end/side (anta), or the “thesis which is not one-sided.” Balcerowicz coined the elegant formula “doctrine of multiplexity.” But in order to figure out what this concept stands for, it suffices not to simply break its name apart and to dissert about its linguistic features. Its dialectical strategy must be examined in the light of the worldview it carries. At its core, the anekānta-vāda is motivated by a realist assessment of the limitations of language and, by extension, of the limited reach of the conventional mind. It accepts that words and thoughts can capture but a fraction of anything, of any object. Only an omniscient person, like the Jina, could grasp all the many faces of an object at once. For, as Balcerowicz explains, since the Jainas maintain that everything is interrelated, one knowing a single entity would also know all its modes, in the three times, and all its possible and impossible relations with everything else in the universe.37 From that perspective, one who knows one thing knows all. In other words, to use the expression of the influential fifteenth-century German theologian Nicholas of Cusa, known for his dialectical mysticism, one who fully sees one thing has “divine sight.” Cusa further explained, But Sight that is free from all contractedness – as being the most adequate Measure, and the most true Exemplar, of all acts of seeing – encompasses at one and the same time each and every mode of seeing. . . . And it remains altogether free from all variation. For in Absolute Sight every contracted mode of seeing is present uncontractedly.38 To fully know or see one thing is a mystical feat. But at the other end of the spectrum, ordinary communication, ever ambiguous at best, is unable to convey a full account of reality, of a single object. Upon such a realization, a scrupulous person anxious to behave in a manner which does not suggest that one knows more than one does, so as not to misbehave either verbally or intellectually, because it incurs karma, is led to consciously limit the scope of one’s discourse, in one way or another. It is perhaps why, to paraphrase Balcerowicz, for reasons less related to epistemology and semantics but rather connected with ethics and monastic discipline, the Jainas devised three different strategies to ‘restrain’ or appropriately structure their communication. Balcerowicz further suggests, In fact, we could claim that the doctrine of multiplexity of reality was precisely such an attempt to provide formalisation of the semantic process of communication and interpretation.39 The Polish professor helpfully enumerated and detailed the three disambiguation strategies at play within the anekānta-vāda dialectic. Historically speaking,

100  The beginnings of Jaina doxography the first to appear is the “theory of the four standpoints” (nikṣepa, nyāsa).40 The second is the sevenfold method of conditionally valid predications, known as the theory of viewpoints (naya-vāda).41 The third disambiguation strategy, referred to as syād-vāda, consists in converting every naturally ambiguous proposition, thus every linguistic statement, into an unequivocal enunciation by adding the term syāt (“somehow,” “in a certain sense,” “from a certain perspective”) at the end.42 Each of the three strategies focuses on a different component of communication: (1) The four nikṣepa-s are analytical tools to be applied on a single term or expression, in order to distinguish four possible theoretical semantic fields associated with it. (2) The theory of viewpoints is resorted to by the Jainas, since around the fifth century CE, to elucidate a whole sentence by clarifying its proper context among a range of possible meanings somewhat delimited by seven parameters, which are seven basic viewpoints within which a sentence can be made meaningful. (3) The quodammodo strategy (syād-vāda) is concerned with the object of the communication itself. It aims to provide an accurate description of it through a range of formalized sentences which could cover its overall complexity. This strategy is also sometimes called sapta-bhaṅgī (the “seven-figured” model), in reference to the seven basic formal standards which are said to encapsulate an infinite number of sentences one can say about any entity. In other words, it systematizes into seven possibilities the entire range of possible assertions about reality. This last dialectical strategy is particularly important to keep in mind when discussing Haribhadra’s ṢDS. For now, to sum up, this brief overview of the dialectical complexity of the anekānta-vāda serves to remind us about the profound anxiety the Jainas felt towards the limitations of language and conventional thought, and how cautious they were about truth claims. This attitude is not without relation to the way they conceived knowledge and its function within the path to liberation. 2.1.2.3  The yoga of knowledge: reasoning on anekānta-vāda The expression “yoga of knowledge” (jñāna-yoga), though it is also used in other contexts,43 is here taken from Haribhadra himself. It is mentioned in his Śāstravārtāsamuccaya. Qvarnström noticed the significant formula. He renders the Sanskrit expression as “intellectual exertion,” the path of philosophical studies. Qvarnström explains how Haribhadra highly estimated the pursuit of jñāna-yoga: In order to understand the purpose of studying various philosophical systems according to Haribhadra, it is imperative first to be aware of the place of philosophical studies among other human pursuits. [. . .] Basically life in the saṃsāric cycle of death and birth is miserable (duḥkha). Realizing this, man should devote all his endeavours to achieve temporal happiness (sukha) in saṃsāra, or even better, to obtain final release from cyclic life (mukti). Sukha is the outcome of virtuous conduct (dharma) . . . The positive “energy” or “matter” (karma) thus amassed is bound to entail the happiness of a good rebirth whereas negative karma inexorably retaliates suffering

The beginnings of Jaina doxography  101 (duḥkha) upon the “delinquent”. This, then, is the moral aspect of the twofold dharma of mankind. However, in order to escape from the bondage of karma in its various forms, and thus obtain absolute freedom from any kind of further desire for anything (āśaṃsā) in the turmoil of saṃsāra, one must cultivate the second aspect of the dharma. Haribhadra calls it “intellectual exertion” (jñāna-yoga, saṃjñāyoga) and defines it as “pure austerity” (tapaḥ śuddha).44 Needless to say, the anekānta-vāda theorization is a crucial component of this jñāna-yoga, if not its very axis. The vision of the cycle of rebirth as essentially painful, from which one can be released through the practice of dharma, is a topos of Indian philosophy, certainly shared, with variations, by the three traditions which produced doxographies. The same can be said, more or less, of the idea that to follow a certain code of conduct and to engage in intensive intellectual exertion (jñāna-yoga) are two symbiotic aspects of a ‘dharmic’ life conducive to freedom from sorrow. But has it always been so? It is significant that Haribhadra considers dialectical exercises as pure austerity (tapaḥ śuddha). In doing so, he fully legitimizes such intellectual practices as philosophical studies and debate within the Jaina path, the ethos of which is otherwise built around rigorous ascetic discipline.45 One is led to wonder, however, if Haribhadra did so in order to make an emerging socioreligious necessity acceptable in Jaina praxis, or is jñāna-yoga a natural unfolding of the Jaina path. The anekānta-vāda itself arose as a response to the ubiquitous dialectical phenomenon of debate. Therefore, W. J. Johnson argued, [O]ne of the reasons anekāntavāda was developed was to justify and protect identity-defining ascetic practice. . . . Clearly such a defence only becomes necessary in the context of a formal and public debate between various traditions. And it is precisely at that time of philosophic system-building (what Dixit calls “The Age of Logic”) that anekāntavāda and its methodology are given their classical formulations.46 What if Johnson’s observation was generalized to most of the major doctrinal innovations taking place within the ideological landscape of the subcontinent? One would say that new ideas mainly arose from the need to protect a school’s main doctrinal asset against potential threats. Yet, that perspective would be onesided (ekānta), for the opposite is also imaginable. New ideas can also emerge as strategies to challenge contenders on their own soil. It is even imaginable that a same strategy plays on both registers at the same time, displaying both negative and positive apologetics, depending on the context.47 The anekānta-vāda seems to offer such multiplexity. What is more, it simultaneously percolates through the Jainas’ soteriological commitment. According to Barbato, anekānta-vāda is the ‘link’ which connects Jaina epistemology and soteriology.48 One could also look at it as a ‘lever.’ Indeed, the ultimate goal of the Jaina path, liberation (mokṣa), coincides with perfect knowledge (kevala-jñāna), becoming omniscient (sarvajña). In this context, anekānta-vāda

102  The beginnings of Jaina doxography simultaneously expresses both the realization of the multiplexed experience of the liberated soul and the dialectical propaedeutic designed to cultivate that same realization. As mentioned previously, the followers of the Jina translated and systematized his dialectical approach, which captures his personal insight, into a coherent propaedeutic, a therapeutic process of learning. The theoretical result was anekānta-vāda. Though post-canonical, this crown jewel of Jaina thought is coherent with the overall conceptual armature of Jainism. And, like the “emptiness of emptiness” of Nāgārjuna, it is also coherent in itself. Not only is anekānta-vāda theorizing about the multiplexity of reality, but it is also multiplex itself. For example, looked at from a historical perspective, this doctrine can be seen as serving to assert the spiritual authority and genuineness of the Jaina leaders and path, but it also appears to be inclusive. From the angle of ontology, it proposes a solution to the problem of causation by maintaining that reality is both unchanging and ever-changing.49 From the perspective of epistemology, it is a sceptical outlook on human cognition nevertheless applying complex analytical tools onto the various components involved in knowledge transmission, to allow a safer cognition to be transmitted. In the realm of eristic, it can be either positive or negative apologetics. From the point of view of conventional reality, it expresses the limitations of knowledge and the risks involved in its communication. From the ultimate perspective, it is an unambiguous all-encompassing knowledge; in other words, it is the perspective of the soul. From the point of view of the ascetic path, it is both a contemplative intellectual training and a moral discipline. The austerities endured by the Jainas on their ascetic journey have but one goal, to cleanse the soul in order to liberate it from the shackles of the karma which keeps it bound forever to the spinning wheel of existence. Jainas believe that the natural state of the human soul is omniscient. Through time, this all-encompassing sight is obscured by the many layers of accumulating karma which slowly conceal its ability to see clearly, to know immediately. Austerities are supposed to remedy that situation. In this way, a jñāna-yoga formulated along the lines of anekāntavāda, like the one presented as “pure austerity” by Haribhadra, is a remedy. It is a medicine. To be precise, one does not ‘gain’ knowledge through anekānta-vāda, but one rather removes ignorance by eradicating karma, like one recovers health by curing a disease. Hence, from the point of view of asceticism, anekānta-vāda is a cleansing device. It appears to perform a similar function as the doctrine of emptiness (śūnyatā-vāda) championed by Bhāviveka; that of an ‘eye medicine’ applied to sight (darśana), for the removal of a defect, grasping; and, especially, grasping unto views. In Jainism, the soul itself, though covered by a thick layer of karma, is said to remain eternally pure, trapped within an obscuring outer shell of substantial karma. What asceticism does – and hence the intellectual exertion involving anekānta-vāda – is to remove the karmic layers of deluded views (darśana-moha) which generated a false knowledge (mithyā-jñāna) and a consequentially perverse behaviour (mithyā-cāritra). These are the three karmic causes of the stained outer

The beginnings of Jaina doxography  103 shell obscuring the soul. In the end, the result of austerity is to recover pure sight (samyag-darśana), which reveals pure knowledge (samyag-jñāna), opening the way for the establishment of right conduct (samyag-cāritra), the last and immediate condition for emancipation.50 In the next discussion, the central place of anekānta-vāda within Haribhadra’s ṢDS will be made visible by analyzing its method and purpose, suggesting that the text is not to be considered as a mere historiographical account of darśanic philosophies, and especially not as a debate manual, but as a literary technology designed for the contemplation of anekānta-vāda taken as a meta-philosophy, the study of which may lead to samyag-darśana.

2.2  Haribhadra’s ṢDS in context While introducing Haribhadra’s ṢDS, Folkert mentioned its problematic nature, beyond the issue of its authorship, which he intended to leave aside in order to focus on the features of the text alone.51 Here as well, those textual features will be of primary concern. For, indeed, they are problematic and have led to hasty interpretations which are worth revisiting. At first sight, this concise metrical work of eighty-seven verses (kārikā-s), consisting mainly of the anuṣṭhub kind, besides three āryā upagīti (v. 24, 40, 43), two āryā (v. 61, 66), one upajāti (v. 41), and one viyoginī (v. 82), which are likely to be fragments extracted from various unidentified texts,52 appears rather modest. Nevertheless, it has been considered significant enough to be commented upon several times: most notably by Maṇibhadra’s (date unknown53) Laghuvṛtti, Guṇaratna-Sūri’s (1343–1418) Tarkarahasyadīpikā, and Somatilakaka Sūri’s (1355–1424) Laghuvṛtti, but also by later commentators.54 These scholastic commentarial works, however, appeared several centuries after Haribhadra, and besides elucidating technical vocabularies and restating the obvious in more detail, they contribute little in elucidating the dialectical strategy of the ṢDS. It should be noted that Haribhadra’s doxography served as a model for a number of later Jaina doxographical treatises, though other strategies of a more polemical facture were also adopted by Jaina doxographers.55 Balcerowicz, in an entry on doxography in his Polish collection of volumes on classical Indian philosophy, suggests that the origins of Indian doxography are related to Jainism in general and to anekānta-vāda in particular.56 Though they did not invent it, the Jainas certainly contributed to the richness of the doxographical genre and, indeed, their leading philosophical theorization, anekāntavāda, played an axial role within such compositions. But, as the present research highlights, each philosophical school which contributed to the genre adopted a dialectical doxographic pattern suiting its own philosophical tenets. Thus, a problem in Balcerowicz’s approach, which he more or less shares with Nicholson and Qvarnström, is that he discriminates between what he considers to be ‘pure’ doxographies (sensu stricto), which are supposedly ‘objective presentations’ (obiektywne zaprezentowanie), and between lesser compositions, which depict philosophers’ views for polemical purposes, like Bhāviveka’s MHK, for example.57 The problem

104  The beginnings of Jaina doxography appears to have a twofold origin. First, like many commentators, Balcerowicz’s remarks about ‘objectivity’ reproduce a common expectation of modern scholarship towards Indian doxographical schemes: that of reading them as historiographical accounts of Indian philosophy instead of as purely dialectical schemes. This perception of doxography reproduces the peculiar connotations which the genre carries in Western philosophy. For example, Jorge J. E. Gracia plainly stated what is expected of the Western doxographer: The main characteristic shared by all doxographical approaches is their emphasis on uncritical description. In contrast with the more probing and critical character of the polemical approaches . . . , doxography aims to present views and ideas in a descriptive fashion without aiming to evaluate them critically. Indeed, in keeping with its historical emphasis, doxography often discourages interpretation.58 What Indologists call “Indian doxographies,” however, should not be mistaken for their Western counterparts. For, as has been shown, ‘history,’ especially as understood by moderns, is simply not part of their sources’ concerns, just as it has actually generally not been in ancient Indian literature. To read all doxographies as somewhat historiographical and ‘objective’ is an unwarranted bias. Moreover, as discussed previously, recent philological studies on classical doxography have shown that several of these ancient texts adopted a dialectical structure and were not intended as purely descriptive accounts. The second problem with Balcerowicz’s depiction of doxography is that it fails to see that the so-called objectivity of the Jaina model (which appears only in some of them, following the lead of the ṢDS) is neither ‘neutral’ nor would it make any sense in the context of non-Jaina doxography. For example, a mere listing of competing schools of philosophy without any refutation would not suit the dialectical methodology of a Mādhyamikan. Indeed, Madhyamaka thought generally states no view of its own but expresses itself through the refutation of the views of others, leaving no view standing at the end. Thus, naturally, in the context of the Madhyamaka, a refutation is to be expected. Therein, polemics are apparent, but, more importantly, the overall dialectic is that of ‘no-view,’ the complete silencing of verbal proliferation (prapañca). As shall be seen in the next chapter, within an Advaita context, another dialectical strategy is set forth which implies a gradual process of refutation, or ‘removal’ (apavāda). However, in the context of the anekānta-vāda, as we now discuss, it is dialectically relevant to present views one by one without any stated refutation. But, even then, this does not mean that no dialectical strategy is at play. In other words, one defining doxography sensu stricto as ‘objective’ and ‘non-polemical’ is both biased towards a certain doxographical model found within the Jaina tradition and missing the dialectical point of that same model of doxography. Perhaps, to excuse this peculiar myopia, one may admit that Jaina compositions which reproduced the model of the ṢDS can be misleading. Indeed, their apparent

The beginnings of Jaina doxography  105 absence of refutation or criticism towards the many theories which they present, and their more or less accurate descriptions of the same, tend to resemble the many uncritical and largely unoriginal works on the history of philosophy manufactured by historians of ideas. Yet, the assumption is anachronistic, is culturally biased, and tends to depict such luminaries as Haribhadra in a rather pejorative light, both in terms of literary creativity and philosophical acumen. To imagine that Haribhadra, had he given himself the daunting task of ‘historically’ recording Indian philosophy as he had come to know it firsthand, like Thucydides with the Peloponnesian war, would have done so in merely eighty-seven verses, many of which directly borrowed from others, is not a particularly flattering compliment to his intellectual abilities, unless one is seriously given to minimalism. I suggest rereading the ṢDS as a brilliant dialectical product of anekānta-vāda, clearly displaying the philosophical superiority of the Jaina path. To do so, I first analyze the title and structure of the text in the light of what we know about the development of other tārkika systems. Then, I give a voice to the rhetoric of silence at play in the ṢDS before proceeding to harness the ṢDS’s classification scheme to its therapeutic function. 2.2.1 From the Ṣaṭ-Tarkī to the Ṣaḍdarśanasamuccaya: the rise of the Mīmāṃsā As an illustration of its conceptual complexity, regardless of its pithy and modest appearance, any discussion of the ṢDS does well to initiate a detailed analysis of the components of its title, in relation to the content of the text. The last word of the compound Ṣ-D-S, samuccaya, is itself composed of sam + uccaya. The preposition sam, like the Greek syn (σύν) and Latin cum (con), conveys the sense of ‘with,’ ‘together,’ ‘altogether,’ and ‘completely.’ As for uccaya, both Apte’s and MonierWilliams’s dictionaries indicate that it refers to a ‘collection,’ a ‘multitude,’ or a ‘heap.’ Together, sam + uccaya can designate whatever ‘aggregation’ or ‘collection’ but, more specifically, the word suggests a ‘totality.’ In the present context, it is appropriate to retain both its values and to understand it as an ‘extensive digest’ or a ‘comprehensive anthology’ made of six (ṣaḍ) philosophical views (darśana-s). Both the term darśana and the number six require more attention. The evolution of the diffuse notion of darśana within Jainism, until its usage in philosophical compendia, has received an extensive coverage from Folkert. He observed that Jaina doxographies, more than any other Indian sources, used the term darśana early on. Indeed, as was seen with Bhāviveka, the term was not previously part of the established Buddhist terminology to designate individual philosophical viewpoints or doctrines.59 Folkert also noticed the significant relation, in Jainism, between the term darśana and the notions of mithyātva (wrongness) and samyaktva (rightness). The link is already established in the TAS, with the definition of the mokṣa-mārga (the path to liberation) as that which “consists in the perfection of view, knowledge and conduct.”60 In this context, a common though imperfect translation of darśana is ‘faith,’ which, when understood as a set of beliefs, serves to convey the cognitive

106  The beginnings of Jaina doxography dimension of darśana, effecting behavioural responses (cāritra). More precisely, the TAS defines ‘right faith’ (or ‘right view’) as the “belief in (or longing for) the true nature of things.”61 These realities (tattva-s), of which the true nature should be known properly within Jainism, are listed as the seven fundamentals: jīva (sentient), ajīva (insentient), āsrava (inflow of karma), bandha (bondage), saṃvara (stoppage of karma), nirjarā (eradication of karma), and mokṣa (liberation).62 In brief, ‘right view’ (samyag-darśana) is the apprehension which aptly grasps the true nature of these seven realities, while the opposite, mithyā-darśana, is a mistaken ‘belief’ in their erroneous nature. Hence, the presence of tattva-s is a compulsory component in relation to darśana, for tattva-s are the marks of a darśana, what the worldview seeks to see. The term darśana never totally freed itself from this early usage, even when it came to designate any philosophical view. A second early usage of the term is also observed in canonical context. In this case, it refers to ‘perception’ rather than ‘faith.’ A relation between these two connotations of darśana is not impossible but remains ambiguous within Jaina scriptures. When it comes to compendia, Folkert insists that the term darśana is to be understood in its relation to ‘faith,’ where the notions of samyak and mithyā are involved. Prior to the TAS, samyag-darśana was especially understood in relation to the ‘perception-obscuring karma’ (darśana-varaṇa karma). The latter is one of the eight types of basic karma affecting the soul (jīva), called ‘delusive karma’ (mohanīya-karma). Within this context, samyag-darśana consists in the removal of that karma which obscures the jīva’s natural vision (darśana). Folkert argues that, with the introduction of the seven fundamentals (tattva-s) in the mokṣa-mārga, samyag-darśana is then put in a very different light, where its ties with karma are loosened.63 However, it is not impossible to imagine a doctrinal development tying these diverse theorizations, where the apprehension which aptly grasps the true nature of the seven principles would have equally been conceived as cleansing the soul from a delusive type of karma. In fact, this would be in tune with Haribhadra’s conception of jñāna-yoga as an intellectual austerity, wherein valid apprehension would lead to right conduct and liberation. It would also recall other developments in Indian philosophy, such as the path to ‘supreme felicity’ (niḥśreyasa) proposed within the Nyāyasūtra of Gautama, where knowledge of the sixteen dialectical categories (padārtha-s) is said to lead to the final goal.64 It also recalls the isolation (kaivalya) of puruṣa realized through the discriminative knowledge distinguishing the twenty-five tattva-s of the Sāṃkhya, which is obtained from the ‘practice of truth’ (tattva-abhyāsa).65 In fact, the idea that a certain knowledge of reality (tattva-jñāna) can be instrumented by a dialectical method (ānvīkṣikī) and directed towards self-realization (ātma-vidyā) appears to be at least as old as Kauṭilya,66 and to have been shared in one way or another by most of the early advocates of what came to be known as Indian ‘philosophical systems’ (darśana-s), with the only exceptions of the Mīmāṃsākas to confirm the rule,67 upon which we shall come back in a moment. Thus, like anekānta-vāda, the term darśana most likely evolved out of a Jaina environment both involved in its own doctrinal systematization and witnessing that of other dialectical schools (jñāna-mārga-s), as a convenient designation for

The beginnings of Jaina doxography  107 the discriminative faculty which they were cultivating in relation to their given objects of knowledge (tattva-jñāna). According to Folkert, within Jainism, the term slowly moved from designating ‘faith’ in general, to ‘a faith’ (a set of beliefs having darśanatva) in particular, a development which he considers to be “an extraordinary point in the Jaina view of others.”68 What has just been said concerning samuccaya and darśana should now inform the way to interpret the mysterious number six. As mentioned in the previous chapter, the number led to much speculation and appears as an aporia. For Bhāviveka, the number did not seem to play any significant function, but with Haribhadra the picture is different. The author felt the need to explicitly mention the figure in his title, and even justified himself for not strictly keeping his promise of presenting six darśana-s, since he ended up listing seven in total. Thus, in Haribhadra, one feels that the author is trying to adhere to a pre-existing template. But which one? A good example of an unsuccessful attempt to elucidate the mystery of the repeated references to the ‘six darśana-s’ in various Indian taxonomies, post Haribhadra, can be found in Prem Pahlajrai’s conference paper on the subject,69 which is itself a kind of non-exhaustive ad hoc doxographical account of the number six in Indian literature. The paper has the advantage of listing a series of well-known occurrences of the number in doxographical literature, including some exceptions, along with a limited sample of other Indian lists of six elements. But, like most writings on darśana-s and their classification, it ends up repeating verbatim what either Halbfass or Qvarnström already mentioned on the issue. To its credit, however, besides acknowledging its inconclusive nature, the paper raises the idea that Perhaps doxographers could not resist a pun, intentionally taking advantage of the phonological similarity between ṣad, “six” and sad, “true”.70 Unfortunately, the suggestion is dropped as soon as pronounced, without further development, as if revealing an Indian philosophical sense of humour was anathema. The meaning of sat, in relation to darśana, however, would go in the same way as that of samyak in the previously discussed compound samyag-darśana. Moreover, sad-darśanaṃ is indeed the first compound opening the ṢDS. Thus, the pun could provide an additional interpretative option. But this interpretation nonetheless fails to explain why the number six was so popular in darśanic taxonomies, even when it is not explicitly mentioned, as in the MHK. Before moving on to the most plausible explanation of the number six in reference to philosophical systems, it is not totally devoid of interest to follow the lead of Pahlajrai’s paper and to note the occurrence of the same number in other classificatory schemes, if only to observe some common threads. Without attention to chronology, nor claiming exhaustivity, the following list provides twenty-four known occurrences of a sixfold taxonomy within Jainism and, when relevant, within the broader Indian literature, moving from those with the closest relation to the subject of darśana to more remote ones:

108  The beginnings of Jaina doxography Occurrences of the number six in Indian philosophy and literature:   1 The six tārkika-s (to be discussed).   2 The six non-Buddhist teachers in the Sāmmaññaphalasutta of the Dīghanikāya (Buddhism).   3 The six disācara-s (ascetic philosophers) which visited Gosāla (the Ājīvika, as mentioned in the Bhagavatīsūtra) in the twenty-fourth year (4  ×  6) of his asceticism.   4 The six vedāṅga-s (interpretative sciences applied onto the Veda).   5 The six Tantra-s (particularly significant among the Tibetan Nyingma tradition and recalled in the six yoga-s of Naropa).   6 The six true faiths (samyaktva) (Jainism).   7 The six substances (Jainism).   8 The six kinds of jīva-s (Jainism).   9 The six colours (leśyā) of karma (Jainism). 10 The six saṃsthāna where a human being can be reborn (Jainism). 11 The six types of Ārya-s (Prajñā-panā-sūtra and Umāsvāti’s bhāṣya – Jainism).71 12 The six classes (abhijāti) of humans (Ājīvikism).72 13 The six osseous structures (saṃhanana) of living beings (Jainism). 14 The six processes of acquiring (paryāpti-s) vital force (Jainism). 15 The six realms of activity of the land of action (karma bhūmi) (Jainism). 16 The six inevitable life factors (anaïkkamaṇijjāïm) (Ājīvikism). 17 The six senses (pan-Indian). 18 The six directions (pan-Indian). 19 The six seasons (pan-Indian). 20 The six ascending and six descending arcs of the wheel of time forming a kalpa, each spoke of the wheel being characterized by a specific kind of human (Jainism).73 21 The six primary functions of the jīva (Jainism). 22 The six objects of taste (Jainism). 23 The six rasa-s (in Āyurveda and kāvya literature). 24 The six daily duties (Jainism). Though a mere list of patterns hardly constitutes a proof of anything, the preceding is revealing in three ways. (1) The number six is shown to be a recurrent symbolic pattern within Jainism, particularly present in classifications related to mankind. (2) The first six items reveal an ancient pan-Indian pattern of regrouping representatives of various beliefs or ‘view-holders’ into groups of six. (3) The remaining elements, more heterogeneous, nonetheless tend to express a ‘totality,’74 in terms of either time, space, types of particulars (souls, human, substance, senses), or actions. Moreover, a certain notion of ‘cyclicity’ within the whole can be evidenced by the seasons, the spinning wheel of time, the transmigrating kinds of souls (jīva-s), the habitual types of humans determined by the karmic colouration (leśyā-s) of their soul,75 the inevitable life factors, and the daily duties. One can imagine that, based on the principle of correspondence, since the Jainas’ ontology

The beginnings of Jaina doxography  109 is fundamentally divided between six eternal substances, this division somehow reverberated throughout their cosmos, generating six kinds of souls, six processes of acquiring vitality, six senses, six objects of taste, six karmic colourings (leśyā-s) of souls likely to taint at least a similar number of views (darśana-s).76 It is also probable that such sixfold listings are somehow reproducing the early sexagesimal numeral system which significantly impacted the ritual culture of the ancient world. Though this numerological exercise is highly speculative, to say the least, one can conservatively advance that the number six, in relation with the term samuccaya, is meant to express a fundamental ‘integrity,’ and, in relation to darśana, a comprehensive account of all ‘basic’ types of views one can encounter in philosophy. Within this context, it is thought-provoking to keep the idea of ‘cyclicity’ as suggestive of the notion that, at whatever period one finds oneself into, those six basic views will more or less reproduce themselves in new avatāra-s, as if they represented some recurring archetypes of the human psyche. It appears in fact that, by the time of Haribhadra, the sixfold scheme of ideological classification had already become an easy and recognized way of summarizing the diversity of views, compared to the earliest and more cumbersome Jaina account of 363 views, a topic on which Folkert extensively dwelled. Leaving aside the limited input gained from speculating on the numerological bearings of number six, one gains deeper insights into the sixfold philosophical taxonomy upon realizing that it is grounded in historical developments known to have occurred within the ideological landscape of medieval India prior to Haribhadra, leaving a lasting impact still observable in later centuries. These ‘developments’ were the focus of Gerdi Gerschheimer’s second year (2000–2001) of readings on Indian philosophico-religious systems and doxography at L’École pratique des hautes études, the most relevant research to date on the topic of the sixfold taxonomy of Indian philosophy, augmented by Gerschheimer’s 2007 article. The study shows that, at least around 900 CE, six philosophical systems, including Buddhism, Jainism, Lokāyata, Sāṃkhya, Nyāya, and Vaiśeṣika, all of which appear in the ṢDS,77 were commonly designated as the ṣaṭ-tarkī, the ‘six dialectical’ (disciplines).78 This early designation excluded the Mīmāṃsā and predated the term darśana along with the taxonomies found in later doxographical accounts. The study of Gerschheimer also investigated the notion of tarka, the etymology of which likely derived from the Indo-European root meaning to ‘twist,’ ‘tordre’ in French, ‘torquère’ in Latin, suggesting that ‘deliberation,’ ‘examination,’ or ‘reflection’ involves an act of mental and dialogical ‘churning’ or ‘spinning.’ It is significant to recall this meaning of tarka within the ascetic context of Haribhadra’s jñāna-yoga and to read it along the line of Bhāviveka’s understanding of inference (anumāna). Therein, like the mythical ocean in heaven, churned by Mount Meru, the ‘chaos’ of competing scriptures must be churned by the pillar of reasoning, for ‘the corroboration of the conventional’ (v. 9.49: saṃketasyānuvāda) to produce the nectar of immortality (amṛta). The fact that the tarkī-s appear in group attests the ambient heterodoxy which has likely set in motion the various formulations of these dialectical disciplines.

110  The beginnings of Jaina doxography In turn, since tarka deals with heterodoxy, the multiplicity of views, its methodological scope does not apply to ultimate truth per se, which is non-conceptual (nirvikalpa-jñāna), but rather focuses on breaking apart (vibhajya), as by ‘mental churning’ (tarka), the linguistic conventions laid down in scriptures, or any particular argument of a competing school. In return, tarka, being based on arguments, becomes a yoga (method) which ties linguistic conventions to logical ones. In tune with MHK 9.49, one can say that, because it is a ‘corroboration’ (anuvāda) of its inbuilt conventions, tarka facilitates the understanding (pratipatti) of a given multifaceted object of reasoning. The nature and scope of tarka are saṃsāra, the conventional world. One may wonder, who or what determines conventions in the first place? Which of the two has precedence over the other, tarka or āgama? This discursive exploration is helpful to draw our attention to the problematic and highly debated role of tarka (and, in extensio, of anumāna) in relation to scriptural means of knowledge (śabda-pramāṇa) within Indian philosophical traditions. Can tarka, in the form of anumāna, produce valid knowledge (pramā) in and by itself and thus be an independent means of knowing (pramāṇa)? If Buddhist logicians like Bhāviveka promoted the authority (pramāṇatva) of inference-based reasoning (anumāna), second in line only after direct perception (pratyakṣa), schools of thought attached to the Veda, like the Mīmāṃsā and Vedānta, had more reservations. For example, in his famous Bhāṣya, Śaṃkara condemns the kutarka (sophistry) which is not guided by śruti or āgama. Otherwise, he engages himself in such dialectical discourse while presenting his views and refuting those of others. This might help to explain why the Mīmāṃsā-related Brahmanical systems (Pūrva and Uttara) were not included in the early six tarka-s. Yet, Gerschheimer also mentions that, in Manusmṛti 12.106, a significant Brahmanical treatise from around the second to third centuries CE,79 one witnesses the possible alliance of tarka and āgama in the pursuit of dharma.80 Thus, there too, tarka came to be given a religious function. In any case, the six original tarka-s agreed in holding ‘valid knowledge,’ achieved primarily through analysis or logical probing, as crucial for spiritual liberation. The old school of Sāṃkhya, for example, attributed so much importance to dispassionate rational thinking that it came to be known as the ‘path of knowledge’ (jñāna-mārga).81 As observed in the MHK, Indian doxographies themselves are a testimony of the heated debates surrounding the use of reason, a topic particularly present in their exposition of the Mīmāṃsā, a late comer among the tārkika-s (logical disciplines). Gerschheimer has found several mentions of the six tarka-s in texts which differ both in nature and eras. The Kāvyamīmāṃsā of the tenth-century poet Rājaśekhara, for example, emulates Kauṭilya’s Arthaśāstra (AŚ) in regrouping the ṣaṭ-tarkī under the umbrella of ānvīkṣikī, glossed by Gerschheimer as a science of the self (ātma-vidyā) oriented towards liberation. That tarka-vidyā and ātma-vidyā are related is also attested in the Amarakośa (ninth century CE). The correlation of the two indicates that the six tarka-s were seen as dialectical systems directed towards self-realization.82 This might explain why the Mīmāṃsā, which had other religious motives, was originally excluded from the group. In fact, Rājaśekhara mentions the Mīmāṃsā within a wider group referred to as pramāṇa-vidyā, ‘sciences based

The beginnings of Jaina doxography  111 on valid means of knowledge,’ which Gerschheimer compares to Haribhadra’s ṢDS, wherein pramāṇa-s, though not explicitly mentioned as a decisive taxonomic criterion, are nevertheless the most systematic features of the text. Else, the compound ṣaṭ-tarkī occurs most clearly in Jayanta Bhaṭṭa’s Nyāyamañjarī (ninth century CE) and in Hemacandra’s Abhidānacintāmani (twelfth century CE), where the six earlier-mentioned schools are explicitly listed. Otherwise, it is mentioned once without any list in an eleventh-century CE inscription from Gadag (Karnataka); once in Maṅkha’s Śrīkaṇṭhacarita (thirteenth century CE); and three times in Munisundara’s Gurvāvalī (fifteenth century CE).83 Interestingly, Jayanta, contrary to Rājaśekhara, does not accept the status of pramāṇa-vidyā attributed to the Mīmāṃsā, which he rather considers a vākyārtha-vidyā, an exegetical discipline dedicated to elucidating the meaning of Vedic propositions. This shows that the status of the Mīmāṃsā school among the other dialectical disciplines was still problematic. Gerschheimer concludes his survey of sources by confirming that, at least by around 900 CE, the expressions ṣaṭ-tarkī or ṣaḍ-darśana were commonly used to refer to Buddhism, Jainism, Lokāyata, Sāṃkhya, Nyāya, and Vaiśeṣika systems as a group. Concerning the later appearance of the Mīmāṃsā among the group, he explains, Un groupement de six systèmes philosophiques a existé, bien établi dans l’usage courant vers 900 de n. è., sous l’appellation ṣaṭtarkī ou ṣaḍarśana; conformément à la théorie des 14 “sièges du savoir”, ce regroupement, motivé par l’emploi du raisonnement, de la réflexion, de l’examen rationnel des données de l’expérience (tarka s’opposant sans doute ici à āgama, la parole révélée, acceptée telle quelle), n’inclut pas la Mīmāṃsā, qui est principalement une exégèse des textes védiques (āgama, précisément). Mais il est avéré que la Mīmāṃsā a depuis longtemps développé elle aussi des thèses proprement philosophiques, et qu’elle est une partenaire obligée des débats entre les adeptes de systèmes philosophiques (les tīrthika dont parle Siddharṣi). Aussi n’est-il plus désormais possible de la laisser en dehors de ce groupement. On aboutit ainsi à une aporie: il y a six (nombre maintenant traditionnel) tarka ou darśana, mais sept systèmes à prendre en considération!84 In other words, Gerschheimer suggests that, though it differs in nature and purpose from the other six tārkika-s, which favour reason (tarka) over scriptures (āgama), the Mīmāṃsā eventually imposed itself as an unavoidable adversary in debate and came to be integrated as a troublesome addition to the group, a scenario which explains both the ‘six plus one’ list of Haribhadra and the revulsion felt by Bhāviveka towards the ritual-oriented Brahmanical system, for which he spared no effort in showing ridicule. The socioreligious developments taking place on the subcontinent, from the sixth century onwards, mentioned in Chapter 1, corroborate Gerschheimer’s hypothesis while suggesting an earlier date for the emergence of the Mīmāṃsā as an inevitable dialogical partner of the six tārkika-s. With the political agitations of militant Brahmanism acting as a social impetus, imposing its ideas at royal courts,

112  The beginnings of Jaina doxography not the least through debates, the ritual-injunction-oriented speculations born of Vedic exegesis progressively forced their way to the pantheon of Indian philosophies, otherwise wary of the bounds of karma and of metaphysical claims solely supported by scriptural authority. Gerschheimer reasons that the intrusion of the Mīmāṃsā within the six tarka-s or darśana-s set in motion the evolution process of the categorization of the philosophical systems known today.85 To their credit, early Indian doxographies, though not historiographical per se, bear witness to the philosophical disturbances created by the ritualist newcomer, each in their own way. Haribhadra suggested two solutions to the problem created by the new Brahmanical input. They were to inspire similar classifications within the later doxographical tradition. At verses 78–79, after having described the sixth darśana, the Jaiminīya-s (Mīmāṃsā), and before introducing the Lokāyatas as the seventh system, Haribhadra explains, Others do not recognise the rupture from the doctrine of the Naiyāyikas by the Vaiśeṣikas. In their opinion there are only five faithful (āstika) disputants. Thus, the number of six views, according to them, is completed from adding the doctrine of the Lokāyatas. Therefore, that opinion will be described.86 What this means is that, in order to keep a sixfold taxonomy, Haribhadra had two choices. He could either exclude the materialist Lokāyatas, or he could join together the sister schools of Nyāya and Vaiśeṣika. Both ways allowed him to gain an additional entry and to include the Mīmāṃsā within a sixfold scheme. The idea of coupling sister schools, as is well known, developed further in later doxographies, particularly among the promoters of Vedic views, bringing together Saṃkhyā and Yoga, Pūrva and Uttara Mīmāṃsā, which, along with the NyāyaVaiśeṣika couple, eventually formed the sixfold taxonomy of Brahmanical doxographies, often taking the name of āstika systems, in contrast with the nāstika-s which were accused of having no faith in the Veda. Nāstika views also eventually came to be grouped in a sixfold scheme, with four schools of Buddhism coming along the Cārvāka and Jaina opinions. This sectarian twist of the expression āstika had not already been adopted by Brahmanism at the time of Haribhadra, as can be witnessed in the previous passages.87 The study of Gerschheimer, without solving all questions, has shed a muchneeded light on the historical development of the sixfold categorization of Indian philosophical views. Yet, beyond the traditional grouping of six tarka-s, the fact that authors like Haribhadra were so concerned with keeping the same number hints at its additional symbolic significance, perhaps related to some fundamental divisions of mankind related to cosmology and ontology. Unfortunately, as done previously, one can only conjecture on the topic. Concerning the significant role played by the Mīmāṃsā in the development of doxographical classifications, it motivated the textual selection of the present study, using the Mīmāṃsā segment of each doxographical text to illustrate their content. Now, concerning the ṢDS,

The beginnings of Jaina doxography  113 another misunderstood feature of its dialectical methodology must be explored: the rhetoric of silence. 2.2.2  Syāt Lux: the rhetoric of silence88 Besides its sixfold taxonomy, a noticeable feature of the ṢDS is its apparent absence of refutation, of the kind of uttarapakṣa found in Bhāviveka’s MHK. Each of the seven systems listed within the ṢDS is presented according to a given set of criteria,89 but no explicit criticism of it is formulated. This peculiar nonargumentative literary style appearing within a known socioreligious context of philosophical polemics and debates has led to all kinds of appreciative remarks about the ‘inclusive,’ ‘tolerant,’90 and ‘objective’ spirit of the author and of those who emulated his model; comments which presuppose an unreflective and purely descriptive function of such writings.91 Others, as mentioned already, even saw in the phenomenon the beginnings of some ‘history of philosophy.’ Fittingly, Folkert ironized, The tendency to see the compendia in the light of what one is seeking is perhaps the single outstanding characteristic of most studies of them. Visions of them as a whole, as a problem unto themselves, are rare. This has left a great many loopholes in the current view of these compositions.92 The state of research on Indian doxography has made little progress since Folkert’s work and, though desirable, no general account of the genre, comparable to what has been achieved with Greco-Roman sources in classical studies, is even remotely available. The present study can merely contribute another pebble throw in the elusive river. Somewhat in tune with previous scholarship, the present chapter will discuss Haribhadra’s doxography in the light of the doctrine of anekānta-vāda. However, contrary to previous opinions, I will argue that precisely because it is a typical manifestation of that worldview, just as the MHK’s strategy reflected the general attitude of Madhyamaka Buddhism towards philosophical opponents, the dialectical strategy of the ṢDS, involving a rhetorical silence, has little to do with ‘inclusivism,’ ‘tolerance,’ or some kind of ‘historical objectivity.’ It is rather an ingenious and coherent rescript of the method, efficiency, and superiority of the Jaina path in the face of heterodoxy. In the beginning, the first glance of the neophyte upon the ṢDS is stricken by the deep argumentative silence hovering over the chaotic waters of randomly disposed philosophical views. The critical mind of the interpreter, seeking but perceiving no clear λόγος, confined to disturbing darkness, can either dismiss such nonsensical literary creation as useless or comfort itself with a moralist reading. The latter, at least, has the advantage of safekeeping creation, however ambiguous, and to elevate its author to a status gratifying the views of the exegete. Behold, the tolerant, the merciful! But, if the ways of the creator are mysterious, light shines through his creation. Hence, it is within the words (λόγοι) of Haribhadra that one

114  The beginnings of Jaina doxography should find the meaning of his silence, and thus, acting like the δημιουργός, shall I separate light from darkness. Just as the author carefully chose his words and mode of speech while composing the ṢDS, one is led to believe that he intentionally decided to remain silent when he did. The intentional silence of a speaker is a rhetorical one. In his entry on ‘rhetorical silence,’ within the Encyclopedia of Rhetoric and Composition, Barry Brummett explained, Most rhetoric takes the form of words or gestures with meaning. But silence can also be rhetorical. Not all silence is rhetorical. Silence functions rhetorically when it is used strategically (rather than happens randomly) in political (rather than interpersonal) contexts. Rhetorical silence has four characteristics: It violates expectations, it is intentional, it generates predictable meanings, and it influences the meanings of its context more than context influences its meanings.93 In the same way, I argue that the argumentative silence of the ṢDS violates the expectations of its dialogical context, where similar philosophical compositions, like the MHK, extensively involved themselves in refutation. This is precisely why commentators advanced various theories on the subject. Hence, Haribhadra’s silence is intentional. As I shall defend, the meaning of his silence is predictable, based on a definite set of factors. In the end, what Haribhadra does not say influences the content of the ṢDS more than the other way around. For, it sets the ultimate perspective from which all the listed competing philosophical views are to be contemplated. Let’s begin with what Haribhadra did say. Basically, in order, the ṢDS lists seven systems: (1) Bauddha; (2) Nyāya; (3) Sāṃkhya; (4) Jaina; (5) Vaiśeṣika; (6) Mīmāṃsā; and (7) Lokāyata. There are some, like Halbfass, who see this taxonomy as a proof that Jaina doxographies do not follow any explicit hierarchical schema.94 To put it plainly, against anything in the κόσμος, such doxographies would be anarchic. In contrast, we have seen how the MHK proceeded from an introductory presentation of the author’s own view, if only through a dialectical breakdown of the Abhidharma, and refuted all following views, moving from the less problematic ones (the Buddhist views) to the most troublesome (the Mīmāṃsā). On the other hand, as shall be seen in Chapter 3, Advaita types of doxography tend to do the opposite, by presenting their own position in conclusion, proceeding gradually from the most problematic systems to the better ones, leaving the Advaita dominating at the very end. Now, if these doxographical nomenclatures follow a precise dialectical teleology in tune with their worldview and their overall dialectical method, why should it be different for Jaina doxographies? Evidence suggests that it is not. Let’s look at the ṢDS again. The author’s position is listed as number four, which stands exactly between the first three and last three ones. It is madhyastha. Are we to believe that it is a mere coincidence if the Jaina position appears midway? I do not think so. As discussed already, in dialogical context, Jainism adopted

The beginnings of Jaina doxography  115 the dialectical attitude of its founder, who sought to establish a middle way, somewhat like the Buddha, but differently. I mentioned how Matilal had remarked that the way of the Buddha was rather based upon rejection and exclusion, whereas the one of Mahāvīra involved acceptance and inclusion.95 This apparent ‘inclusive’ middle way appears to be the exact model reproduced here in the ṢDS. Occupying the middle position between all philosophical systems, for a Jaina, is not a random occurrence but the best situation imaginable. In other words, just like our Madhyamaka and Advaita authors within this study, Haribhadra gave his position the best-suiting location, matching both his worldview and dialectical strategy. The scenario is thus not one of so-called inclusivism or tolerance but rather one of dialectic and rhetoric. If one is not yet convinced that Haribhadra strategically situated his position in the most favourable light, one only needs to read how the author introduced the Jaina position and compare this with how he presented the other views. Opening the discussion of the Jaina view, verses 45–46 read as follows: Therein, the Lord Jina is the master. He, who has abandoned craving and aversion, is the great wrestler who destroyed ignorance, whose vision and knowledge is the highest. He is to be honoured by the lords of both demons and gods, the teacher of things as they truly are, he who, upon having accomplished the destruction of all karma, has obtained the supreme station.96 This eulogy of the Jina is the only praise introducing any traditional authority (devatā)97 within the text. Therein, the view (darśana) and knowledge (jñāna) of the Jina are explicitly mentioned as the highest, just as his final abode. In other words, Haribhadra is not exactly impartial. He gave the best position to his own school, which stands like Mount Meru at the centre of world philosophy, and he attributed the highest beatitude to his ‘ford-maker’ (tīrthaṅkara) alone. Yet, one might insist that, nonetheless, since there is no explicit refutation, the ṢDS depicts the Jaina position on an equal footing with all the others and is thus ‘inclusive’ and ‘tolerant.’ This is where it becomes useful to analyze the rhetorical silence of the author. Another hint as to its interpretation is given in the opening benedictive (maṅgala) verse of the text: Having paid respect to the Jina, holder of the true view, the virile instructor of the syād-vāda (method), the meaning expressed by all worldviews will be briefly proclaimed.98 With this scant but dense formula alone, Haribhadra stated his intention in writing the ṢDS. First, let’s observe that the introductory verse opens with sad-darśanaṃ, which cannot fail to recall the title of the text ṣaḍ-darśana. Indeed, there seems to be a pun here.99 But what does it mean? Although there are six other worldviews, the only true (sat) one is that of the Jina. Perhaps Haribhadra really felt as if he dealt only with six views in the end, which could explain his title (ṣaḍ). If he took his position as being the sole truth (sat), or samyak, the other six, taken

116  The beginnings of Jaina doxography independently, would altogether be erroneous (mithyātva), while his would stand alone in the middle as the exact opposite, not as a mere view, but as truth itself, in all its shades. In fact, the dialectic is subtler than this, as we have seen already while discussing anekānta-vāda. The ṢDS does not merely exclude, or include, but it uses the other positions as necessary segments of an overarching truth. In this way, standing in the middle of all views, Haribhadra’s position acts as the axis mundi of world philosophy, the very pillar of tarka, subsuming all systems in a totality transcending their limitedness which, if left alone, would not only be false (mithyā) but poisonous. At the same time, Haribhadra justifies the very raison d’être of these darśana-s within the domain of philosophy, as various possible steps within an ‘omni-movement’ of the mind towards omniscience. The axis of a wheel is related to every of its segment, but it is not the same. Its central position gives it primacy of movement. It defines the whole. It is essential and not merely possible. If a pragmatic or somewhat anti-intellectual mind, ever sceptical towards philosophical speculations, was to ask why to spend the effort of studying every segment of the philosophical wheel if the middle alone is enough to know the truth, the metaphor of the wheel, in relation to anekānta-vāda, could provide a convincing answer. It is because for a wheel to spin correctly, in other words for a mind to be in tuned with ‘reality,’ all the sections of the wheel have to stand in their right places. Philosophical studies, the jñāna-yoga of Haribhadra, which the ṢDS serves to support, is a method designed to fix the wheel of life, once and for all, to the chariot of liberation. The axis of the wheel, in Sanskrit, is called kha. When the wheel is not well balanced, it is duṣ-kha, which together comes to mean suffering (duḥkha). However, a well-balanced wheel is su-kha, which translates as ‘happiness’ (sukha). For lasting happiness to happen, one must thread the path (mārga) on a well-harnessed vehicle (yāna) bound by the yoke of reason (yukti). Seen in this light, the Middle Way is more than a dialectical strategy, it is the religious path itself, and philosophical studies its quintessence. This is jñāna-yoga. The purpose of the ṢDS, as stated in the introductory verse, is to proclaim (nigadyate), in an abbreviated manner (saṃkṣepeṇa), the ‘meaning’ expressed (vācyaḥ) by all view (sarva-darśana). It is as if this ‘meaning’ was unknown by those in question, the followers of the mentioned systems, or by anyone else, perhaps by the neophytes of religious life. The idea appears bogus. Another possible double entendre within the same introductory verse suggests that the purpose of the ṢDS is not simply to ‘summarize’ the meaning of systems, but to proclaim, in a nutshell, their true utility, their true place within the world (κόσμος) enlightened by the only ‘teacher of things (artha-s) as they truly are’ (v. 46a sadbhūtārthopadeśaka), the Jina. And how is Haribhadra to proceed in doing so? How could he, in merely eighty-seven verses, fix all philosophical systems? Simply by following in the footsteps of the ‘virile instructor of the syād-vāda’ method (vīraṃ syādvādadeśakam), which he praised in his introductory verse. We have seen already that the quodammodo strategy (syād-vāda) is concerned with the object of the communication itself. It aims to provide an accurate

The beginnings of Jaina doxography  117 description of it through a range of formalized sentences which could cover its overall complexity. This strategy is also sometimes called sapta-bhaṅgī (the ‘seven-figured’ model), in reference to the seven basic formal standards, which are said to encapsulate an infinite number of sentences one can say about any entity. In other words, it systematizes into seven possibilities the entire range of possible assertions about reality. Is it imaginable that, in listing seven views, Haribhadra was again playing with numbers? Could it be that, within the ṢDS, he was offering a basic philosophical template which, although it does not exactly match the seven logical divisions of the sapta-bhaṅgī, nevertheless mirrors them through the reflection of seven actual philosophical positions dividing the world of philosophy known in the days? Thus, the ṢDS would be an attempt to schematize the entire range of possible darśana-s, offering a template much easier to handle than the canonical model of 363 views. But, what is more important, in doing so, Haribhadra would have had no need for any refutation. For, the only necessary refutation, from the syād-vāda perspective, is inbuilt within the divisions of the sapta-bhaṅgī. We have seen previously that syād-vāda consists in converting every naturally ambiguous proposition, thus every linguistic statement, into an unequivocal enunciation, by adding the term syāt (‘somehow,’ ‘in a certain sense,’ ‘from a certain perspective’) at the end of it. Just as for the sapta-bhaṅgī, from the anekāntavāda perspective there is nothing wrong per se in the seven darśana-s. They are all different perspectives. The problem only comes when they are reified, turned into absolute positions. Then only they become wrong (mithyā). Thus, Haribhadra would have only needed to voice a refutation if the views which he listed had been said to be ultimate. In this case, he would have needed to say but one thing alone: syāt. This is the only predictable meaning of Haribhadra’s silence: syāt. For Jainism, syāt is the light of philosophy, the all-powerful verb of the Jina hovering over the chaotic ocean of heterodoxy, the thundering sound of the conch shell which quells the quarrels of sophistry. Now that we can predict the meaning of the rhetorical silence of Haribhadra, it remains to be seen how it influences the content of the ṢDS more than the other way around. I suggest that syāt is silent in the ṢDS because the text seeks to induce the perspective of the absolute (niścaya- naya) in the reader through a careful depiction of conventions representing various relative perspectives (vyavahāranaya) on reality which are actually in vogue in the world. Syāt, in itself, is the conventional expression of omniscience. One must arrive at it through the channels of worldly conventions. In the case of the ṢDS, syāt is silent because one must learn to pronounce it by oneself while studying the many segmented views (darśana-s) of philosophy. It is a silence waiting to be uttered. Rhetorical silence is for the auditor to fill, explained Brummett: Rhetorical silence is theoretically interesting because it is centred on the audience rather than the rhetor. Although silence may certainly be used as a strategy, the rhetor only initiates it. What it means, and how those meanings influence perceptions of the particular context within which silence occurs, is

118  The beginnings of Jaina doxography determined by a process of attribution and meaning construction carried on by the audience itself.100 When one finally hears syāt after every stated position, one has elevated oneself to a worldly level of omniscience, a necessary step towards the dawn of absolute omniscience. Let’s now put all of this in context. The distinction between two levels of realities was developed within Jainism by Kundakunda, who lived somewhere between the second to the fourth century CE,101 particularly in his Samayasāra, which has been acclaimed as the most profound exposition of the Jaina doctrine,102 but also in the Dravyasaṅgraha. His epistemology is said to have influenced Śaṅkara’s distinction between vyāvahārika and pāramārthika, as well as the corresponding divisions in Buddhism, pivotal in Bhāviveka’s work.103 According to Kundakunda, omniscience can be looked upon from two perspectives. From a relative point of view, the kevalin, one who has attained omniscience, can be said to know all things. In this way, he would simultaneously see the many aspects of reality, a sight which ordinary human language can only hope to approximately render with unequivocal enunciations through syād-vāda. From the ultimate perspective, however, the kevalin does not see a plurality of aspects of reality. At this level, there is no distinction between darśana, jñāna, cāritra, and ātman. Omniscience is then organ-less, it is immediate knowledge. It is the perception of the soul cognizing itself. Nothing else but the soul remains. The knower knows no distinction. Knowing the soul, as discussed already, one knows all. But this knowledge, being immediate, cannot be mediated through human words. It must thus remain silent. If it was to speak, since it cannot deceive, being samyak-cāritra, it can only speak in syād-vāda, careful of the limits of conventions. Seen that way, one realizes that, by training in syād-vāda, one actually trains in being sarvajña (all-knowing). Dialectic is the path. The ṢDS offers exactly that kind of propaedeutics. It is not a mere introduction to the ‘history of philosophy’ or a ridiculously helpless debate manual, but a training into truth (tattva-abhyāsa), a jñāna-yoga fully coherent with the Jaina path. It uses heterodoxy to establish orthodoxy and orthopraxy, turning poison into medicine. But let no one be fooled. Though silent, the ṢDS is extremely critical. Syād-vāda implies that, left alone by themselves in the world, the various darśana-s which dispute their sole authority are poisonous. Their bias, like any strong attachment, vitiates the mind and pollutes the soul with vision-obscuring (darśana-moha) karma-s. Only syāt, like a magical formula, can turn water into wine. And only the Jaina path teaches this most desirable of formula (yukti). Both from a dialectical and rhetorical perspective, the ṢDS is a masterpiece of subtlety. Appearing completely inoffensive, it subjugates all its opponents. With a seemingly ‘anarchical’ random structure, it positions itself as the very axis of world philosophy. Thought to be ‘unreflective,’ it ends up answering every philosophical quarrel. It is meaningful to reflect upon the ṢDS as a metaphor for the socioreligious position of Jainism, at least at the time of Haribhadra.

The beginnings of Jaina doxography  119 Here again, the concept of rhetorical silence is eloquent. Brummett observes, Talk is a way of defining relationships in politics; rhetorical silence therefore means mystery and uncertainty among people, in place of relationship. Talk is a form of action in politics; rhetorical silence therefore means passivity and relinquishment of initiative in politics.104 It is to be remembered that the way of the mauna-s, of those who took the great vow of silence (muni-vrata), the muni-s, as Jain monks like Haribhadra are called, is ultimately the relinquishment of all action (karma) whatsoever, well beyond mere political activism. That Jain religious life can culminate in sallekhanā (lit: to thin out), abandoning all food intake and all activity until death ensues, is perhaps the most extreme expression of this principle. If public debates are seen as political activity, one can imagine that at least a certain portion of Jain monkhood had reservations towards the practice and preferred a silent truth to a boastful ignorance. The legend of Haribhadra himself depicts the sufferings caused by such worldly disputes. Is the ṢDS the expression of a cry for peaceful coexistence, safeguarding the authority and coherence of one’s own tradition while seemingly laying down the weapons and presenting no threat to anyone? Is it a Trojan horse of conquest? Or, is it the silent victory cry of the recluse, who found peace, if only in isolation? It is striking that, though the ṢDS seems to inclusively engage with other views, it does nothing of the sort. It is in fact a monologue. Therein, views are austerely listed and left for themselves, isolated in their limited perspective, with not a single word of praise or blame, besides for the Jina, who stands upright in the midst of all, pure as white marble, motionless. From this perspective, the ṢDS is the exact opposite of the MHK, which amused itself in a dialogical atmosphere, by times quasi-carnivalesque,105 if only to laugh out in irony. This might serve to reflect differently on the observation of Matilal concerning the Buddha and the Jina. As mentioned, he suggests that, unlike the Buddha, who tended to withdraw from philosophical disputes by avoiding any commitment to extreme positions, the Jina rather espoused commitment. With the respective doxographies coming out of their lineages, we now witness the exact opposite scenario. Dialogue, in the ṢDS, can only begin when and if the silent syāt is uttered. Did Haribhadra somehow reproduce the famous silence (tuṣṇīm-bhāva) of the Buddha, also a muni, kept when the latter was confronted with the fourteen undetermined questions (avyākṛta-vastūni), the disquieting rhetoric of which influenced Madhyamaka and Yogācāra up to Ch’an Buddhism?106 This seems unlikely. Haribhadra’s silence is not paradoxical. Its meaning, if not explicit, is predictable. Moreover, Haribhadra did say a few words by presenting the tattva-s and devatā of Jainism along those of others. But these words engaged in no way with the other systems. They gave no directions as to what to do with the knowledge provided by the ṢDS on each school. What is missing, that which remains silent, is the expression of the higher omni-vision able to perceive each darśana in its true form, for what it is worth, as part of a cohesive unequivocal discourse on reality. This

120  The beginnings of Jaina doxography perspective, since silent, has to be intuited by the reader. If the silence of the ṢDS has any mythological model whatsoever, it comes closer to the way a Tīrthaṅkara is said to directly transmit his omniscient knowledge to his chief mendicant disciples (gaṇadhara-s), as if by telepathy, without uttering a word. Though the master does not speak, he is able to directly transfer his insight to those who have the spiritual capacities to receive it. Reading the argumentative silence of Haribhadra within this mythological context supports the idea that the correct understanding of the ṢDS is meant to be acquired through introspection, and thus imparts a contemplative function to the text. The hypothesis is not as farfetched as it appears. In their study on the role of silence within the main religious traditions of India, Nemi C. Jain and Anuradha Matukumalli have discussed the function of silence in Hindu jñāna-yoga, observing similar connotations in Jainism: We suggest that the functions of silence at the individual level can best be understood in terms of the Hindu concept of yoga . . . Yogas are of many kinds. Silence is also an integral part of jnana yoga which is the path of right knowledge. . . . Silence facilitates the contemplation and introspection which enables the person to distinguish between the real and unreal until he or she realizes “the one nameless, formless, indescribable, absolute Being of the universe, where the subject and the object, the knower, knowledge and its object losing their relativity merge into the ocean of the absolute Brahman” (Abhedananda 1936: p. 60). Like Hinduism and Buddhism, Jainism also recognizes the value of silence in attaining the higher aims of life. . . . Introspection is purifying.107 In Jainism, the vision of the absolute is not expressed in terms of brahman, but of kevala-jñāna or sarvajña, being all-knowing. Nevertheless, the relation of silence to contemplation is the same. It facilitates the introspection required for one to distinguish between the real (sat)108 and the unreal (asat). As discussed already, the process by which the vision (darśana) of the soul is made to perceive tattva-s in their true light, producing samyag-jñāna and the correlated samyag-cāritra, is portrayed as a purifying asceticism, which Haribhadra referred to as jñāna-yoga. It is in this way that I am inclined to read the ṢDS and similar doxographies as literary technologies designed as further soteriological devices, or spiritual exercises, supporting a jñāna-yoga engaged in purifying views. A short note on the ṢDS’s classification criteria and the way they harness conventional views to operate an ultimate perspective now remains to be provided. 2.2.3 Harnessing conventions: yoking Devatā and Tattva to the absolute The criteria selected by Haribhadra to distinguish and classify the various darśana-s have already been abundantly discussed by scholars.109 There is little to add to the subject, beyond a comprehensive summary of what was said already, which is nevertheless necessary in order to understand how the text proceeds in its presentation

The beginnings of Jaina doxography  121 of views. What the present exposition intends to contribute, however, in continuity with the previous discussion on the MHK, is a reflection on the conventional nature of the views listed in Jaina doxographies and their relation to the absolute. In the second verse of the ṢDS, Haribhadra explains, There are only six views, depending on their fundamental distinctions, divided in terms of authority and realities, to be known by learned ones.110 The six views are, in order, Bauddha, Nyāya, Sāṃkhya, Jaina, Vaiśeṣika, and Jaiminīya (Mīmāṃsā), with the Lokāyata in addenda. Now, Haribhadra suggests that learned people distinguish these ‘six’ according to their fundamental distinctions (mūlabhedavyapekṣayā), which are expressed in terms of authority (devatā) and realities (tattva-s). The term devatā, most commonly translated as ‘deity’ or ‘divinity,’ can also be understood in terms of ‘lordship’ or ‘mastery’ (aiśvarya). The commentator Maṇibhadra relates devatā to the quality of being deva, ‘shiny’ or ‘glorious.’ Folkert has shown the relation of devatā to the notions of āpta (authority), āptatva (reliability), āgama (scriptures), and guru (important) in various Jaina treatise. He also indicated that a number of discussions on samyaktva seem to refer to the Jina as devatā. Hence, in the present context, I chose the translation of devatā as ‘authority’ in order to refer to it as a general category of ‘knowledge standard,’ able to subsume an actual divinity, but also authoritative scriptures or human teachers. For, it is how Haribhadra uses the term, depending on each darśana. For example, the devatā of the Buddhists is the Sugata (v. 4); Śiva stands as the elected divinity of the Nyāya system (v. 13); the Sāṃkhya school is distributed between those who accept Īśvara (the Lord) and those who don’t (v. 34); Jina is the devatā of the Jainas (v. 45–46); the Vaiśeṣika school has the same divinity as the Nyāya, but different tattva-s (v. 59); the Jaiminīyas have no other devatā than the Veda (v. 68–69); while the Lokāyata have simply no external devatā at all (v. 80).111 Folkert has also shown that the notion of devatā is intimately related to samyaktva, as found in the important threefold parts of the path to liberation (mokṣa-mārga) formulated in the TAS, beginning with samyag-darśana. Ultimately, what matters is for the seeker to distinguish which of the devatā-s is most reliable. In order to do so, one must examine their moral qualities, concerning ‘non-violence’ (ahiṃsā) for example, and all of their promulgated tattva-s. Tattva-s are the second fundamental distinction (mūlabheda) between the darśana-s. More than mere intellectual categories, they correspond to the fundamental realities which every system designates as the core of its teachings, the doctrinal elements which guide contemplative practice. The discriminative knowledge (viveka-jñāna) born out of meditating on the tattva-s (or padārtha-s) is an active ingredient essential to the ‘cognitive therapy’ set forth by each darśana. For example, again, the Buddhists are said by Haribhadra to have the four noble truths, the twelve āyatana-s, and two pramāṇa-s, as tattva-s (v. 5–12); the Nyāya systems refers to sixteen padārtha-s (v. 14–32); the Sāṃkhya distinguishes between twenty-five tattva-s (v. 35–43); the Jaina system has nine tattva-s (v. 47–58); the

122  The beginnings of Jaina doxography Vaiśeṣika has six padārtha-s (v. 60–67); if any, the Mīmāṃsā has only its dharma and six pramāṇa-s (v. 70–79); while the Lokāyata refers to the four elements as tattva-s and to pratyakṣa-pramāṇa alone (v. 83–87).112 Pramāṇa-s can be said to be tattva-s, or at least to be related to them, in the sense that, like direct perception or inference, they act as means to acquire knowledge and to determine its validity. Hence, like the tattva-s, they are discriminative tools. Haribhadra’s distinct darśana-marks are thoroughly systematic. The two classifying criteria of devatā and tattva explicitly describe a darśana as a belief system centred on a sapiential authority which instructs a given set of fundamental theories,113 altogether forming a ‘gnoseological’ prerequisite to salvation. It should be observed that, like in the MHK and most of the later Indian doxographies, there is strictly no further discussion about discipline, rituals, dress code, and so on, whatever accidents114 a modern-day encyclopaedia of religions would be eagerly engaged in discussing. The only concern of the ṢDS is knowledge (jñāna) and its source (devatā) and content (tattva). Hence, one can safely conclude that, for its author, a darśana is, at the very least, a belief system or, better, a theōría, a speculative contemplation, and hence a proponent or another of the therapeutic path of knowledge (jñāna-mārga), with the Mīmāṃsā in an ambiguous position. The notion of tarka, as found within the six tārkika-s, in relation to ānvīkṣikī, seems to have conveyed exactly that meaning. For, even the ‘materialist’ Lokāyata-s, enjoining self-reliance and critical thinking in terms of knowledge sources (devatā), had their basic principles (tattva-s) upon which they sought to free the human mind of its unnecessary torments (false beliefs and so on). But, with the introduction of the Mīmāṃsā, the ‘path of action’ (karma-mārga) par excellence, there is a slight shift, or one could say an ‘opening up,’ of the darśana-s. Yet, even the Mīmāṃsā, especially but not only with the later developments of the Uttara-Mīmāṃsā (Vedānta), has undergone a set of doctrinal changes which indicates an intentional and systematic adaptation to the dialectical ways of the path of knowledge, most likely because the latter retained its prestige among the Indian socioreligious elites over the centuries. Hence, one could say that the influence went both ways. The Mīmāṃsā adapted to the other darśana-s, and, in response, they adapted to the Mīmāṃsā, engaging it in the unforgiving fray of debate and, eventually, in doxography. Folkert sought to clarify the relation of Haribhadra’s darśana-marks to samyaktva and its key role within the mokṣa-mārga. He mentioned a recurring pattern coming from as early as Samantabadhra (around 450 CE) and still noticed as late as Vasunandin (around 1100 CE), of defining samyaktva as faith in āpta, āgama, and padārtha (or tattva), a triad also recalled in Hemacandra’s list of deva, guru, and dharma.115 He also noticed that a discussion found in Haribhadra’s Lokatattvanirṇaya indicates that a devatā needs to be unsullied in order for a darśana and its instructed tattva-s to be acceptable. The given criteria of a proper devatā, and the criticism of improper ones, are generally in tune with the condemnation of the gods and so on found within the MHK and in later Jaina doxographies, particularly in Bhāvasena’s Viśvatattvaprakāśa (fourteenth century CE). Thus, one should not assume that, because Haribhadra listed several devatā-s

The beginnings of Jaina doxography  123 without criticism, he saw them all as equal. Their only similarity lies in being perceived, by some, as sources of knowledge. But there is more to the relation between the two darśana-marks of Haribhadra and the threefold division of samyaktva found in the mokṣa-mārga of the TAS than what Folkert needed to mention in his demonstration. The threefold division of samyaktva has close resonances with the theoretical framework adopted by Bhāviveka to outline the pivotal role of reasoning and hermeneutics in Mahāyāna religious praxis, the three levels of wisdom (prajñā) which he borrowed from the Yogācārabhūmi. Therein, we find the wisdoms gained from hearing (śrutamayī), from critical analysis (cintā-mayī-prajñā), and from contemplative insight (bhāvana-mayī-prajñā). The first one is a wisdom gained from hearing an authority (āpta/āgama/devatā), and the second one is a critical analysis of the content of such hearing, precisely of its tattva-s. The subtlest of the three wisdoms, contemplative insight, said to be non-conceptual, is the outcome of the two earlier ones. Descriptions of its non-conceptual content naturally varies from one tradition to another. But it is possible to approximately equate these various appellations. For example, the bhāvana-mayī-prajñā of the Buddhists, acting as the last step into liberating truth, could be related to the Jaina notion of pure conduct (samyakcāritra), the culmination of dharma. Both involve a non-conceptual relation to reality having emancipative behavioural consequences made possible through a valid knowledge promulgated by an equally valid source. Thus, the two darśanamarks of Haribhadra, devatā and tattva, could correspond to the two preliminary levels of conceptual wisdom necessary to reach a third non-conceptual and emancipative wisdom, the realization of the ‘Tattva of tattva-s,’ to borrow from Biblical formulations (Sanctum sanctorum). In an earlier paper, I argued that Haribhadra referred to an identical threefold process of insight in his Yogadṛṣṭisamuccaya (v. 101), when he claimed, Through scriptures, inferences, and the essence of yoga practice, they succeed at the threefold wisdom and obtain the highest reality (tattva).116 Accordingly, to hear the scriptures (āgama), to reflect upon their tattva-s through inference (anumāna), and finally to engage in yogic contemplation (yogaabhyāsa) are the threefold components of the well-balanced therapy called yoga, what could be referred to as jñāna-yoga.117 This threefold division of the path appears to paraphrase the mokṣa-mārga of the TAS. In the same article, I also highlighted a similar pattern found in Śaṅkara’s Upadeśasāhasrī, within a verse of deep Buddhist resonance (US: 19.25), stressing the significance of scriptural studies and reasoning in the soteriological process leading to self-discovery according to the Advaita: Thus, through reasoning and scriptures, the awakened one, seeing [the Self] as homogeneous and hence as forever luminous, free from the duality that fancies existence and non-existence, undergoes complete extinction (nirvāṇa), like a lamp.118

124  The beginnings of Jaina doxography These examples suggest that the three thinkers envisioned their path as resting firmly upon hermeneutic praxis, making use of both authoritative scriptures (āpta/āgama/devatā) and discriminative knowledge (viveka-jñāna) as means to support contemplative life and insights into truth (tattva). As for the ṢDS, it lists the first two components of the path to liberation (mokṣa-mārga) – if samyag-darśana is related to devatā (because ‘right view’ is defined by the given ‘authority’) and samyag-jñāna to tattva (for ‘perfect knowledge’ is the knowledge of ‘reality’) – found within each darśana, leaving out the non-conceptual ultimate perspective. The previous analysis of the dialectical teleology of the ṢDS, along with its argumentative silence, has shown that the ultimate insight which Haribhadra was interested in communicating through the ṢDS, which was left silent in order to stimulate introspection, was the one born of syād-vāda. From a Jaina perspective, this is the epistemic insight which leads to samyak-jñāna and -cāritra. The fact that doxographies involve the tattva-s of opposing views in their overall systematic contemplation (θεωρία), which is dialectically designed to establish the ultimate perspective (niścaya naya/paramārtha), though innovative,119 follows an ancient ubiquitous pattern of Indian philosophical speculation. The pursuit of philosophy in India is no mere intellectual activity. It has a practical purpose, which I define as therapeutic. Knowledge, in this practical context, primarily serves as a discriminative tool to discern between what is conducive to the desired goal and what is not, between what leads to sukha (pleasure) and what leads to duḥkha (pain); it discerns between dharma (righteous) and adharma (unrighteous), sat (true) and asat (false), understood as rooted in concrete experiential events. A typical example of this practical-mindedness can be found in the Sāṃkhyakārikā (SK)120 of Īśvara Kṛṣṇa, the earliest surviving text of the system: By virtue, ascent to higher planes, and by vice, descent to lower planes take place; by knowledge release is obtained while by the reverse of it (i.e. by ignorance) one gets bound.121 Every activity has a consequence. Every pursuit leads to a certain end. Within the Sāṃkhya system, propounding a philosophy which Gerald James Larson described as ‘primarily soteriological,’122 in the sense that it seeks an end to the suffering of existence, liberation arises out of the discriminative knowledge (viveka-jñāna) which clearly distinguishes puruṣa from prakṛti. Here again, the principles of dialectic are called upon in order to discriminate pure consciousness (puruṣa) from everything that it is not (prakṛti). The whole soteriology of the Sāṃkhya, its therapy, consists in mastering this contemplative dialectical process. According to Sāṃkhya, before aiming for the station of no station, the experience of pure consciousness (puruṣa), the neophyte must learn to recognize his own position (sthiti/pakṣa) in the world (within prakṛti). Only then can he progressively negate what he is not (prakṛti) and proceed to his goal (puruṣa). Discriminative wisdom is gained from contemplating the given tattva-s and liberation ultimately ensues from their ‘negation,’ whence one is said to abide in pure consciousness (puruṣa) alone, the only permanent and unchanging tattva, considering that prakṛti is permanent but changeable.

The beginnings of Jaina doxography  125 Another Sāṃkhya treatise, perhaps the most informative commentary, the Yuktidīpikā (YDī), mentions vīta and avīta (or āvīta) as means by which the system argues with other schools. The pair is said to work together in establishing the reasoning employed by Sāṃkhya to prove the existence of primordial matter (prakṛti), as the prime example. The second term, āvīta, is of significant interest in the context of doxographical dialectic. Shujun Motegi explains, The essence of the concept of āvīta is said to be in denial of the concept propounded by an opponent, by which the standpoint of a proponent is established.123 Within the reasoning method of Sāṃkhya, a set of supporting proofs (vīta) is first given in an argument before an equally numbered set of counter-possibilities (āvīta) drawn from opponents is negated. To use the negation of opposing views in support of one’s own position also appears to be the method of doxographical dialectics. Indian doxographies reproduce the same dialectical scheme with the tattva-s promoted by competing systems. The neophyte who studies them must first gain the wisdom born from hearing about the devatā and tattva-s of every school. In a second movement, he must reason to discriminate the only valid ones. Then only can he finally transcend them all, leaving only the ‘Tattva of tattva-s’ to be experienced, the underlying reality behind plurality and change, the definition of which depends on the tradition to which one belongs, śūnyatā for some, puruṣa, ātman, and so on for others. Hence, one could say that, by providing all possible competing positions regarding the nature of reality, Indian doxographies act as performative texts which, by simulating various dialectical approaches to truth, however misleading at first, make use of opposing views as many stepping stones to be identified and abandoned (or adjusted) on the journey towards a ‘meta- position’ transcending the yoke of dialectic and conceptuality altogether, which coincides with omniscience in Jaina context. In brief, doxographies serve to illustrate how the sapiential foundations (the tattva-s) of conventional views (darśana-s) must be dialectically harnessed (through yukti) in order to rein in their non-conceptual ultimate perspective, the triumphant chariot of liberation. Thus, the dialectical practice of philosophy as captured in doxographies is essential to every darśana, gnostic by nature. This is the jñāna-mārga, or the jñāna-yoga of Haribhadra. Therein, even the perspective of the Mīmāṃsā must be seen for what it is worth. Hence, to conclude this discussion on doxography in Jainism and to illustrate Haribhadra’s doxographical style, we will now look at how he represented the Mīmāṃsā system.

2.3  ṢDS Chapter 6: the Jaiminīya Compared with the 167 verses of the MHK on the Mīmāṃsā, the ten verses of the ṢDS require little comment. Their formulations are straightforward and easily understandable. Their topics are clearly identifiable. Perhaps one should note that

126  The beginnings of Jaina doxography Haribhadra does not refer to the Mīmāṃsā directly, but to its founder, Jaimini, the author of the Mīmāṃsāsūtra (MS), a text of 2745 sūtra-s, whose chronological composition it is hard to determine, ranging from the third century BCE to the third century CE.124 We can safely assume that the MS was a source of information for Haribhadra. Being a Brahmin by birth, it is not impossible that he studied the text in his youth. Haribhadra makes it clear that the followers of Jaimini do not believe in any omniscient divinity (devatā), unlike Bhāviveka’s description of the system. Hence, knowledge about reality, and particularly about dharma, which cannot be directly perceived, is said to be acquired through the Veda, which de facto becomes a devatā. Thus, inquiry into dharma, motivated by the desire to know dharma (dharmasya jijñāsā), is the foundation of the school: dharma acting as a tattva. The MS indeed begins by stating the necessity of such a desire (MS 1.1.1.: athāto dharmajijñāsā). The defining characteristic (lakṣaṇa) of dharma is to impel ritual action through ‘injunction,’ rendered as nodanā, but also often expressed in literature by the equivalent term vidhi. Having clarified the devatā and tattva of the darśana, Haribhadra proceeds to list and define its six means of knowing (pramāṇa-s): perception (pratyakṣa), inference (anumāna), verbal testimony (śabda), analogy (upamāna), presumption (arthāpatti), and absence (abhāva). The most important of these, for the Mīmāṃsā, is śabda, since it is the knowledge born of the Veda, the devatā of the school. Debates about these various pramāṇa-s have preoccupied Indian epistemology through the ages. With this extremely brief summary, Haribhadra concludes his presentation of the school, following a template which he reproduced throughout the text. Having finished his presentation of the pious (āstika) systems (darśana-s) which entertain faith in an afterlife, Haribhadra moves on to discuss the unfaithful (nāstika) view of the Lokāyata-s. The suggestion that such a meagre description of views may have served as a debate manual is ludicrous. Equipped with so little information, one is in no position to engage in the kind of intense logical argumentation upon which the fate of one’s community, or of oneself, is at stake. That it be the only content of some history of Indian philosophy is also hard to imagine. Surely, one is to look for another explanation.

Notes 1 In his PhD thesis (1998), Jon M. Yamashita argues against the attribution of the SDS to Mādhava. His conclusion goes as follows: “In conclusion, both internal and external evidence suggests that the author of the Sarvadarśanasaṅgraha was not Mādhavācārya, the elder brother of Sāyaṇa. Rather, the author must have been at least a generation later than Mādhavācārya. Two possible candidates which have been proposed are [1] the son of Sāyaṇa – whose name Māyaṇa is to be considered a corruption or regional form of ‘Mādhava,’ and [2] Cannibhaṭṭa, the naiyāyika author of the Tarkabhāṣāprakāśikā – either independently or as part of a committee. There is no direct proof to support either choice; the available evidence, however, allows each to be given serious consideration as possible candidates for the authorship. At most, we can say with certainty that under either claim, the Sarvadarśanasaṅgrahawould have

The beginnings of Jaina doxography  127 been composed during the fourteenth century under the patronage of the Vijayanagara court” (1998: p. 31). 2 Halbfass 1979: p. 196. 3 Qvarnström (1999: p. 179) argues that Bhāviveka’s MHK is the first Indian doxography. Following the broad definition of doxography which I gave in the beginning of this work, insisting on the dialectical nature of doxography, underpinning its systematic form, and so far as the actual state of research goes, I agree with him. 4 On Siddhasena’s pre-doxographic contributions, see Qvarnström 1999: p. 177–179. 5 Nicholson 2010: p. 147. 6 Folkert 1975. 7 Folkert 1993. 8 Folkert 1975: p. 257. 9 Folkert 1975: p. 258. 10 Haribhadra cannot be said to be a mere doxographer, for, like Bhāviveka, he produced genuine independent philosophical materials. Yet, the present discussion is mainly interested in his doxographical work, hence the title. 11 Jacobi 1926: p. xxff. 12 Another helpful summary of Haribhadra’s life and work can be found in Chandra 1988. 13 Chapple and Casey 2003: p. 1–14. 14 As mentioned in Shukla 1989: p. 1–2. 15 Paul Dundas (1996: p. 147) read this hagiographic detail as somewhat ironic, a signal that the Jaina tradition was unwilling to see Sanskrit triumph over its ancient linguistic custom. 16 A gaccha, in Jainism, refers to a monastic order. 17 The text was edited by Jacobi (1926). 18 For a detailed list, see Shukla 1989: p. 7–10. 19 See Chandra 1988 and Shukla 1989: p. 5. 20 Shukla 1989: p. 6. 21 Barbato 2017: p. 33. 22 The expression is from Thomas 1960. 23 In Majjhima-nikāya, sutta 99, as quoted by Matilal 1981: p. 7. 24 Dīgha-nikāya, 1, 191. 25 See Matilal 1981: p. 7. 26 Matilal 1981: p. 8. 27 Matilal 1981: p. 23. 28 Matilal 1981: p. 18. 29 For example, reading anekānta-vādaas a socioreligious function, as opposed to a purely religious one, W. J. Johnson (1999: p. 105) claims that its primary function “is not philosophical or even ethical, but is a means of defending, rather than defining, an already established Jaina social identity.” On the motive and method of rational enquiry in ancient India, see the contribution of Ganeri 2001b. 30 As can be attested in the work of the famous Jain logician Siddhasena Divākara, also a Brāhmaṇa by birth. See Dundas 1996: p. 147–148. 31 Bronkhorst 2010a: p. 8. 32 The term ‘outsiders’ is used here to refer to any Indian religious group of non-Jaina denominations. To qualify such outsiders as ‘heretics,’ as commonly done, would be problematic, since there are no Sanskrit terms fully conveying that meaning. The term tīrthika, also common in Buddhist literature, refers to people who follow a way (tīrtha) other than one’s own. It is often rendered as ‘ford-maker.’ Peter Flügel (2009: p. 85, n. 20), quoting Padmanabh S. Jaini (1979: p. 80), explains that, based on common heritage, contemporary Jain ‘orthodoxy’ classifies only originally non-Indian traditions as truly heretical. On the Prākrit term pavayaṇa-nihṇaga (concealers of the doctrine), see Dundas 2006a. 33 Bronkhorst 2010a: p. 1–2.

128  The beginnings of Jaina doxography 34 Bronkhorst 2010a: p. 10–11. 35 The dates of the Tattva-artha-sūtra are debated. Robert Zydenbos (1983: p. 9–12) dates its commentary (bhāṣya) to the fifth century and hence the sūtra text must be earlier. It is generally accepted to have been written between the second and fifth centuries. Paul Dundas (2006b: p. 395–396) argues for the fourth through fifth centuries. 36 Balcerowicz 2017: p. 77. 37 Balcerowicz 2017: p. 79. 38 De Visione Dei, Capitulum II Visus autem absolutus ab omni contractione simul et semel omnes et singulos videndi modos complectitur quasi adaequatissima visuum omnium mensura et exemplar verissimum . . . et manet ab omni varietate penitus absolutus. Sunt enim in absoluto visu omnes contractionum modi videndi incontracte.

The translation is from Hopkins 1985: p. 683. 39 Balcerowicz 2017: p. 80. 40 Balcerowicz 2017: p. 80–81. 41 Balcerowicz 2017: p. 71–89. 42 Balcerowicz 2017: p. 89–97. 43 Notably, the expression occurs in the Bhagavad Gītāand the Upaniṣads upon which Śaṅkara commented. 44 Qvarnström 1999: p. 170–171. Although Qvarnström gives tapaḥ śuddha for “pure austerity,” one would rather expect tapaḥ śuddhaṃ. 45 Johnson (1999: p. 105) argues, “the Jains’ identity as a religious group largely resides in asceticism.” 46 Johnson 1999: p. 105. 47 Positive apologetics denotes a strategy which constructs arguments to establish one’s beliefs as true and valid. In that sense, it can be construed as an offensive strategy aiming at converting the other. Its negative counterpart, by contrast, criticizes arguments formulated against one’s belief and, in that sense, constitutes a defence seeking to dismay outside attacks as unwarranted. Both have in common to use reasoning as tools in the service of religion. 48 Barbato 2017: p. 92–96. 49 Matilal 1981: p. 28. 50 On the three causes of bondage and their corresponding threefold purification, see Tatia 1951: p. 144–155. 51 Folkert 1975: p. 252. 52 Qvarnström 1999: p. 181. 53 Maṇibhadra is listed in Karl H. Potter’s Encyclopedia of Indian Philosophy, Volume I, within the list of texts whose authors’ dates are unknown, under the entry: “DU425. Maṇibhadra: 1. Laghuvṛtti on Haribhadra’s Ṣaḍdarśanasamuccaya. See e410.16:5, 13, 18, 25.” It can be accessed online at https://faculty.washington.edu/kpotter/xtxtdu.htm, as per January 2018. 54 For a complete list of commentators, see Balcerowicz 2003: p. 147. 55 For example, among the five Jaina compendia studied by Folkert, Merutuñga’s Ṣaḍdarśananirṇaya, written in prose, is the only one to present a refutation on non-Jaina positions, whereas Rājaśekhara’s Ṣaḍdarśanasamuccaya, Jinadatta’s Vivekavilāsa, and the anonymous Sarvasiddhāntapraveśaka follow the model of Haribhadra and do not state any refutation. Bhāvasena’s Traividyaviracita adopts the model of Merutuñga, in prose, with refutation. 56 Balcerowicz 2003: p. 146. 57 Balcerowicz 2003: p. 147. 58 Gracia 1992: p. 246. 59 Though Richard King (1999: p. 44–45) suggests that the use of the term darśana, to denote a philosophical school, appears in the MHK, and even if Eckel, based on the

The beginnings of Jaina doxography  129 commentary on verse 9.17, has shown that Bhāviveka equates the term with siddhānta, one should note that the practice of using the term darśana to refer to a ‘philosophical school’ was only emerging. Bhāviveka did not use it in any of his works’ titles nor in his subdivisions of philosophical schools within the MHK, preferring the term tattva. Yet, this reference to the tattva-s of each school already signals a relation between established philosophical tenets (siddhānta), their proposed lists of tattva-s, and the emerging use of the general term darśana. 60 My translation of samyagdarśanajñānacāritrāṇi mokṣamārgaḥ (TAS I.1). 61 My translation of tattvārthaśraddhānaṃ samyagdarśanam (TAS I.2). 62 jīvājīvāsravabandhasaṃvaranirjarāmokṣāstattvam (TAS I.4). 63 Folkert 1975: p. 240–241. 64 NS 1.1.1: pramāṇaprameyasaṃśayaprayojanadṛṣṭāntasiddhāntāvayavatarkanirṇayavādajalpa vitaṇḍāhetvābhāsacch alajātinigrahasthānānāṃ tattvajñānān niḥśreyasādhigamaḥ | 65 SK 64 evaṃ tattvābhyāsān nāsmi na me nāhamityapariśeṣam | aviparyayād viśuddhaṃ kevalam utpadyate jñānam|| 66 On the debates around the relation of ānvīkṣikī to ātma-vidyā, see Halbfass 1988: p. 276. 67 Although one would be tempted to include the Lokāyata philosophy to the exceptions, the too limited knowledge which we possess on the school, added to the fact that the system was accepted as one of the six tarka-vidyā, all of which were ātma-vidyā, as shall be discussed within these pages, leads one to consider that there might be more to the Lokāyata ideology than what its opponents tried to depict: being merely atheistic, hedonistic, materialistic, and so on. For example, some have argued that the school presents a kind of spiritualist ‘live in the moment’ philosophy (Kanchi 2016) and that asceticism, not hedonism, was part of its lifestyle (Bhaṭṭācārya 2011; Ballanfat 1997). In any case, it represented a serious threat to Brahmanical authority and is thus unlikely to have been a mere advocate of sensuality. 68 Folkert 1975: p. 279. 69 Pahlajrai 2004. 70 Pahlajrai 2004: p. 13. 71 Kapadia 1933–34: p. 98–99. 72 Umāsvāṭi, in his Saṃbandhakārikā-s (v. 4–6), also offers a sixfold typology of humans, categorized from the point of view of the kind of life spent: (1) adhamādhana; (2) adhama; (3) vimadhyama; (4) madhyama; (5) uttama; and (6) uttamottama. See Kapadia 1933–34: p. 104. 73 Items 10, 13 and 20 are mentioned in Kapadia 1933–34: p. 106. 74 However, one could argue that to represent a ‘whole’ is the function of any numerical grouping. Thus, with a group of six elements, the only difference is the increased discrimination within the whole, in comparison to groups made out of fewer constituents. Nevertheless, since Jains divide the universe into six substantial categories, the number seems to express a more fundamental division of reality expressing a cosmic ‘integrity.’ Further speculations could explore the symbolism of the 2 × 3, symbolized by the hexagram of the two locked triangles, often appearing in Indian iconography, including Jainism. However, any such exploration is unlikely to contribute anything meaningful to the current discussion on Indian doxography. 75 Ājīvikism and Jainism shared many features, among which is the relation between the Ājīvika notion of abhijātiand the Jaina leśyā, which, though similar in many points and probably related, cannot be identified. The issue has been discussed by Basham (1951), Jaini (1979), and Schubring (2000), as reported by Bronkhorst 2012: p. 825.

130  The beginnings of Jaina doxography 76 The correlation between the six karmic colours (leśyā-s) of souls and their correspondingly tainted views (darśana-s) is here conjectured, based on the principle of correspondence which seems to pervade the recurrent sixfold taxonomies of Jainism, yet I have never seen it explicitly mentioned in Jaina literature. 77 Most of these also feature in the MHK, but Bhāviveka seems to have replaced the Lokāyata and Jaina systems by the Vedānta and Mīmāṃsā ones, just briefly hinting at the Jaina in an addenda discussion on omniscience, indicating that the two Brahmanical schools were more significant adversaries. 78 Gerschheimer 2000–2001: p. 183. 79 The dates of the MS are a hotly contested issue. For example, in its entry on the text, the online New World Encyclopediapresents the following series of theories: “For composition between 200 B.C.E. and 200 C.E. see: Avari, 142. For dating of composition ‘between the second century B.C.E. and third century C.E.’ see: Flood. An Introduction to Hinduism (Cambridge: Cambridge University Press, 1996), 56. For dating of Manu Smritiin ‘final form’ to the second century C.E. see: John Keay. India: A History. (New York: Grove Press. 2000), 103. For dating as completed sometime between 200 B.C.E. and 100 C.E. see: Hopkins, 74. For probable origination during the second or third centuries C.E., see: Hermann Kulke and Dietmar Rothermund. A History of India (New York: Barnes & Noble, 1986), 85.” See www.newworldencyclopedia.org/entry/Manusmrti, note 6, as of 23 January 2018. 80 Gerschheimer 2000–2001: p. 184. 81 On the place of analysis, logic, and reason among the schools of Indian philosophy, including the comment on Sāṃkhya and the ‘path of knowledge,’ see Aklujkar 2017: p. 47–50. 82 This might lead us to doubt the negative stereotypes forced onto the Lokāyata worldview by its opponents. A valuable recent assessment of scepticism and materialism in ancient India can be found in Doniger 2018. 83 Gerschheimer 2000–2001: p. 186. 84 Gerschheimer 2000–2001: p. 188. 85 Gerschheimer 2000–2001: p. 189. 86 ṢDS 78–79 naiyāyikamatād anye bhedaṃ vaiśeṣikaiḥ saha | na manyante mate teṣāṃ pañcaivāstikavādinaḥ || ṣaḍdarśanasaṃkhyā tu pūryate tanmate kila | lokāyatamatākṣepāt kathyate tena tanmatam || 87 On the use of the term āstikain Haribhadra, having no reference to the Veda but rather referring to a faith in the afterlife and so on, see Nicholson 2010: p. 172–176. On the labels āstika and nāstika in general, see Aklujkar 2017: p. 43–45; Doniger 2018: p. 131–136; Nicholson 2012. 88 A shorter version of this chapter won the first prize of the Graduate Essay Contest of the Society for Asian and Comparative Philosophy (SACP) Conference, in Krakow, 2018. 89 These criteria, tattva and devatā, are the topics of the next chapter. 90 Perceptions of tolerance in Jaina doxographies were likely induced by the way Matilal (1981) and Kapadia (1947) presented anekānta-vāda. Discussing Haribhadra’s doxographies on yoga, Christopher Chapple (2003) is also prone to see various instances of ‘tolerance.’ Claims that Jainism as a whole is overly tolerant and nonargumentative have, however, been criticized and moderated, for example in Barbato (2017) and Cort (2000). Balcerowicz (2015: p. 49), openly sceptical, talks of a ‘tolerance myth.’ 91 Folkert criticized the notion of tolerance projected on the compendia. He also singled out A. K. Warder’s view of doxographies, for granting “to their discussions of various viewpoints an unreflective, descriptive character that cannot properly be assumed to

The beginnings of Jaina doxography  131 exist without further investigation of the texts and what they represent” (Folkert 1993: p. 230). 92 Folkert 1975: p. 228. 93 Brummett 1996: p. 627. 94 Halbfass 1988: p. 351. 95 Matilal 1981: p. 18. 96 ṢDS 45–46 jinendro devatā tatra rāgadveṣavivarjitaḥ | hatamohamahāmallaḥ kevalajńāna darśanaḥ || surāsurendrasaṃpūjyaḥ sadbhūtārthopadeśakaḥ | kṛtsnakarmakṣayaṃ kṛtvā saṃprāptaḥ paramaṃ padam ||

Qvarnström (1999: p. 194), at verse 46a, gives prakāśakaḥ,translated as ‘he explains,’ instead of upadeśakaḥ, as given in Murty (1957). Both meanings are similar, but I kept the older version of Murty, as I generally did for my reading. 97 The reason why I translate devatā as ‘authority’ and not with the more common meaning of ‘deity’ shall become clear in the next chapter. 98 ṢDS 1 saddarśanaṃjinaṃ natvā vīraṃ syādvādadeśakam | sarvadarśanavācyo ’rthaḥ saṃkṣepeṇa nigadyate ||

99 Which was generally hinted at without any reference or comment by Pahlajrai 2004. 100 Brummett 1996: p. 628. 101 For the dates and a discussion on sarvajña in Kundakunda, see Fujinaga 1999; Petit 2014. Fujinaga refers to a period around the second century CE, while Petit opted for around the fourth century. 102 The expression, from Jaini (2000: p. 30/92), is also quoted by Bronkhorst (2016b: p. 40). 103 Barbato 2017: p. 108–111. 104 Brummett 1996: p. 628. 105 This remark is to be read in the light of Bhaktin’s theory, where ‘carnival’ represents the reversal, the upside down, of ‘real’ life, organized on the basis of laughter and dedicated to the world’s renewal. See Renfrew 2015: p. 129–144. 106 On the Buddha’s silence, see Park 2006; Nagao 1991; Organ 1954. 107 Jain and Matukumalli 2014: p. 249–251. 108 To distinguish between sad-darśana and a-sad-darśana, for example, as suggested in the introductory verse of the ṢDS. 109 Folkert 1975: p. 251–265, 1993; Gerschheimer 2000–2001: 176–179; Qvarnström 1999: p. 181–182. 110 ṢDS 2 darśanāni ṣaḍ evātra mūlabhedavyapekṣayā | devatātattvabhedena jñātavyāni manīṣibhiḥ || 111 The sceptical, pragmatic, and materialist leanings of the Lokāyata worldview suggest that, if there is no external devatā, the authority in matters of knowledge is none but ‘oneself,’ one’s own critical thinking. Verses 81 and 82 of the ṢDS, though somewhat ironical, indicate that the system enjoins to have faith in no one else than oneself and to be critical. 112 Concerning the Lokāyata-s, Folkert seems to have exaggerated their nihilism to suit his theory of classification concerning the darśana-marks. He says that the Lokāyata-s have no tattva-s when the text, at verse 83, explicitly mentions the four elements and the pratyakṣa-pramāṇa, as it did with previous schools. 113 The word ‘theory’ should here be taken in its Greek sense of θεωρία. The Blackwell Dictionary of Western Philosophyintroduces ‘theōría’ as “Greek, vision of the real

132  The beginnings of Jaina doxography in the mind, hence contemplation or speculation, from theorein, to contemplate and theasthai, to gaze on, giving contemplation visual associations” (Bunnin and Jiyuan, 2004: p. 684). It is perhaps the term which best renders the meaning of the Sanskrit darśana, besides ‘worldview.’ However, for moderns, the ancient Greek semantic of the term ‘theory’ is not obvious, and its use to render darśana could become misleading. Too often is ‘theory’ opposed to ‘practice.’ 114 The term ‘accident,’ from Latin accidens, ‘something that happens,’ related to the Greek sumbebekos, from the verb sumbainein, ‘to come together,’ ‘to happen,’ and better translated as ‘coincident’ or ‘concomitant,’ is here to be taken in its Aristotelian sense. See Bunnin and Jiyuan (2004: p. 7). 115 Folkert 1975: p. 261. 116 YDS 101 āgamenānumānena yogābhyāsarasena ca | tridhā prakalpyam prajñāṃ labhate tattvam uttamam ||

The translation is from Chapple and Casey 2003: p. 125. The second part of the verse might be more literally rendered by “The threefold wisdom being accomplished; he obtains the highest reality.”   However, as Philipp Maas kindly pointed out to me, YDS 101 is not an original part of the YDS but a quotation of a stanza occurring in PYŚ 1.48. The earlier quotation in the PYŚ reads a slightly different text with dhyānābhyāsarasenain pādab and yogam uttamamin pāda d. More important in the present context is, however, the reading prakalpayan (active present particle) instead of prakalpyam (passive future participle) in the rendering of the YDS, which makes the stanza unmetrical. I believe that this borrowing suggests that Haribhadra accepted the structure of a threefold process of insight which he likely identified in the PYŚ. 117 Bouthillette 2017b: p. 112. 118 US: 19.25 samaṃ tu tasmāt satataṃ vibhātavad dvayād vimuktaṃ sadasadvikalpitāt | nirīkṣya yuktyā śrutitas tu buddhimān aśeṣanirvāṇam upaiti dīpavat || 119 Śāstra-s generally focused on the tattva-s of their own tradition, often accompanied by the objections of others (pūrvapakṣa) and their refutation (uttarapakṣa). To systematically engage with the tattva-s of competing traditions is an innovation of the ‘age of doxography,’ appearing around the time of Bhāviveka. 120 On the dates of the text, not later than 569 CE, when it was translated into Chinese, see Larson 1969: p. 4. 121 SK 44 dharmeṇa gamanam ūrdhvaṃ gamanam adhastād bhavatyadharmeṇa | jñānena ca apavargo viparyayād iṣyate bandhaḥ || 122 123 124

Translation taken from Virupakshananda 1995: p. 93. Larson 1969: p. 201. Motegi 2010: p. 362. Verpoorten 2014: p. 58.

3

The beginnings of Advaita doxography Śaṅkara’s SSS

3.1  Advaita doxography in context After having examined such famous and important doxographies as Bhāviveka’s MHK and Haribhadra’s ṢDS, it might come as a surprise to some that the present chapter is now dedicated to a text which has been mainly ignored or downplayed by modern scholarship, the Sarvasiddhāntasaṅgraha (SSS). After all, Advaita’s most famous doxography is incontestably the fourteenth-century Sarvadarśanasaṅgraha (SDS), which Halbfass esteemed so highly.1 Yet, the Advaita tradition claims that its first saṅgraha was composed by its leading theologian, Śaṅkara, who, though he is not the first historical Advaitin, is revered as the lineage’s (parampara) prime authority (ādi-ācārya), the universal teacher (jagad-guru), next to the Prasthānatrayī, of course, its three sets of authoritative scriptures. It is Śaṅkara who gave birth to the lineage’s institutions and fame. But this traditional claim concerning the SSS has generally been dismissed by scholars, for serious reasons which will be summarized within these pages. Having lost its initial prestige, the SSS ironically fell in the shadow of Śaṅkara and eventually in the forgotten realms of ‘spurious’ literature. Yet, as I will argue, the text most likely preceded the SDS, from which the latter borrowed on several occasions. To begin with, the full name of the SSS, gathered from its colophon at the end of each chapter, is in fact the Sarvadarśanasiddhāntasaṅgraha (SDSS); and the SDS appears to pay homage to its predecessor by adopting an abridged version of the same.2 The fact that the SSS is written in verse and that it limits its exposition to a concise, schematic, and repetitive set of doctrinal elements within each darśana, more or less like the earliest doxographies of other schools, also presents an early stage in Advaita doxography. The SDS, on the other hand, is much more elaborated, written in prose. However, for these compositional criteria to be used in determining the chronology of doxographies, further research would need to be done on a scale which the present study cannot afford. But, more on that later. Before moving further in discussing the authorship of the SSS, my only point here is to state, at the outset, that I take the SSS as one of the earliest known samples of Advaita doxography. With or without the associated aura of Śaṅkara, it constitutes a rich and valid sample of Advaita doxography, displaying a dialectical model which defines the genre.

134  The beginnings of Advaita doxography As was done with previous chapters, I will first examine the ideological context in which Advaita doxography evolved. Since Śaṅkara is such a prominent figure in the tradition, and even if the authorship of the SSS is uncertain, I will examine some elements of his biography which can help shed more light on the central place of debates and apologetics within his pedagogy and conception of religious life. As usual, I will then proceed to highlight the ideological context in which Advaita doxography evolved, focusing on a set of fundamentals essentially encapsulating Śaṅkara’s teachings. I will finally comment upon the two chapters of the SSS related to Mīmāṃsā, to illustrate the style of the text. As will become clear, the relation of Advaita to Mīmāṃsā is an intimate one. Thus, this chapter will add further contrast to my portrayal of the Brahmanical exegetical school. So far, its sketch has been rather skeletal. What I aim to achieve within the next pages, in addition to what was already drafted, is to bind the many sociohistorical, philosophical, and religious threads of the previous discussions into a coherent narrative. Here, doxographical literature shall serve as a window into a turbulent ideological landscape in which a dominant Brahmanical community rechristened its identity, both through a renewed interpretation of Vedic scriptures and by borrowing doctrinal elements and exegetical tools from the many competitors it sought to subordinate or censure, in order to consolidate its grasp on power and assert its authority over both the physical and spiritual realms of Āryāvarta.3 In brief, to recall a recent title from Bronkhorst, the dialectic of Advaita doxography is part of a powerful rhetorical arsenal reinforcing the ideological weaponry through which Brahmanical ideology ‘won’ over its opponents in India.4 3.1.1  The conquest of Śaṅkara So considerable was the intellectual activity of the ‘Hindus’ around his work that commenting upon the great teacher Śaṅkara (Śaṅkara-ācārya) became as humbling a task as, to paraphrase Alfred North Whitehead, adding a mere footnote to a long series of commentaries on Plato. Yet, one should take courage in considering that no canvas holds without its many threads. Ironically, in spite of his reputation, knowing the man has become as difficult an enterprise as knowing the ‘self’ (ātman), the supreme religious goal that Śaṅkara invited his Brahmin followers to achieve; through a careful exegetical removal (apavāda) of the net of delusory superimpositions (adhyāropa) projected upon it. Like the Buddha or Jina, the historical character of the wandering teacher, crushed under his own fame, has long been eclipsed by the legendary man’s persona. Thus, the historian is left with the daunting task of critically engaging the many reports involving the elusive preceptor. Fortunately, many have already begun the work, though surely none have ever found Śaṅkara’s pure ‘identity’ (śuddha-brahman), nor will that be the case here. In terms of primary sources, we can count three baskets through which information can be gathered about Śaṅkara: his own writings, the traditional hagiographies, and the accounts of his own pupils. For our purpose here, the whole gamut of possible biographical details need not be listed. I will simply select a

The beginnings of Advaita doxography  135 few examples which serve to illustrate the central place of debates and apologetics within the teacher’s pedagogy and give an idea of the religious atmosphere in which he gravitated. For, this information can already serve to analyze the dialectic at play in Advaita doxographies. Conveniently, Jacqueline Suthren Hirst has given a compelling account of Śaṅkara’s life, borrowed and augmented from David N. Lorenzen,5 while acknowledging that it remains a ‘plotted version.’ She listed eleven significant biographical elements, eight of which are worth reproducing here: The facts that we can accept with confidence are [i]

that he was born in a Brāhman family from the Kerala region [South West India] but left home at an early age to become a wandering ascetic (saṃnyāsin); [ii] that he became a student of a teacher named Govinda, a pupil of Gauḍapāda [the directness of the connection with Gauḍapāda is questionable]; [iii] that he wrote various philosophical and devotional works including commentaries on the Upaniṣads, Bhagavad Gītā, and Brahmasūtra; [iv] and that he travelled throughout India with his own disciples defeating rival theologians. It is probable though not at all certain [viii] that one of his most important conquests was a man named Maṇḍana Miśra, a mīmāṃsā follower of Kumārila; [ix] that his most important disciples were named Ānandagiri, Padmapāda (Sanandana), Sureśvara (doubtfully identified with Maṇḍana Miśra and Viśvarūpa [identifications disproved]), Hastāmalaka and Toṭakācārya [some stress only four, omitting Ānandagiri or identifying him with Toṭaka, and relating one pupil to each of the maṭhas mentioned in [x]]; [x] that he established various religious centers, especially at Śṛṅgerī in the South, at Puri in the East, Dvārakā in the West, and Badarikāśrama in the North [a strong claim is also made for Kāñcī in the South, but some versions mention none at all]; [xi] and that he died at a young age, thirty-two according to most accounts, at either Badarikāśrama [in the Himalayas], Kāñcī [in modern Tamilnadu] or somewhere in Kerala.6 In a nutshell, the list provides the armature of the revered teacher’s biographical chronicle. The points on which I particularly want to insist now are i, iv, viii, and x. Together, they have been instrumentalized early on by the Brahmanical tradition to convey a narrative of conquest, one in which Śaṅkara wanders victoriously throughout the subcontinent and subdues, one by one, the many enemies of the Veda, and those who entertain a divergent reading of the same scriptures.

136  The beginnings of Advaita doxography Such a diegesis insinuates that Śaṅkara’s life enacted the kind of symbolic powerconferring rituals used by the kings of lore, to project their sovereignty, as in the most important of all sacrifices, the Vedic horse sacrifice (aśva-medha), but also the king’s consecration (rāja-sūya), both promoted in the Mahābhārata and the Purāṇa-s. Through their myth-making, the narrators of Śaṅkara’s life story came to conjure the spirit of a divine overlord whose greatest life achievement, besides his religious consummation, was to purify the land of the Ārya-s from swarming demonic ‘foreigners.’ When his initiatory journey to the four directions of the subcontinent is read through the lens of the aśva-medha, Śaṅkara himself acts as the sacrificial horse. His life sacrifice is offered for the Veda. This ennobling consecration has the double performative effect of elevating the Veda – as embodied by its Brahmanical depositaries, to whom Śaṅkara belongs – to the status of cakravartin of the physical and spiritual realms and of declaring its enemies vanquished. Thus, within this magical ritual play, in accord with the sacrificial economy which informs the concept of the ‘cosmogonic puruṣa,’ the primordial male, Śaṅkara appears as both the ‘sacrifiant’ and ‘sacrifié,’ just as Prajāpati, the lord of creatures, which instituted the original sacrifice by offering himself at the dawn of creation. Śaṅkara himself associated the sacrificial horse with Prajāpati in his commentary on the first adhyāya of the Bṛhadāraṇyaka Upaniṣad (BAU), where contemplation upon the sacrificial steed is said to represent the universe. Śaṅkara comments, The horse which is a part of the sacrifice has to be purified; hence its head and other parts of its body are to be looked upon as certain divisions of time, etc. (and not vice versa). And it will be raised to the status of Prajāpati by being meditated upon as such. In other words, the horse will be deified into Prajāpati if the ideas of time, worlds and deities be superimposed on it, for Prajāpati comprises these.7 Hagiographers have purposefully depicted Śaṅkara as the stallion of Brahmanism, the Prajāpati of the teachings, wandering about the land to fertilize the minds of believers with the sole true Vedic faith, the ‘royal’ creed (rāja-yoga?).8 Śaṅkara precipitated the defeat of his enemies, those who interfered in the enthronement of the Veda as king among scriptures. Śaṅkara’s conquest is couched in spiritual terms. By adopting saṃnyāsa, the life of a renouncer, Śaṅkara became the sacrifice. Having no more to expect from mundane existence, he is the liberated self (ātman) who has burned all superimpositions from the realm of consciousness, through the purifying fire of apavāda. Here, one can notice the bivalence of the concept of superimposition (adhyāropa), which can be applied both to (1) external ‘nay-sayers’ (nāstika-s), holding onto false beliefs, and to (2) internal obstacles (moha), false cognition. Apavāda, allegorized as the fire of Brahmanical knowledge (brahma-jñāna), allows for no heterodoxy. Salvation can only pass through the caste-exclusive door of Vedic revelation (śruti). We shall return to these notions momentarily. In brief, what hagiographers seek to convey is

The beginnings of Advaita doxography  137 that the conquest of Śaṅkara was total (temporal and spiritual), Brahmanical, and sanctioned by God. He is said to have brought forth the ultimate end, the culmination of the Vedic path (veda-anta), instituted over land and souls. This symbolic narrative might also explain the enigmatic mythological death of Śaṅkara in the flower of youth, as if executed while still in full potency, like a sacrificial stallion, to impregnate the Aryan land, the sacrificing queen, with his fertilizing vitality. Jonathan Bader, in his doctoral dissertation on the traditional accounts of the life of Śaṅkara, commented extensively on the obvious symbolic connections between the traditional accounts of Śaṅkara’s conquest and the royal rāja-sūya sacrifice.9 Faced with such a heavily charged symbolic narrative, the historian is compelled to take the hagiographical accounts with a grain of salt. Yet, whether true or false, the fact that the word vijaya (conquest) features in the titles of twelve of Śaṅkara’s hagiographies, which together total about twenty, including the famous one of Mādhava-Vidyāraṇya, the alleged author of the SDS,10 the Śaṅkaradigvijaya (Śaṅkara’s Spatial Conquest), speaks volumes about the triumphant spirit of the Advaita community and its political aspirations. We shall soon observe how Advaita doxographies exuded the same triumphalism. For now, let’s return to Hirst, who summarized Vidyāraṇya’s plot: In this Life, Śiva takes birth as Śaṃkara in response to a plea from the gods that the Vedic path is being neglected, for it is being criticized by Buddhists who are converting people away from it and rejected by Vaiṣṇava and especially Śaiva sectarians who are spreading their own degraded practices (Canto 1.27–29). Like Viṣṇu who takes birth to restore dharma and (for Śaṃkara) to teach the Advaitin path, Śiva descends as a great teacher ‘to save the world floundering in the ocean of transmigratory existence (1.84).11 Thus, the hagiographer equates Śaṅkara with an avatāra of Śiva, descended upon the world stage from his heavenly abode to save the dharma from the demons of delusion. His enemies, the revilers of the true faith, can easily be identified by the audience – the legend is explicit. This spirit of conquest is not limited to hagiographical literature alone. Hirst also noted how, in the introductory verses to his subcommentary on Śaṅkara’s Taittirīya Upaniṣad, for example, Sureśvara, a direct disciple of Śaṅkara, after having praised brahman, clearly hails his master as the one “by the axe of whose speech the opinions of the logicians (tārkika-s) are destroyed.”12 There is little doubt that these logicians regroup the same six tārkika-s discussed in the previous chapter on Haribhadra; the assembly of which, as we explained, was somewhat ‘forcefully’ extended to include the Mīmāṃsā within the nascent doxographical compositions, even though it is not a tarka per se. With Sureśvara’s bellicose praise, we are now witnessing how the followers of Śaṅkara themselves entertained the idea that they did not belong to the group, but that they were dealing with an altogether different kind of knowledge, an undoubtedly higher one, beyond the reach of mere rational enquiry (tarka). Not only did they believe that the ‘non-dual’ (Advaita) teachings of their guru defeated non-Brahmanical creeds, but they were also adamant that it

138  The beginnings of Advaita doxography subordinated all the other inferior Brahmanical ‘sciences’ (vidyā-s); hence the victory cry of the tradition over the alleged conversion of Maṇḍana Miśra, a Mīmāṃsā follower of Kumārila (item viii on Hirst’s list). The introductory chapter of the SSS, for that matter, provides but one sample of the mythical pyramid of eighteen Vedic fields of knowledge which such thinkers designed to schematize a progressive elevation towards the ultimate revelation of the Uttara-Mīmāṃsā. The term Uttara-Mīmāṃsā designates the later exegesis dedicated to the knowledge content (jñāna-kāṇḍa) of the Veda, a synonym for Vedānta (which includes but is not limited to the Advaita kind). It contrasts with the Pūrva-Mīmāṃsā, the former exegesis of authors like Jaimini and Kumārila, dedicated to the elucidation of the ritual portion (karma-kāṇḍa) of the Veda. This pyramid of knowledge, with Vedānta theology at the apex, reminds one of the curriculums of Middle Ages universities, structured by Christian theology. The division of the Mīmāṃsā into a former and latter section, the latter of which only entertains the self-perception of its penultimate station, cannot fail to recall a similar pattern of sectarian contempt within Buddhism, where a selfproclaimed community took the name of the ‘Great Vehicle’ (mahāyāna) in pejorative contrast to a previous ‘Lower Vehicle’ (hīnayāna). Obviously, none of those designated as ‘former’ or ‘lower’ ever perceived themselves as underachievers, but the emerging narratives of systematization, which enforced a theoretical uniformity upon the given traditions, most likely to present a coherent and legitimate facade to the outsiders, could not suffer the equivalence of contradictory doctrines. There needed to be a final authority. In the Brahmanical context, the proponents of the Advaita Vedānta claimed that crowning position as their legitimate victory, mythically achieved by their conquering ascetic teacher. Indeed, as noted by Bader, epithets designating Śaṅkara as the king or emperor of ascetics are commonly used by hagiographers.13 Moreover, as their doxographies make clear, Advaitins had no qualms in extending this spiritual authority well beyond the confines of their own traditional walls, to assert it over all socioreligious domains. According to Bader, the dramatic portrayal of Śaṅkara’s life as a heroic war against the forces of evil is a “conspicuous feature of Śaṅkara’s hagiographies.”14 His depiction of the symbolic at play in the gest of Śaṅkara captures the many layers of temporal and spiritual subordination instigated by the Brahmanical hero: Although some of the texts do not refer to the digvijaya until later on in the narrative, it effectively begins the moment Śaṅkara sets forth from the abode of his guru. The journey is in itself an essential component of the digvijaya. Indeed this ritual conquest involves much more than simple combat. It represents, above all, a movement towards unity, in both the political and metaphysical spheres. In the first place, disparate kingdoms are always a potential source of conflict. The world conqueror brings together the warring states under the aegis of his superior authority. His emblematic white parasol signifies the protection and harmony he will afford them. Metaphysically, he becomes the sacred centre, the hub of the wheel, and restores the original

The beginnings of Advaita doxography  139 unity which was lost in the demarcation of space. This is accomplished by establishing his presence in the four quarters of the land and yet at the same time remaining at the centre.15 As we have seen previously, the aspiration to become the sacred centre of the world, the hub of the sacred wheel of life, is surely one equally shared by the Buddhists and Jainas, which they strategically aimed at through various dialectical means observable within their doxographies. Haribhadra’s doxography, for example, explicitly positioned the Jainas at the centre of world philosophy. Yet, with Śaṅkara, beyond philosophical dialectic, the political overtones of this governing centrality are particularly explicit. The climax of Śaṅkara’s glorious hagiographical narrative takes place when the wandering stallion of Brahmanism is finally crowned. In Hirst’s words, Finally, Śaṃkara ascends the so-called throne of omniscience at the temple to Mother Śāradā in Kashmir, the only learned man from the South to be given such admission, earning his right to enter by answering opponents from all the major schools (16.54 f).16 In other words, after having subdued all opponents, the king is enthroned and a new heavenly kingdom is born. As Bader explains, Śaṅkara’s digvijaya fulfills the ritual expectations of the rājasūya and repeats the heroic behaviour exemplified in Mahābhārata and Raghuvaṃśa. First of all, he establishes his presence throughout the whole of India. In the course of his journey, he is victorious on the battleground of learned assemblies. Śaṅkara defeats his rivals, the leaders of numerous sectarian groups. The very existence of these sects indicates the chaos resultant from the fragmentation of the Vedic tradition. In restoring the authority and correct understanding of the essential sacred texts, he re-establishes the true centre of the Hindu tradition.17 Hagiographers symbolically proclaimed the young Brahmin stallion emperor of all ascetics, embodying the only centrifugal force powerful enough to ‘pacify’ (sarvaśāntikṛt) – a classic Brahmanical euphemism – once and for all the demonic (asura/rākṣasa) wagers of spiritual wars, the miscreants who refuse to kneel to the supreme authority of the Veda.18 Not only is Śaṅkara’s rulership divinely sanctioned, but also he is divinity incarnated. To morally educated eyes, having learned of the unspeakable human tragedy left by the repeated attempts at tyranny, not less in India than elsewhere, this grand scenario can hardly be of good omen. Prefiguring the agonistic world envisioned by the icon of modern popular culture, John Ronald Reuel Tolkien, the narrators of Śaṅkara’s epic life story cast the Advaita theology into a ‘ring of power’: the one ring to rule them all. As shall be seen, Advaita doxographies, partaking of that same conquering mythology, offered the very blueprint of Sauron’s (Rudra?)19 projected spiritual empire. Regardless of the arguments which they may or may

140  The beginnings of Advaita doxography not formulate, their hierarchical structure explicitly assigns the respective position benevolently allotted to every subject by the all-seeing (sarva-darśana) eye of the Lord, throning highly on the towering ‘seat of the omniscient’ (sarvajña-pīṭha). Along the lines of Marshall McLuhan’s famous theory, when it comes to doxographies, it is their structure (medium), more than their minute arguments, which constitutes their message.20 Like Śaṅkara himself and his Brahmanical kin, Advaita doxographies are neither afraid of hierarchy nor ashamed of the elevated status which they proclaim for themselves. History shall remember that these were not the only men claiming to be the chosen ones. How to forget, also, that like others before and after him, the wise universal teacher (jagad-guru) oversaw that purity of blood and virility of gender, solely determined by birth, would not only dictate the social rank and livelihood of every incarnated soul, but also discriminate who is worthy of entering the heavily guarded and secret gateway to ritual and gnoseological power, acclaimed as the only way to freedom, the Veda. Such is the irony of man, forever binding his very freedom to an external authority ratified in texts – made in his image through his own making. Some may argue that the mythical conquest of Śaṅkara was not couched in words (śabda) alone. Beyond symbols and narratives, throughout his peregrinations, the wandering teacher supposedly left behind numerous and concrete evidences of his authority. It is said that he established ten ascetic orders of loyal ‘knights’ and four ‘fortresses’ of knowledge, at the four gates of India: at Śṛṅgerī in the South, Puri in the East, Dvārakā in the West, and Badarikāśrama in the North. Frank Whaling argues that the famous theologian thus incorporated into Hindu asceticism some of the distinctive features of the Buddhist saṃgha: As a result of Śaṅkara’s organizational activity, temple-colleges began to spring up, and monasteries became recognized as having an educational function. Traditionally, Śaṅkara is said to have organised the ascetics into ten orders with the names; Araṇya; Āśrama; Bhāratī; Girī; Parvata; Purī; Sarasvatī; Sāgara; Tīrtha and Vana. Each centre had its own special deity, sacred water, Veda, etc. They became centres for the spread of Advaita Vedānta in India, and for the spiritual welfare of the four corners of India that were assigned to them. Other centres were later set up and many of them still exist in India today. In other words, Śaṅkara introduced Buddhist principles of organisation and lifelong asceticism into Hindu monastic life, and provided for the first time some sort of guiding authority to lay down and preach right principles of philosophy and religion.21 The pattern of establishing new modus operandi and foreign ideas within Brahmanical customs and ideology, generally borrowed from competitors and particularly from Buddhists, is a pervasive feature of the Advaita. It is well known that Rāmānuja, a competing Vedāntin of the Viśiṣṭādvaita filiation, pejoratively labelled Śaṅkara a ‘crypto-Buddhist’ (pracchanna-bauddha).22 Following the logic of Advaita’s founding mythology and general dialectic, it should come as no surprise to see the emperor appropriating whatever he sees fit

The beginnings of Advaita doxography  141 within his ‘kingdom.’ After all, if the crown jewels of England are made of precious stones from Africa and other colonies, it does not mean that the monarch is African and so on. On the contrary, it serves as a stark reminder of the extent of his domination. For the colonizer to adopt the valuable products of the colonized is the very motive of colonization. India knew the pattern well before the Muslim and British invasions. The relation of Advaita to Buddhism goes deeper than mere cultural appropriation. A helpful comparison comes from reading it along the lines of the conflictual relationship between the Catholic Church, in its infancy, and Gnosticism. The Protestant Church historian Adolf von Harnack, in his Dogmengeschichte (1885), famously described Gnosticism as “the acute Hellenization of Christianity.”23 In her book on Gnosticism, Karen King quoted a passage from von Harnack which has considerable resonance in the present context of Advaita versus Buddhism: The struggle with Gnosticism compelled the Church to put its teaching, its worship, and its discipline, into fixed forms and ordinances, and to exclude everyone who would not yield them obedience. . . . It had to pay a heavy price for the victory which kept that tendency at bay; we may almost say that the vanquished imposed their terms upon the victor: Victi victoribus legem dederunt. It kept Dualism and the acute phase of Hellenism at bay; but by becoming a community with a fully worked out scheme of doctrine, and a definite form of public worship, it was of necessity compelled to take on forms analogous to those which it combatted in the Gnostics.24 Although von Harnack’s reading is tendentious, since it is subtly formulated as a critique of the Catholic Church from a Protestant perspective, its description of the process of identity formation underwent by the Church in its struggle against the Gnostics is historically insightful. In the present context, one could almost interchange the words of Church and Gnosticism (Hellenism) by those of Advaita and Buddhism, leaving the rest untouched. Von Harnack’s words are even more compelling when compared to those of Whaling: Buddhism lost the battle in that it disappeared from India, but it gained the victory in that it won its way, through Śaṅkara, into the very heart of the Hindu faith.25 Further on, Whaling suggests that without Buddhism there would be no Vedānta.26 Bhāviveka said more or less the same. In other words, one could say that its profound anxieties towards Buddhism drove Advaita to become the mirror reflection of its own fears. The Swiss novelist Max Rudolf Frisch once said, “jealousy is the fear of comparison.” In the long history of their relation, Buddhism and Brahmanism continuously espoused each other’s ideas and practices, yet mutually agreed to disagree and compete over land and souls. Unfortunately, this disagreement was to have devastating consequences for the fate of Buddhism in India, the disappearance of

142  The beginnings of Advaita doxography which is traditionally assigned to Śaṅkara.27 At the same time, banning Buddhism from India might have been the greatest gift Śaṅkara could have ever done to his nemesis. For, by migrating out of India, in a fruitful process of internationalization, Buddhism became what it had always pretended to be, an impermanent dweller, a groundless refugee.28 As for the prospect of a unified Brahmanical land and creed, it never materialized beyond ideology. It is as if implosion and fragmentation were the logical consequences of a vision solely looking inward, cherishing one’s self and one’s view as supreme. It is perhaps a thought worth contemplating by the modern saffron fold known as the Hindutva, which claims Śaṅkara as one of its heroes. Beyond the myth of Śaṅkara the conqueror, which tells more about his devotees than himself, the exegetical genius and intellectual acuity of the man who irrevocably transformed the religious landscape of India is better appreciated when looked at through his own ideas, formulated in his prolific work. A man of his time, he was well versed in the many ideological currents which fashioned the controversies over which he is said to have risen. He took his Brahmanical culture at heart, mastered its scriptures, and sought not only to protect it from collapse and subordination, but also to provide it with all the tools it needed to fulfill its religious and worldly aspirations. Looked at from within the tradition, and not from the outside as an inquisitive and invasive opponent, his many reforms and accomplishments reveal a learned man, who dedicated his life for the benefit of his people. In a nutshell, his message is the superiority of the Vedic knowledge and culture. In the next section, I will examine how Śaṅkara placed scriptures (śabda) above reason (anumāna), knowledge (jñāna) above action (karma), and ultimate truth (pāramārthika-satya) above conventional truth (vyāvahārika-satya), suggesting that the Vedānta stands above all other established views (sarva-siddhānta).

3.2 The hierarchical context of Advaita doxography: above the above The primary purpose of this section is not to provide a full overview of the Advaita, but to introduce some of its doctrinal features which are relevant to better understand the dynamics of Advaita doxographies. Combined with the mythology which surrounds the persona of its founder, these elements should help clarify the nature and purpose of Advaita teachings and thus better orient us to the dialectic and rhetoric at play in the SSS. Naturally, each of the subtopics discussed here could be the subject of an independent monography, thus it should be understood that the following is an effort in summarization. 3.2.1 Scriptures (śabda) above reason (anumāna) Although he expressed many of his ideas through a philosophical language, Śaṅkara is not exactly a philosopher, as often depicted.29 It is in fact not so easy to clearly define the nature of his work, since he uses many genres. Nevertheless, the nature of the sources from which his thought arose and developed is

The beginnings of Advaita doxography  143 indubitable. For, it is upon these sources, the Upaniṣads, the Bhagavad Gītā, and the Brahmasūtra, that he composed his famous commentaries. It is the ideas formulated in these texts which he analyzed, promoted, and defended. Thus, in theory, Śaṅkara is not expressing his own ideas. Though, in practice, many will argue against his interpretations and accuse him of novelty, for his problematic concept of māyā (illusion), for example. That is to say, opponents of Śaṅkara will point out that his reading might be coherent with his own opinion, but that it is not the case with the scriptures which he had set out to comment upon in the first place. If he had been perceived as a philosopher, in other words, as a tārkika, novelty of thought would not have been much of an issue for Śaṅkara. Like Bhāviveka, he could have equated the validity of reason with that of scriptures, and claim that, if reason establishes a proposition, even if it is not supported by any scriptural authority, it is still valid, because scriptures themselves must inevitably be reasoned upon to be established. For Bhāviveka, valid reasoning is āgama. But, precisely, this is not the case with Śaṅkara. His self-proclaimed duty is to clarify a set of pre-existing irrefutable scriptures and not to create a new doctrine or to speculate upon new ideas which have no place in the traditional Vedic corpus. In other words, whereas authors like Bhāviveka and Haribhadra wrote as tārkika-s, openly trusting in the work of reason, composing independent treatises on speculative and logical problems, even while operating within the confines of their scriptural lineage; Śaṅkara, on his part, accepted reason only if guided by scriptures, to clarify a scriptural passage, for example. In brief, Śaṅkara subordinates reason, in its logical and inferential form (anumāna), to scriptures (śabda). Unless, of course, the topic under consideration is not peculiar to the realm of scriptural knowledge but immediately accessible to the intellect through other valid means (pramāṇa-s). For these reasons, I believe that one can unproblematically describe the great Advaitin as a theologian, or, more precisely, as an exegete. His main activity in life was to interpret Vedic scriptures, to defend them against opponents, and to explain them to worthy students. Hirst insisted upon the pedagogical dimension of Śaṅkara’s work and presented him as a “reflective religious teacher,” in contrast to a “systematic philosopher,”30 the latter being more suited for thinkers like Bhāviveka and Haribhadra, though they were also both religious teachers. If one accepts that Śaṅkara’s main activity was scriptural exegesis, then one agrees to say that he was a Mīmāṃsaka. For exegesis, the science of the semantic analysis of sentences (vākya-artha- vidyā), if we remember Jayanta’s definition, is precisely the domain of the Mīmāṃsā. In the next section, I will examine why Śaṅkara’s thought is not simply classified as Mīmāṃsā. But beforehand, it remains to be seen what exactly is the nature of that knowledge which is circumscribed by scriptures. In what way is it not accessible to reason? What is the domain of reason, and who has access to scriptural knowledge if reason is of no avail? First of all, what are scriptures? What do we mean by Veda? In his BAU Bhā,31 Śaṅkara’s definition of the Veda follows the ancient one already given by the Āpastambaparibhāṣāsūtra (1.33): “The designation of Veda applies to both the mantra-s and Brāhmaṇas.”32 This traditional definition was

144  The beginnings of Advaita doxography unproblematically adopted by Jaimini, the author of the Mīmāṃsāsūtra (MS), and his commentator Śabara, by the famous grammarian Kātyāyana, by the Veda commentator Sāyaṇa,33 and by numerous Mīmāṃsakas, like Kumārila,34 in different wording. It is meant to say that the Veda contains two types of statements of different natures. However, there exists no precise definition as to what exactly can be said to be a mantra and a Brāhmaṇa, except that mantras are what is so called by those who know them, and that Brāhmaṇas are the remainder.35 A more helpful definition of Veda might be found in Laugakṣi Bhāskara’s Arthasaṅgraha (AS): “The Veda is a non-human (apauruṣeya) sentence (vākya). It is divided into five: injunction (vidhi), sacrificial formula (mantra), designation (nāmadheya), prohibition (niṣedha), and explanatory passages (arthavāda).”36 Here, we can see that mantric formulas are found in the Veda along with a range of statements of other natures whose functions is either to enjoin (vidhi) action (karma), to designate (nāmadheya) a specific act, or to prohibit (niṣedha) it, or else it serves to provide some general information (arthavāda) about something else mentioned within the Veda. The descending order of enumeration of each of these types of sentences, as found in the AS, already indicates the priority which the Mīmāṃsā will give to each kind when performing its exegetical task. Again, we see that lists are not made at random, but rather tend to follow a given logic, as in doxography. Bhāskara’s definition is helpful in defining the nature of the Veda, according to the Mīmāṃsā, in that it adds the notion of ‘non-human’ (apauruṣeya) agency. This Mīmāṃsā doctrine is central to its claim that the authority of its scriptures is above all other forms of speech. In a nutshell, it maintains that the Veda has no author, not even a divine one, like the Nyāya and others had suggested. The Veda, it is believed, was once perceived by the ancient Vedic ‘seers’ (ṛṣi-s) but is otherwise eternal. Everything within it is in fact ‘constant’ (nitya), ‘fixed,’ or ‘innate’ in meaning,37 and authorless (apauruṣeya), from its individual words to their meaning and even their sound. Based on this peculiar mythology, worshipping the ‘word’ (śabda) itself, the Mīmāṃsā contends that, unlike ordinary human speech (pauruṣeya-vākya), which cannot be trusted, the Veda is faultless (nirdoṣa) and consistent (avirodha). It is truth (satya). As we have seen, Bhāviveka found the claim irrational and attacked it head on at the beginning of MHK 9. The move was strategic. Once this claim is refuted, the Mīmāṃsā crumbles and becomes just like any other system: a mere set of more or less logical propositions. It should be noted that every other Indian system of thought accepted a certain set of scriptures (āgama) as authoritative, yet none other than the Mīmāṃsā maintained that these scriptures were authorless. Eltschinger helpfully defined the general nature and function of these other scriptures in other contexts: Les Écritures sont généralement dites (au moins exclusivement) porter sur l’ordre des états de fait reconnus comme suprasensibles ou radicalement imperceptibles par la tradition, c’est-à-dire sur l’ordre des réalités demeurant inaccessibles aux lumières ordinaires de la connaissance humaine :

The beginnings of Advaita doxography  145 elles informent alors sur les modalités de la praxis éthico-rituelle et ses conséquences eschatologiques, sur des faits de cosmologie ou de théologie, sur des événements passés et futurs. Surtout, avant d’exiger d’elles une doctrine du monde, plusieurs parmi ces écoles (Yoga, Nyāya, bouddhisme, jainisme) demandent à leur personne d’autorité d’administrer une thérapeutique à la douleur existentielle, d’articuler une sotériologie.38 Thus, generally speaking, scriptures are said to provide information of a metaphysical nature, to guide a given conduct towards some kind of a ‘therapeutic’ path, formulated as a ‘soteriology.’ Generally, these scriptures are reasoned upon to fathom their meaning. Yet, as stressed by Chakravarthi Ram-Prasad, liberation in Advaita is no soteriology, since there is no saviour.39 It is also no “cosmic psychotherapy.”40 Albeit, the nuance is so subtle with the latter as to be almost irrelevant. The fact here is that Śaṅkara does not take suffering as a kind of cognitive disease justifying an aetiology and driving one towards a doctor, as articulated in Buddhism or in Nyāya, for example. Rather, for him, the cause of all frustration, the origin of ‘bondage,’ lies in a fundamental ‘misunderstanding’ (adhyāropa) of the revelation. It is removed (apavāda) when the correct understanding of the knowledge conferred by the Veda dawns upon the soul, which is the direct experience (anubhava) of the soul. Advaita liberation is the recovery of universal consciousness, as expressed in the Veda. Only once this consciousness is recovered will peace be established. Thus, the fundamental problem in this narrative is an ignorance (avidyā) of a religious kind, redeemed through a religious authority, the Veda and its (Brahmanical) interpreters, which paradoxically acts as a mirror without acting, being the ultimate indicator of a self (ātman) which has no descriptive indication (nirguṇa). The role of the Veda, in relation to liberation (mokṣa), is somewhat ambiguous, not to say irrational. It lies in between that of a doctor and a saviour, just as Śaṅkara’s notion of māyā (illusion), being neither ‘is’ (sat) nor ‘is not’ (asat), neti neti (not this nor that). In the end, like his gnostic competitors, Śaṅkara is adamant that liberation is not caused, else it would not be everlasting. However, here, as in the Mīmāṃsā, one needs not reason upon the Veda to gain some derivative knowledge (jñāna), in the sense of validating the meaning of the Veda, against other means of knowledge and so on, or of inferring a hidden or secondary meaning. One only needs to be ready to receive the Vedic instructions and be immediately transformed by its negation of ignorance (avidyā), as if by command (codanā): “you are that!” (tat tvam asi);41 according to a set of preliminary preparations and indications already found within the Veda. In brief, all one needs to know about heaven (svarga) or liberation (mokṣa), and the way (mārga) to get there, is in the Veda. This is its domain. The Mīmāṃsā system presents a kind of circular thinking where coherence (avirodha) can only be established from within, by accepting its numerous doctrines outright, like the eternity of sound (śabda), the superiority of scriptural authority (śabda-pramāṇa), and so on. Like any theology based on revelation, it requires an act of faith to begin with, even if at the cost of reason. And, like similar theologies, its coherence is threatened by an outside reasoning which rejects its internal

146  The beginnings of Advaita doxography premises (tattva-s). This explains the long resistance of various Vedic proponents to the work of reason; their desire to separate themselves from those which they pejoratively labelled as tārkika-s. It also explains Śaṅkara’s desire to apologetically defend his theology using the argumentative tools of his opponents, but not to validate his own system. An ironical mind could argue that Śaṅkara’s theology requires no supporting arguments. As observed by Bronkhorst: Paul Hacker (1968: 120 [214] ff.) has expressed surprise about the fact that Śaṅkara offers few if any rational arguments in defence of monism, which is yet a central part of his philosophy. Hacker looks for a solution in a hypothesis concerning the biography of Śaṅkara (first Yogin, then Advaitin). However, Śaṅkara the Mīmāṃsaka had no need for proofs of monism. What is more, knowledge derived from the Veda should be unobtainable by other means. This includes knowledge of monism.42 This passage serves to illustrate the problems which come out of wanting to frame Śaṅkara as a philosopher. One would expect him to support his ideology with arguments. But he needs not do so, for his authority derives not from reason (yukti), but from the words (śabda) of the Veda, which is a non-human (apauruṣeya) revelation serving as the only means of knowing the only valid path to the only possible liberation. As Bronkhorst explains, in this context, “scripture legitimizes the use of reasoning by providing cases to which it may be applied.”43 Thus, reason is legitimate in the confines dictated by scriptures. In a perfectly circular way of thinking, the Veda is said to exist from all eternity solely to explain itself. As such, it is perfectly coherent and faultless, and whatever is found therein is true, point. Reason, if needed, is secondary. What, exactly, is found therein? What can one know through the Veda? According to the Mīmāṃsā, it is dharma, that which is the object of enquiry of the system, as stated in the opening verse of Jaimini’s MS (1.1.1): “Thus begins the enquiry into dharma.”44 This dharma is subsequently defined as a “purpose indicated (lakṣaṇa) by a command (codanā).”45 In other words, dharma is a purposeful conduct enjoined by a Vedic command. It is exactly this dharma which is the sole domain of the Veda, according to the Mīmāṃsā. This is to say that the only means of knowing (pramāṇa) dharma are the words (śabda) of the Veda. For other domains of knowledge, however, other means can and should be used. But, again, what is the use of conforming to the purposeful conduct enjoined by Vedic commands? Ultimately, it is the promise of heaven (svarga) for the one who desires it (svarga-kāmaḥ), if he should sacrifice (yajet) accordingly; however, more mundane desires are also said to be fulfilled by suitable Vedic rituals, such as having children, acquiring wealth, and so on.46 Thus, a Mīmāṃsaka’s aspirations are generally ‘this-worldly,’ if one considers that heaven is simply an extension of this world, a higher abode. This is where Śaṅkara will considerably alter the picture. The scripture which preoccupies him, on which he composed his main commentary (bhāṣya), his

The beginnings of Advaita doxography  147 magnum opus, the Brahmasūtra, contrary to the MS, begins by indicating an altogether different focus of enquiry: “Thus begins the enquiry intro brahman.”47 At this point, we can start to appreciate the way in which Śaṅkara sought to distinguish his ambitions from those of his fellow Mīmāṃsakas. It is not that he rejected dharma, but he subordinated it to the ultimate goal of ‘knowing brahma.’ Bronkhorst spoke of Śaṅkara as nothing less than a “palace revolution” inside Mīmāṃsā.48 The revolution lies in this, that whereas the prior focus of the Mīmāṃsā was on dharma, the innovation of Śaṅkara is to subordinate the former to his enquiry into brahman. This is no small feat. Whereas the prior Mīmāṃsā had originally little concern for ideas such as bondage (bandha), rebirth (punarjanma), and liberation (mokṣa), but rather focused on dharma, as a code of conduct indicated through Vedic injunctions (vidhi/codanā), where karma appeared somewhat as a transactional ritual currency, Śaṅkara revalues the nature of the Veda as an emancipating teaching, teleologically oriented towards the knowledge of brahman, the only knowledge able to free one from the bondage of karma. Śaṅkara did not change the prerequisites to knowledge from those of the former Mīmāṃsā. He even added additional ones. As explained in his Upadeśasāhasrī (Chapter 1), the one who is fit (adhikārin) to receive the liberating knowledge of the Veda is a pure male Brahmin, who approached his teacher in a proper way and was scrutinized with respect to his caste, occupation (or ritual action), conduct, knowledge (of the Veda), and family; all conditions directly or indirectly related to birth.49 Moreover, he must be a renouncer (saṃnyāsin) fulfilling four additional conditions which Hirst explains as having: (i) (ii)

Discrimination between eternal and non-eternal matters Dispassion towards the enjoyment of the results (of actions) here and hereafter (iii) Accomplishment of practices such as mental tranquility and control of the senses (iv) Desiring liberation.50 Thus, the candidate should more or less be a Brahmanical male Mīmāṃsaka knower of the Veda, desirous of directing his attention beyond the realm of ritual activity (karma). For sure, only a twice-born (dvi-ja) male can study the Veda, a masculine member of one of the three upper varṇa-s.51 In BSB 1.3.34–39, along the lines of Bādarāyaṇa, Śaṅkara insists that the Veda is strictly forbidden to śūdra-s and women, restating the same exclusions which were ridiculed by Bhāviveka in his reply to the uttarapakṣa of MHK 9.11, the one which generated the longest pūrvapakṣa of the chapter. Indeed, gender and caste exclusions are no small affairs here. We will conclude on this last point. By adding together all the listed prerequisites to Vedic liberating knowledge, there logically remains but a tiny minority of eligible potential candidates. Yet, at the same time, Śaṅkara argues tooth and nail that Vedic knowledge is the only way to liberation and that strictly nothing else, especially not independent reasoning, could ever lead one to the highest goal. One is thus led to wonder if

148  The beginnings of Advaita doxography Śaṅkara purposefully wanted to move his tradition in the opposite direction of the Buddhist Mahāyāna of his time, which was advocating a way (mārga) for all (sarva-sattva).52 As a marketing strategy, exceptionality will ever remain a lure for the elite. The contrast with Buddhism is remarkable. For example, advocating the awakening potential within everyone,53 Buddhism conceived early on the hypothetical category of the lone Buddha (pratyeka-buddha), one who could reach enlightenment by oneself, alone, in distinction to one who would do so through hearing (śravaka)54 the dharma of the Buddha. Thus, Buddhists could imagine, at least theoretically, that one obtains nirvāṇa without being one of their followers. But this is inconceivable within the Advaita, for whom ‘freedom’ is ironically the exclusive property of a religious authority, the caste-exclusive Veda and its priests, like a peaceful dove in a sacred golden cage. Śaṅkara’s path, in comparison to the Mahāyāna, is a way for no one but a tiny minority among priests who, by birthright, claim to be the ‘seat of omniscience’ on earth. In this context, the accusation of Bhāviveka, who qualified such Brahmins55 as ‘shameless’ (MHK 9.1: anapatrapa), takes all its flavour. I read these doctrinal differences as further escalations of a long history of ideological confrontations between Brahmins and their opponents, especially but not exclusively Buddhists, for the control of ritual power.56 With Śaṅkara, furthering that which had begun under Kumārila, the war becomes total. It is now waged on all fronts. A Brahmanical elite emerging from the fold of the Mīmāṃsā now claims for itself not only its right to ritual performance, which it never abandoned, but what was until then a domain in which it had little interest, liberation (mokṣa).57 Śaṅkara’s hagiographers may have properly understood his ‘conquest’ after all. When transposed in the context of social conflicts, the hermeneutical strategies of the ‘universal teacher’ are meant to subjugate. They are designed to culminate in domination, by promoting a myth of blood purity and a karmic mystic of social exclusion.58 Not only are the ‘lowly’ banned from any dignified social status in life, but they are barred entry to heaven (svarga) at death, and Śaṅkara oversees that they forever remain in bondage (saṃsāra) by preventing them from even hearing about the way to release (mokṣa). Surely, as discussed in the chapter on Bhāviveka, there are sinister sociohistorical dynamics at play in this new exegetical shift, going beyond the realms of philosophy and theology. 3.2.2 Knowledge (jñāna) above action (karma) As we have just seen, Śaṅkara redirected the focus of the Mīmāṃsā from dharma to brahman, from ritual action (karma) to knowledge (jñāna). To explain this shift, he justified himself by claiming that whereas Mīmāṃsā focuses on the exegesis of the ritual portion (karma-kaṇḍa) of the Veda, the Vedānta is rather dedicated to the study of its knowledge segment (jñāna-kaṇḍa), as found in the Upaniṣads, the Bhagavad Gītā, and the Brahmasūtra, together forming the Vedāntic scriptural corpus designated as the Prasthānatrayī (the Three Sources). Śaṅkara called his form of exegesis the śārīraka- or brahma-mīmāṃsā. The practice of using the terms of pūrva (prior) and uttara (later) Mīmāṃsā to unify, while distinguishing,

The beginnings of Advaita doxography  149 the Mīmāṃsā and the Vedānta, was unknown to Śaṅkara and appeared later in the history of the tradition. Discussing Parpola’s hypotheses,59 Bronkhorst explained that the terms seem to stem out of an erroneous analysis of the titles of the Pūrvamīmāṃsāsūtra and Uttaramīmāṃsāsūtra, which are but two sections of one single work called Mīmāṃsāsūtra.60 Originally, for other Vedāntic thinkers,61 like Bhāskara, Yāmuna, and Rāmānuja, the distinction was meant to indicate that the former (pūrva) study of the Veda had to precede the latter (uttara) in time, in the course of one’s curricular study, and not that the latter study was the ultimate end-goal of the Veda (veda-anta). Following this line of thinking, Bronkhorst argues that the Mīmāṃsā was not originally, if it ever was, interested in the kind of Vedic exegesis formulated in Śaṅkara’s terms. It always kept its ritual focus. It was the emerging Vedānta exegesis who sought authority by adopting the means of, and thus claiming its belonging to, an established exegetical tradition, while introducing new game-changing doctrinal elements, like bondage (bandha) and liberation (mokṣa), gathered from a section of Vedic compositions (the Upaniṣads) which were originally meant as a criticism of the very Vedic path propounded by ritualists following the karmamārga, inspired from ideas originating from outside the Vedic fold (Jainism and Buddhism being their most successful proponents). This new form of Vedāntic exegesis strategically positioned itself as the highest (uttara) Vedic interpretation, thus becoming the ultimate Brahmanical achievement and claiming to be the only valid path to liberation among all competing religious communities. In doing so, it ensured the central role of the Veda and of their own interpretation of it in Indian religious life. Bronkhorst noted, Vedānta conceived of as Mīmāṃsā is not a form of philosophy which uses various means of knowledge to establish its positions; quite on the contrary, it is Vedic interpretation which starts from the assumption that knowledge correctly derived from the Veda cannot but be correct itself.62 Hence, Vedānta is not philosophy but theology. It claims liberating knowledge as the product of its exclusive revelation and turns freedom into a by-product of the sole Veda. This interpretation developed even while the Veda themselves had originally no interest in liberation. They rather proclaimed an altogether opposite worldview, focusing on fulfilling respectable and more or less universal worldly human goals, through ritual performances proceeding as magical acts, binding cosmic energies to the officiating will, through a skillful process of manipulations of symbolic correspondences, generally combined to a petition of favour to divine agents. How is one obtaining the salvific knowledge promoted by Śaṅkara once one is entitled to it? How does this knowledge accomplish what ritual business (karma) alone supposedly could not? Once more, the path laid forth by the ‘great teacher’ amounts to a rehashing of the jñāna-mārga of its ‘enemies,’ cast in Brahmanical terms. Ram-Prasad explains that Advaita generally acknowledges three means of

150  The beginnings of Advaita doxography obtaining liberating cognition, the same three which we saw previously in Buddhism and Jainism: hearing, thinking, and contemplation.63 In this regard, RamPrasad quotes the Advaitin Dharmarāja: What is called hearing (śravaṇa) is that mental act which is conducive to the determination of the purport of the Vedānta [texts] on the non-dual brahman. What is called thinking (literally: ‘minding’) is that mental operation which, when there is a doubt that there could be conflict between meaning determined through text and [that determined through] the other means of knowledge, generates cognition of the nature of argumentative analysis (tarka) which assists in its (the doubt’s) removal. What is called contemplation (nididhyā-sana) is that mental operation which, when consciousness is dragged towards objects by a beginningless bad inclination, assists in dragging it away from objects and establishing it in the matter of the self (ātman) (VP IX. 22–4, p. 160).64 This summary perfectly captures the practice of jñāna-yoga (intellectual exertion) mentioned previously. The definition provided here, like the previous ones, binds the process of insight into truth to textual hermeneutics. Obviously, in the case of the Vedānta, the texts to be contemplated upon are those of the revelation (śruti), the Prasthānatrayī. Unfortunately, Dharmarāja’s passage is not explicit as to how insight is actually produced, but it seeks to shed some light on the nature of the third step (nididhyāsana), which resembles that of all jñāna-yogin-s. Here, the term nididhyāsana, which is an action noun derived from the desiderative of the root dhyā (thinking or meditation), is said to refer to a sustained recollection pertaining to the content of what was previously heard and reflected upon, where that meaningful content is used as a meditation support receiving the full attention of consciousness, freed from other mental disturbances. Furthermore, Walter Menezes explains, Nididhyāsana is a technical term used in [Vivekacūḍāmaṇi] VC and also in Advaitic literature with the meaning “constant meditation”, “repeated concentration” of the truth arrived at the result of śravaṇa and manana. There are other Advaitic literature that give importance to nididhyāsana. According to Vedānta-sāra nididhyāsana is conceived as, “a stream of ideas of the same kind as those of Brahman, the One without a second, to the exclusion of such foreign ideas as those of the body etc.” (VS: 192; Rambachan 1991: p. 109). BṛU emphasises the superiority of nididhyāsana in this way: “The Self, my dear Maitreyī, should be realized, should be heard of, reflected on and meditated upon. By the realization of the Self, my dear, through hearing, reflection, and meditation, all this is known”. Nididhyāsana is a “meditation-in-absorbed concentration” in which sacrificial fire (meditation) inflamed by “That, thou art” has friction with its woods, namely mind and Om and produces the knowledge comparable to the power of a scorching sun that is capable of dispelling every darkness of ignorance (ajñāna).65

The beginnings of Advaita doxography  151 Menezes makes it clear that the content of the nididhyāsana meditation is the stream of thought revolving around the semantic of ātman-brahman. It is that semantic which was previously refined through hearing and analyzing. It is captured within such ‘great sayings’ (mahāvākya-s) as “that, thou art” (tat tvam asi), which serve as the spark, enlightening the mind with the knowledge contained therein. In brief, previous studies (textual hermeneutics) generate a bulk of intellectual knowledge which clarifies the meaning of a ‘mantric formula’ (tat tvam asi and so on) to be contemplated upon until its gnostic content burns away previous misconceptions (avidyā) and leaves one with the direct experience (anubhava) of the knowledge of the true nature of consciousness (ātman-brahman). In other words, liberating knowledge is the awakening of self-consciousness. The threefold process is here obviously linked to the Upaniṣads, establishing its uncontestable authority. Such clarifications are helpful to better understand the nature of jñānayoga ‘praxis,’ even beyond the realm of the Advaita.66 Ram-Prasad explains that Advaita developed two lines of interpretation of the process leading to liberation. The Vivaraṇa stream contends that receptivity to truth, in other words, the fact of being adhikārin, and the very power of that truth are liberating in themselves, while the Bhāmatī line stresses the need of philosophical analysis. But both are clear that liberation is not actually caused by knowledge, else being created it would not be eternal. No, liberation is knowledge, the cessation of ignorance. Nothing is created, but misperception, or superimposition (adhyāropa), is removed. The nuance is significant. For, according to the proponents of the jñāna-mārga, the main problem with the karma-mārga of the prior Mīmāṃsā, as mentioned by Bhāviveka, is that it operates only through causal means, and hence final liberation is out of its reach. It is significant, again, to notice that Śaṅkara’s criticism of Mīmāṃsā and Yoga, as discussed by Halbfass,67 reformulates the one of his Madhyamaka predecessor, Bhāviveka. In fact, it is a logical conclusion to draw, from the point of view of a jñāna-yogin, a philosophizing theologian. Karma, be it in the form of yajña (sacrifice), āsana (ritual posture), samādhi (mental enstasis), or whatever means (yoga), partakes of the very causal nature of saṃsāra. Hence, it cannot bring an end to itself. Yet, all these means can surely act as facilitators or supports of religious life. Through them, one is made ready (adhikārin) to receive the liberating knowledge (jñāna) which alone escapes the binding chains of causality (karaṇatva). Immediate knowledge (anubhava), however, is not caused by any means, for it is said to be the mere disappearance of that (avidyā) which was never really there to begin with, but which was a mere ‘illusion.’ By immediate knowledge, nothing is created and nothing really ceases, neti neti (not this nor that). In that way, Śaṅkara, like the other jñāna-yogin-s, does not negate any validity whatsoever to rituals or to yogic practices, but he subordinates such activities to a higher end, by assigning them a supportive secondary function. The relation of knowledge (jñāna) to action (karma) might be worth meditating on by those who doubt the benefits of ‘hermeneutic praxis,’ or textual studies, and favour more ‘pragmatic’ means of tangible success (siddhi), a tendency often found within the followers of these ancient paths, today like yesterday. There is nothing new under the sun.

152  The beginnings of Advaita doxography In the end, such dialectic serves to clarify the nature of liberation and to make explicit the nexus of interactions involved in religious life, that between moral discipline, ritual activity, conceptual doctrinal studies, and non-conceptual realization, culminating in liberation from bondage. The motor engine of the religious vehicle (yāna) is hermeneutics (tarka/mīmāṃsā), fuelled by scriptures (āgama), supported by the armature of conduct (cāritra) and bound together by rituals (karma), indicated through dharma, itself a by-product of scriptures requiring hermeneutics. The path (mārga) goes along the conventional (vyavahāra) inroads of life which are to be transformed along the way into the ultimate domain, wherein one acquires the perfect sight (samyag-darśana) of pure consciousness, where nothing else remains, not even a path or a vehicle. Similarly, Ram-Prasad insisted, “the way to liberation lies in and goes through unliberated life.”68 What Advaita doxographies are laying out is thus a systematic and gradual process of ascension through the various expressions of that ‘unliberated life,’ using opposing views as many ladders to be examined and corrected through the peculiar exegesis of the Uttara-Mīmāṃsā, until all conventional views (sarva-siddhānta-s) subside into the ultimate experience. 3.2.3 Ultimate truth (pāramārthika-satya) above convention (vyāvahārika-satya) In the previous chapters on Buddhist and Jaina doxographies, I insisted that their authors understood the views listed in doxography as expressing a conventional (vyāvahārika) worldview (darśana), in contrast to an ultimate (pāramārthika) truth which, by nature, lies beyond the realm of language and reason altogether. Nevertheless, I also explained how each thinker perceived his traditional method as the most suited to navigate the endless ocean of conventions, saṃsāra. Each path was promoted as a kind of cognitive therapy, explicitly referring to the semantic of medicine and healing, said to lead one to the reestablishment of the perfect eyesight (samyag-darśana). The dialectic of these doxographies skillfully managed to use the poison of heterodoxy within a therapeutic process by which the mind is trained in discrimination (viveka) and becomes somewhat vaccinated against error (mithyā-jñāna), a process recalling the ancient jñāna-mārga of the Sāṃkhya philosophy. It is as if the doxographical study of philosophy was understood as purifying sight (darśana) from doṣa, a word conveying the double meaning of ‘fault’ and ‘disease,’ suitable in relation to ‘mistaken views’ (mithyā-darśana-s), and to strengthen the student’s certainty (niścaya)69 against the downfall of doubt (saṃśaya), having turned one’s mind into a pure and unbreakable diamond (vajra). At the same time, such dialectical studies confer the sharpness of the diamond unto one’s debating skills, a much-needed weapon in a context of ongoing sectarian rivalries often settled at royal courts, in a language couched in legal terms.70 This ‘diamond-view’ (vajra-darśana), which the authors rather referred to as samyagdarśana, is the ultimate intellectual and religious achievement conceivable in the realm of conventions, a much sought-after weapon (śastra). Skillfully wielded,

The beginnings of Advaita doxography  153 the power of this ‘sharp sword of wisdom’ (MHK 1.11: prajñāniśitanistriṃśa) can be projected over the religious, philosophical, and political domains. Its holder is an āpta, a trustworthy authority, the very mouthpiece of the āgama (scripture) whose words are commands. One thus empowered is an ārya, or arhat, in its noblest sense. His wisdom crown inspires respect. It is exactly along the lines of this jñāna-mārga which we must interpret Śaṅkara’s efforts to cast the Advaita as samyag-darśana. Discussing how Śaṅkara perceived the position of the Advaita in relation to other schools, Hirst observed, This is not just viewed as one school among many schools, comparable with Sāṃkhya or Nyāya, with a well-developed lineage of pupils and teachers. Śaṅkara nicely distinguishes between the viewpoints of these other darśanas (literally ‘seeing’, hence ‘schools’, ‘viewpoints’) and the samyagdarśana of Advaita. The samyagdarśana is the perfect realization of non-duality, the true teaching of Advaita, the correct school that passes this on. It conveys the saṃpradāya, the correct interpretation given by Nārāyaṇa, the one who unites personal and impersonal, scripture and interpretation, in the perfect realization itself. Because it is so-based, it is not prone to the inconsistencies and contradictions of schools that are based in the mere reasoning of human teachers.71 In brief, Śaṅkara clearly distinguishes between the ‘right view’ (samyag-darśana) guaranteed by his own system and the ‘wrong views’ (mithyā-darśana-s) of others. The consequences of this assessment are coherent with the overall self-perception of the Advaita as being the only path to liberation. Thus, in Advaita doxography, all views besides the Advaita are ‘wrong,’ though some are better than others. At the same time, just as with the previously examined doxographies, all views (sarva-darśana-s) listed in Advaita doxographies, including the Advaita teachings, are mere conventional expressions. The ultimate realization occurring through samyag-darśana is itself no worldly perception but an altogether different kind of superior knowledge. Along the lines of the Muṇḍaka Upaniṣad (1.1.4),72 which states that there are two kinds of knowledge, a higher (parā) and a lower (aparā) one, Śaṅkara distinguishes between a conventional knowledge which includes all the sciences listed in the introductory chapter of the SSS, the four Vedas, phonetics, rituals, grammar, etymology, metrics, astrology and so on, and the higher knowledge of the self (ātman-brahman). Thus, even the Veda is classified along the conventional means. Hirst quoted a passage from the BSB (4.1.3) where Śaṅkara defends this position against an opponent who appears to be sceptical about this subordination of even the Veda to a higher knowledge: If you say that, if perception, etc. become non-existent then this will lead to the undesirable conclusion (prasaṅga) of the non-existence of śruti even, [we say] no problem. Because this is what is acknowledged. From the text beginning with “There a father is no father” to “The Vedas are no Vedas” (BAU

154  The beginnings of Advaita doxography 4.3.22), the non-existence of śruti in enlightenment (prabodhe) is acknowledged by us.73 Therein, the discussion comes eminently close to the one taking place in Nāgārjuna’s famous MMK, Chapter 24, where a Buddhist opponent accuses Nāgārjuna of destroying the Bauddhadharma with his teachings on emptiness (śūnyatā). Śaṅkara’s answer, by claiming that his teachings are in fact the very meaning of the Veda, uses the same dialectical strategy as the renowned Madhyamaka śāstra-ācārya. Both are turning the allegations of the opponent against him by suggesting that it is he who does not fathom the true meaning of scriptures, not them. If one were to accept the objection of the opponent, then liberation would become impossible. In turn, just as emptiness is empty of itself, being a purely conventional mean, the Veda is no Veda. As a conceptual unit, a mere particular (viśeṣa), the Veda is not brahman. At best, it is the ‘only’ finger properly pointing at the moon, the radiance of Vedic knowledge (jñāna), which the Veda cannot contain within itself. Here, for the sake of the demonstration, adopting the means of tarka, Śaṅkara becomes a philosopher (tārkika), subordinating every science to truth alone. Even the Veda is mere ignorance (avidyā) in comparison to brahman. But this nuance applies only within the perspective of the absolute, what Śaṅkara expresses in this passage by saying “in enlightenment.” From the conventional standpoint, the Veda and its Advaita exegesis are the ultimate viewpoint, the samyag-darśana. Thus, to repeat, Advaita doxographies, like their cousins from other traditions, are dealing only with conventions, using them as a propaedeutic to establish the ultimate view, the immediate experience (anubhava) of the self (ātman-brahman). The process of moving from the conventional to the ultimate constitutes the yoga propounded by Śaṅkara. In essence, its method resides in the hermeneutical praxis of adhyāropa-apavāda, the ability to eliminate (apavāda) superimposition (adhyāropa) through a scripturally informed discrimination (viveka). Strictly speaking, it is this discriminative practice which makes Advaita a jñāna-yoga, of the likes of Sāṃkhya’s, Bhāviveka’s, and Haribhadra’s jñāna-mārga. Hirst elegantly captured the exegetical process at play in Advaita’s discriminative yoga: There is no dichotomy between the cosmological and the psychological. The crucial point is to use the structures of this world, ordered by scripture, to realize the self within all. Then the endemic misidentifications with body and mind will be removed, as the processes of interiorization and rejection deepen the preliminary teaching tactics. They both further the pupil’s introspection and formalize the dissatisfaction that renounces the things of the world.74 The actual exegetical method of apavāda, making use of Mīmāṃsā-like sentence analysis, applied to Vedic statements about the nature of the self (ātman), to remove all identification which is not ultimately relevant to its nature, as in the mahāvākya “tat tvam asi,” can be examined within the easily accessible Vedāntasāra (VS) of Sadānanda. It need not be presented here. What is worth noting, however, is that the

The beginnings of Advaita doxography  155 VS chapter on adhyāropa uses the views of competing darśana-s as typical examples of superimposition (adhyāropa). Thus, within the Advaita tradition, starting with Śaṅkara, adhyāropa applies both to internal cognitive delusion (moha) and to the external views (darśana-s) of non-Advaita systems. Only once all impositions are removed is one said to experience (anubhava) the self as the dṛśi, described by Hirst as “the witness, that context-free state of consciousness which simply is the self, nonsublatable, ultimately self-validating (svapramāṇaka).”75 Advaita doxographies, being the fruit of such dialectics, are hence intended as a yogic training, suited for jñāna-yogin-s, a spiritual exercise designed to recognize superimposition and to eliminate it. Through them, one is led to adopt various philosophical positions, to examine their inner dynamic (made of tattva-s and so on), and to move on to an ever-subtler (sūkṣma) and appropriate (kṣama) mental disposition, established (siddhānta) by scriptures, but moving beyond (para). Seen in this context, Advaita doxography, as the other types previously examined within these pages, can be said to be ‘performative’ texts. They are meant to be performed in contemplation (nididhyāsana), once having previously heard and reflected upon the teachings found within scriptures. What remains to be seen, now, is how the SSS serves to illustrate the previous discussions and if Śaṅkara himself could have been at its origins.

3.3 Reconsidering the SSS: authorship, pedagogy, and doxography76 The first and only critical edition of the Compendium of Tenets of All Views, commonly known in Sanskrit as the Sarvasiddhāntasaṅgraha (SSS),77 was produced over a century ago, in 1909, by Professor M. Raṅgācārya,78 based on five manuscripts.79 The scholar mentions that the colophons, at the end of each chapter of the SSS, give the longer title Sarvadarśanasiddhāntasaṅgraha.80 The reason for this alternance in titles is unclear. However, we have seen that, at this period, the term darśana (view, doctrine, opinion) had become a generic appellation for speculative systems and was commonly used to designate similar doctrinal compendiums. The term siddhānta (established end, conclusion, tenet) serves to specify that the text deals with the main tenets of these systems, as in an epitome. Relatively less common in similar doxographical titles than the term darśana, it is possible that the word siddhānta was kept as a more useful designator when the title was shortened. In brief, the SSS counts twelve chapters (prakaraṇa-s)81 summarizing eleven doctrines, if we list the two types of Mīmāṃsā treated separately as independent systems. The text holds within 537 concise verses of simple facture. No explicit refutation of each doctrine is formulated, but an implicit hierarchical progression suggests that each following system stands closer to truth, as conceived by the Advaitins, than the preceding one. Since Raṅgācārya, although the text has been briefly debated in the early 1920s (Jacobi 1921; Liebich 1923) and subsequently classified as one of the many spurious literary attributions of the famous theologian Śaṅkara, it received little attention. The exception to the rule is Paul Hacker who, by dedicating a whole three and

156  The beginnings of Advaita doxography half pages to the subject, reprinted in Kleine Schriften (Hacker 1978: p. 55–58),82 struck a final blow to its general relevance by refuting the assertions of Raṅgācārya in favour of the authorship of Śaṅkara. Hacker’s remarks tend to be ignored by more recent scholarship, which rather shortcut their enquiry by referring to Wilhelm Halbfass’s general observations on Sanskrit doxographies, found in India and Europe (1988). One interested in reviewing the scholarship on the SSS thus finds oneself in a labyrinth of scattered footnotes. My previous article on the subject (2019) already analyzed what has been said concerning both the authorship and the possible dates of the SSS, while contributing my own observations in tune with what we know so far about Śaṅkara’s work in particular and the genre of doxography in general. In brief, I suggest that doxography is in tune with Śaṅkara’s pedagogy and that this factor should influence our reception of the SSS, regardless of whether Śaṅkara is the author. 3.3.1  The general authorship of Śaṅkara However complex the issue of the authorship of Śaṅkara is, contemporary scholars tend to consider a relatively limited number of titles as genuine, in comparison to the four hundred claimed by the traditional account. To facilitate the presentation of the entire Śaṅkarian library, including works of debatable authorship, S. K. Belvakar gave a threefold division of its numerous titles.83 The first category regroups the commentaries, consisting of primary interpretation of basic texts such as the vivaraṇa-s and ṭīkā-s. The second group contains the hymns, the poems, the metric incantations, and praises of the gods (stotra, stava, and stuti), while the third consists of independent compositions, treatises, and compendia. It is in this last category that the SSS is classified, along with other titles such as the Upadeśasāhasrī, the Vivekacūḍāmaṇi, Ātmaboddha, Aparokṣānubhūti, and Śatāślokī. However, other scholars frequently choose not to classify the SSS within any category of Śaṅkara’s work.84 Following the lead of Hacker, Mayeda, and Vetter, who applied a complex set of philological criteria to determine Śaṅkara’s authorship, the only texts accepted by contemporary scholars as unproblematically coming from the hand of the great Advaitin fall within the first category. They include the famous Brahmasūtrabhāṣya (BSB), the commentaries on the main Upaniṣads,85 and one commentary of the Bhagavad Gītā. The Pātañjalayogaśāstravivaraṇa (PYŚV)86 remains a matter of contention, even if valuable arguments have been made to support Śaṅkara’s authorship. Of the two other categories delimited by Belvakar, with the exception of the Upadeśasāhasrī, all the texts are either rejected as inauthentic or debated.87 3.3.2  The authorship of the SSS Besides the traditional opinion, when it comes to the authorship and dating of the SSS, the present review of literature has met only two early proponents of the authorship of Śaṅkara: M. Raṅgācārya and B. Liebich.88 Patel Maheshkumar Gokulbhai in a more recent PhD thesis on the SSS (submitted at the University of

The beginnings of Advaita doxography  157 Baroda in 2009) attributes the authorship of the SSS to Śaṅkara. However, on the question of authorship, his thesis does not engage with the opinions of previous scholarship and only presents a list of six external and eleven internal pieces of evidence which, by their general weakness, do not add anything significant to the discussion.89 Hence, although the thesis has other qualities, I did not reproduce its arguments. My article90 already summarized the arguments in favour of the authorship of Śaṅkara that have been articulated by scholars prior to the rejection of the said authorship by Hacker, which became authoritative in general scholarship and which I also summarized. One interested in the details is welcome to consult the article. For the sake of concision, I will only present here the main conclusions of my review. 3.3.2.1  Arguments in favour of the authorship of Śaṅkara RAṄGĀCĀRYA

In brief, Raṅgācārya believes that the authorship of Śaṅkara is supported by a set of solid evidence. In a half-veiled manner, he suggests that those who have doubts might do so out of some general mistrust towards ‘Indian tribesmen.’91 This attitude of suspicion concerning Western Indological scholarship might have contributed to the Indian professor’s rejection of evidence based on stylistic analysis and his overall prognosis concerning the authorship of the SSS. LIEBICH

Liebich agrees that the authorship of Śaṅkara is not definitely proven by his arguments, as it would require a more careful comparison, especially of the Vedānta chapter in the light of Śaṅkara’s Bhāṣya – a task which Hacker undertook almost sixty years later. Yet, his arguments entertain the possibility that the great Indian philosopher Śaṅkara was the author of the text. His most important discussions in this regard, though not conclusive in themselves, are perhaps (1) the suggestion that the fourteenth-century author of the SDS borrowed from the SSS on several occasions and (2) that the number of padārtha-s (6) and guṇa-s (24) attributed to the Vaiśeṣika supports the hypothesis that the SSS was written before the tenthcentury Bengali scholar Śrīdhara. 3.3.2.2  Arguments against the authorship of Śaṅkara HACKER

Concerning the arguments aligned by Raṅgācārya and Liebich, Hacker92 makes no case for the arguments of authority. He spends no time on the general opinion of Liebich, on the grounds that the Vedānta chapter of the SSS was not taken into consideration in his article, while, for Hacker, it is the only one of importance in discussing the alleged authorship of Śaṅkara. Within the arguments presented by

158  The beginnings of Advaita doxography Hacker, four arguments highlight doctrinal developments known to have emerged after Śaṅkara. Together, they contradict the argument of Raṅgācārya and Liebich, stipulating that, within the SSS, one finds no account of thought systems appearing later than the period of Śaṅkara. However, Raṅgācārya and Liebich seem to refer only to the systems of thought used as chapter headings within the SSS. In any case, Hacker’s indications point to serious flaws in their judgment. What is more, Hacker limited his criticism to what he knew best, the Vedāntapakṣa chapter. But, looking at the Patañjali-pakṣa-prakaraṇa, one can identify another set of anomalies which tend to discredit the hypothesis that the author of the BSB, Śaṅkara-Bhagavatpāda, is the same one who compiled the whole SSS. 3.3.2.3 New arguments against the authorship of Śaṅkara: problems of the Patañjali-pakṣa93 It has been acknowledged that Śaṅkara was not completely hostile to the Yoga presented by Patañjali. His Pātañjalayogaśāstravivaraṇa (PYŚV), in the plausible case that it is indeed from his hand,94 bears witness to his interest in the system. This is conceivable if we consider that more recent scholars tend to doubt the authorship of Śaṅkara’s commentaries on texts other than from the Prasthānatrayī,95 usually accepting only twelve of them as authentic.96 If, indeed, Śaṅkara commented on Patañjali’s text, he would have held the śāstra in high esteem, as an authoritative source of knowledge. This might serve to explain the high position of the yoga system within the SSS, itself a testimony of its status within the Advaita community. Hacker even went as far as to suggest that Śaṅkara had been a yogin in his early career,97 before becoming an Advaitin.98 For his part, Govind Chandra Pande, aligning the work of Śaṅkara with Vācaspati Miśra, argues that the Vedāntic rejection of yoga is only partial and concerns mainly the dualistic and realistic elements of Sāṃkhya, whereas yoga’s practical method tended to be harmonized with the early Advaita doctrine, i.e. the one present in the BSB.99 However, my paper showed that, on more than one occasion, the Patañjalipakṣa-prakaraṇaof the SSS mentions elements which are not in line with the YS, but which rather appear to bear the mark of later doctrinal developments. The exact belonging of such references, likely related to a post-Śaṅkarian haṭha-yoga context, would have to be sorted out in a later research. It suffices here to note the general theme of the problematic mentions which have not been thoroughly examined by scholars yet: 1 2 3 4 5

The problematic definition of kriyā-yoga: (SSS 10, 15). The mention of āsana-s absent in the YS and in Vyāsa: (SSS 10, 34). The mention of prāṇāyāma-s absent in the YS and in Vyāsa: (SSS 10, 37). The mention of the nāḍī theories absent in the YS and in Vyāsa. The mention of the alchemical metaphor absent in the YS and in Vyāsa: (SSS 10, 52).

The five anomalies which I have just outlined suggest that Śaṅkara, who supposedly had a good knowledge of the YS, in the event that the PYŚV is his, and who

The beginnings of Advaita doxography  159 would thus not have attributed to Patañjali ideas which were not his, is not the author of those lines. 3.3.2.4  The authorship of Śaṅkara: all or nothing? This being said, now that we know that several portions of the text are not from the hand of Śaṅkara, the author of the BSB, does it refute the hypothesis that Śaṅkara might have composed at least some parts of the SSS? Indeed, is it necessary to adopt the position of all or nothing? Surely, the illustrious Advaita teacher could not have composed the whole text, since a careful reader can distinguish the presence of some post-Śaṅkarian doctrinal developments within it. But does it mean that he had absolutely nothing to do with it? Liebich is prone to believe that later insertions within an original core are a possibility. I argue that, based on the composite nature of doxographies and on the assessment of the function of doxography within the general propaedeutic of the Advaita, which I will present in the next section, rather than relying strictly on the critical analysis of selected passages of the SSS, as has been done before, it is reasonable to admit, at best, the possibility of a text composed in multiple stages, perhaps beginning with a Śaṅkarian impulse of some sort, though it is unlikely to be from his hand at all. When keeping the composite (saṅgraha) nature of the SSS in mind, it is awkward to go about demonstrating the authorship of Śaṅkara based on some verses of the text alone, while we know that most of the SSS is not meant to express the author’s own ideas. At best, most of the SSS’s verses could either be borrowed from competing systems or offer a paraphrase of their main ideas. Thus, analyzing individual verses could help substantiate an approximate datation, for example, as Liebich did. But, in the case of the SSS, the method used by Hacker to disqualify the authorship of Śaṅkara is unidirectional. It can disqualify Śaṅkara’s authorship of some passages, limited to a single chapter of the SSS, the exposition of the Advaita, but it cannot prove his authorship of any passage, especially not of those unrelated to the Advaita. Hence, it can only disqualify some verses. For, even if Śaṅkara’s style and terminology was found in the SSS, nothing can ascertain that it is not a skillful rendition by someone else. Thus, one using Hacker’s method alone is left with the only option of disqualifying Śaṅkara’s authorship of the SSS altogether, based on selected passages which do not reflect his doctrine and terminology, even if most of the SSS is preoccupied with doctrines other than Śaṅkara’s Advaita. As its title explains, the SSS is a mere summative gathering (saṅgraha), a synopsis. It is neither an original work, in the sense that it would formulate new ideas, nor is it a commentary. One can argue that the doxographical nature of the text could have allowed new elements to be easily aggregated to the core text in the course of time. Indeed, the doxographical nature of the SSS, being a compendium, a saṅgraha, would facilitate aggregation. The Nāṭyaśāstra (BhN) attributed to Bharata defines a saṅgraha as follows: “Intelligent men know that a summary is a literary work that concisely [summarizes] the topics that sūtra- and bhāṣya-texts teach extensively.”100 This definition reflects what we know of the genre of doxography. For example, the classicist André Laks describes how doxographies are either

160  The beginnings of Advaita doxography ‘fragments’ or ‘testimonies’ dependent on a certain tradition of transmission. The fragment is a literary quotation from an original source, whereas the testimony is rather an abridged version of, a paraphrase, a commentary, or an allusion to a source.101 Hence, a doxographical compendium102 is more or less a collection of topics gathered from here and there and reorganized according to a given strategy. This format can easily afford continuous additions without breaking the unity of the text, especially if the verses are given in a simplified Sanskrit formulation, as is the case for most of the SSS. Thus, even if the SSS verses had been laid down by Śaṅkara himself, he would have been extracting them from whatever third-party sources he had access to. His authorship, in that case, could only be indirect, as that of a compiler. In other words, the ideas his verses would express would not be his. Only the structural pattern of the SSS and the exposition of Vedānta,103 the one examined by Hacker, could genuinely be his. And this structural pattern, common among Advaita doxographies, is most significant. It thus remains necessary to wonder if Śaṅkara would have designed it himself or if he would have passed on an already established tradition. Else, had he not conceived of it, would he have approved its method? To probe such queries, one must assess the method and function of doxography within the general propaedeutic of the Advaita. 3.3.3 The SSS in the light of the doxographical propaedeutic of the Advaita At this point, more needs to be said on the presence of doxography within the Advaita. As discussed in previous chapters, although the individual topics discussed within each section of a doxography can be aligned in an ad hoc manner, the overall structure of such texts, the sequence of their consecutive sections, betrays a deliberate dialectical strategy. Classicists like Laks, Mansfeld, and Runia argued that dialectic is the Sitz im Leben of doxography. They were the first to notice how doxographical dialectic is teleological.104 This is especially obvious in doxographies produced by the Advaitins. To that effect, Halbfass observed, “the Advaita Vedānta doxographic texts are usually based upon a hierarchical classification at whose apex stands the Vedānta.”105 What is more, not only is the whole hierarchical pyramid of views dominated by the superior position of the Advaita, but each view (darśana) positioned within the text is followed by another which is presented as its logical refutation, what Gokulbhai defines as “the succeeding refuting the preceding,”106 leaving only the Advaita with no reply. Once this structure is understood, it becomes clear why the author of the SSS did not need to engage in the kind of minute uttarapakṣa refutations found in doxographies of the kind of Bhāviveka’s MHK, which entertains its own peculiar dialectical structure. What Halbfass and Gokulbhai failed to notice, however, and what I will now seek to demonstrate, is that this peculiar teleology of Advaita doxographies reproduces a recurring Upaniṣadic pedagogical narrative pattern which Śaṅkara himself commented upon on several occasions and adopted as a teaching device. It is found, for example, in the famous end-story of the Chāndogya Upaniṣad (8.7.1–8.12.6),

The beginnings of Advaita doxography  161 where Indra and Virocana approach their guru, Prajāpati, for instructions about the ‘self’ (ātman). Upon their initial request, they are given a preliminary answer, which happens to please Virocana, the king of demons: the self is like the body which can be seen from a reflection in water. Satisfied with so little, Virocana decidedly leaves his guru in peace and returns home. But, Indra, meditating on the unsatisfactory nature of the answer, comes back to seek further guidance. Several answers are then progressively given to the divine seeker, king among the gods, each time following a preparatory period of purification and contemplation on his part: the self is then said to be like the one who goes happily in a dream; later it becomes like one sleeping a dreamless night. Finally, after a final and shorter interlude, Prajāpati discloses an ever-subtler description of the elusive self, said to be venerated by the gods. Commenting on the story, Jonardon Ganeri made a significant observation worth considering while analyzing the SSS: What is remarkable is the literary device used here, in which progressively more sophisticated accounts of the self are presented as grudging concessions of a recalcitrant god. Indra passes through a sequence of doctrines about the self, each one being an improvement on its predecessor, and perhaps one of the things the story is teaching us is that some such procession is necessary in the quest for self-knowledge. Among two doctrines or propositions, let us say that the first is a ‘preparatory condition’ for a second if understanding that the first is false is required of someone who is to be in a position even to speculate upon the truth or falsity of the second.107 Ganeri’s remarks about the Chāndogya passage is perfectly suited to fathom the dialectic at play in the SSS. It serves to illustrate that the text was not merely meant as (1) a historiographical account of Indian philosophy, nor meant as a mere (2) introductory debate manual, but rather as (3) a pedagogical device to bring about some philosophical insights, for example inquiry into the nature of the self, using competing views as many ladders communicating with one another and leading towards an ever-subtler understanding of reality. Angot (2009) observes that the nature of ‘truth’ and the dialectical process that leads to it, entertained in the myth of Prajāpati, reveals a common Indian understanding of the value of words: Ce qui importe ce n’est pas de dire la vérité – celle-ci, au sens où nous l’entendons généralement, n’existe pas au niveau de la parole – que de dire une vérité qui vient élargir celle à laquelle précédemment on avait jusque-là accès.108 Similarly, for the Buddhists, Jainas, and Advaita Vedāntins who produced doxographical texts, truth (sat) lies beyond the realm of words (śabda) and established opinions (siddhānta). The great Advaita theologian himself relegated the Veda and related scriptures (āgama-s), the very āpta of his system, to the realm of nonultimate truth (aparāvidyā). This understanding relies on the Muṇḍaka Upaniṣad

162  The beginnings of Advaita doxography (1.1.4), for example. Accordingly, only Brahman is the ultimate truth (parāvidyā). Yet, as pointed out by Ganeri, the words of authorities act as indicators. They offer a guidance which, when carefully followed, effects a widening of views previously assumed, until truth is finally met face to face, without the remedy of words or concepts. This ‘guidance’ resembles the Advaita propaedeutic at play within the SSS. The Chāndogya is not the only Upaniṣad to display such a device. Śaṅkara commented on a similar pedagogy in the Bṛhadāraṇyaka (BAU Bhā 3.6.1), in a passage where Yājñavalkya responds to Gārgī’s cosmological inquiry. Hirst (2005) also noted how this process of “successively seeing what is more subtle” is found by Śaṅkara in the Kaṭha Upaniṣad. Hirst adds, This idea of progressing through stages so that the self beyond all can be realized is shown clearly in Śaṃkara’s explanation of what the Upaniṣad means by saṃkrāmati (‘coming together’ or ‘going through’) in Tait Up 2.8.5. . . . Rejecting the views that saṃkrāmati means ‘attainment’ of the self or identification with any one of the five sheaths (of food, breath, mind, understanding, and bliss), Śaṃkara concludes that it means ‘realization alone’. This is the realization that the self is within yet other than the five sheaths which are simply misidentified with, or superimposed on the self. It is to the realization of ‘brahman (as) reality, consciousness, infinite’ (Tait Up 2.1.1), beyond all conceptual construction, that the stages of understanding, through stories of manifestation and interiorization, lead.109 Hirst argues that this ‘process of interiorization’ of progressive doctrinal stages is central to Śaṅkara’s own pedagogical method.110 Beginning at his level of understanding,111 the student must be progressively introduced to higher levels, until the ultimate experience of truth dawns upon him. Hence, it is plausible that the Advaita theologian would have reproduced the same propaedeutic to ‘truth’ within a short doxographical teaching that aimed at clarifying his understanding of the position of the Advaita among the theories of competing groups. If he did not compose the SSS himself, which is most likely, the thinker who did was well acquainted with Śaṅkara’s teaching technique. What is worth emphasizing here is that, just as the teleology at play in Bhāviveka’s MHK reproduced the typical dialectic of ‘no-view’ adopted by Madhyamaka Buddhism, while the teleology of Haribhadra’s ṢDS was in tune with the Jaina dialectic of anekānta-vāda, the SSS perfectly assimilated the dialectical attitude characteristic of its own Vedāntic tradition. Again, the dialectical endeavours found within each of these ‘literary technologies’ were not solely interested in interreligious debates but were rather primarily designed to bring about a specific knowledge of significant religious value, while disposing of heterogeneous distorting views. In that sense, Indian doxographies in general appear as contemplative guides on the path to liberation, a path essentially resting upon knowledge (jñāna-mārga). They somewhat recall Rabbi Moshe ben Maimon’s (Maimonides: 1135–1204 CE) famous Guide for the Perplexed in intent.112 In any case, one can

The beginnings of Advaita doxography  163 safely argue that the doxographical presentation of various philosophical hypotheses, within the optic of a teleological dialectic, is an essential component of the propaedeutic methodology of the Advaita promoted by Śaṅkara,113 itself a hodgepodge of multitudinous doctrines reinterpreted through a Mīmāṃsā-styled exegesis, fundamentally resting upon Vedic scriptures. Bhāviveka, writing in the sixth century, already observed this heterogeneous state of affairs happening under the name of ‘Vedānta’ prior to Śaṅkara,114 in the Vedānta-tattva-viniścaya chapter of his MHK, the first written critique of the emerging Vedāntic system.115 In other words, I argue that the communication of the precise understanding of the tenets of Advaita necessarily passes through a careful examination and rejection of competing religious positions, and that doxographical expositions are no mere coincidences or ‘beginner guides’116 within the system. Teleological dialectic is made especially important in the exegetical and soteriological context of the Advaita, where ignorance (avidyā) is equated with superimposition, i.e. the erroneous attribution (adhyāropa) of wrong belief unto an object, and particularly upon the self (ātman); and where final beatitude (apavarga) consists of gaining the definite knowledge (jñāna) of truth, by removing any superimposed attribution. As Hirst mentioned, adhyāropa-vāda is the “method par excellence” of Advaita.117 By learning about the possible fallacies imputed to the self, through an investigation of non-Advaita positions presented against the background of authoritative scriptures, one becomes acquainted with an ignorance which needs to be identified prior to being neutralized. There is no doubt that adhyāropa is especially equated by the Advaitins with the views of their competitors. The fifteenth-century Vedānta-Sāra (VS) of Sadānanda, for example, following the Advaita soteriological propaedeutics, introduces a doxographical account of competing philosophical systems, including various Cārvāka-s, two Buddhists, the Prābhākara-s, the Tārkika-s,118 and the Bhāṭṭa-s within its exposition on adhyāropa.119 The practice of critically examining competing views is, however, not original to the Advaita or to doxography alone, but was already harnessed by Buddhists, among others, appearing under the name of parīkṣā, ‘investigation,’ ‘examination,’ an important component of the Yogācārabhūmiśāstra’s (YBhŚ) definition of the hetuvidyā (lit. ‘the science of causes’), ‘dialectics,’ which had a significant impact on later Buddhist philosophers.120 For our present purposes, let’s observe, along with Vincent Eltschinger, that hetuvidyā did not only serve apologetic purposes but was also held to support religious praxis as a precondition to the reflective and meditative parts of the path,121 by removing erroneous views and their corresponding ill-conceived actions. The removal of erroneous attribution (adhyāropa) through exegetical methods, as advocated by the Advaitins, thus shares close similarities with its Buddhist elder brother.122 Similarities between the two worldviews are indeed many.123 In this light, one better understands how the exposition and refutation of competing views are a significant part of the propaedeutic of the Advaita path. In debate, to establish one’s view, one must take position, against an opponent. Coming back to Śaṅkara himself, his own BSB is replete with doxographical materials.124 The second chapter of the BSB, on avirodha (lit. ‘non-contradiction’),

164  The beginnings of Advaita doxography directly addresses most of the views enumerated in the SSS, namely Sāṃkhya, Yoga, Vaiśeṣika, Buddhist realism and idealism, Jainism, and even the Bhāgavata view (absent in the SSS). For the missing Mīmāṃsā (Prabhākara-pakṣa and Bhaṭṭācārya-pakṣa) and Vedavyāsa-pakṣa sections, separately identified in the SSS, one may observe that the whole BSB is more or less an ongoing display of a Mīmāṃsā-styled exegesis of the Brahmasūtra, ascertaining the Advaita Vedānta perspective as ultimate. However, within the Bhāṣya, clear references to Kumārila or Prabhākara are hard to secure.125 As for the Lokāyāta-pakṣa, meagrely presented in the SSS, it is implicitly refuted by the BSB, especially in the third section of the second chapter, which discusses the origin of the elements, of the mind, of the soul, and so on. Regarding the need for such dialectical discussions within the BSB, one can refer to the passage at 2.2.1, where Śaṅkara answers an opponent who asks, as if vexed that heterodox views were constantly being addressed, whether the views of the Sāṃkhya-s and so on had not already been refuted earlier and what is the need of discussing them all over again. Hence, Śaṅkara replies, “What we have done so far was just to prove that their interpretations were mere fallacies and not the correct explanations. But here follows a refutation of their reasoning independently of the [Upaniṣadic] texts.”126 Thus, Śaṅkara explains that his previous evaluation of opposing views refuted fallacious interpretations of scriptures on the ground of (1) a critical examination of authoritative scriptures alone and that he is now providing a second attempt of refutation on the ground of (2) reason, independently of scriptures; by examining the consistency of the views to be refuted. The first discussion was meant for the doctrinal education of his pupils, the second is a direct attack on the doctrines of his opponents. This method recalls the Buddhist definition of parīkṣā, provisionally defined by Eltschinger as the evaluation, by means of reason(ing) (identified or not to the pramāṇas) and/or scripture (provided the opponent belongs to the same confessional denomination), of an opponents’ theoretical and/or practical tenets in order to assess their rationality and, further, the soteriological relevance of the system as a whole.127 There should be little doubt that the insistence of the author of the BSB on the critical examination and refutation of the views of others, as it was for the Buddhists, does not merely serve apologetic purposes, but that it is also understood as a crucial component of the religious praxis he is promoting. Hence, the idea of a separate doxographical treatise on his part, dedicated exclusively to this dialectical exercise, would not be impossible. In brief, the composition of a doxographical treatise is in tune with the general Advaita propaedeutic. For, I have argued that (1) Advaita doxographies reproduces a recurring Upaniṣadic pedagogic narrative pattern which Śaṅkara himself commented upon on several occasions and adopted as a teaching device; thus (2) the SSS perfectly assimilated the dialectical attitude characteristic of its own Vedic tradition; hence (3) doxographical texts like the SSS were not meant as (a) a

The beginnings of Advaita doxography  165 historiographical account of Indian philosophy, nor as mere (b) introductory debate manual, but especially as (c) pedagogical devices to bring about philosophical insights of soteriological value. 3.3.4  Summary on the authorship: unsettled Considering the relevance of doxographical lists within the Advaita’s religious training, I am inclined to believe that Śaṅkara participated in the Advaita doxographical practice himself, either orally or in writing, as seen in his BSB. This does not mean that he participated in the formulation of the SSS. A more thorough research should compare the presentations of opponents found in the BSB and in other works of Śaṅkara with the ones in the SSS, along with a general review of all the SSS manuscripts available to date. It is nonetheless significant, for now, to observe that a doxographical impetus was present within the Advaita from its very beginnings, in concordance with its pedagogical and soteriological objectives. Thus, aware of how contentious this may sound, as it mitigates a sound line of interpretation established since Hacker, I conclude that no definitive judgment can be made so far based on either the arguments in favour or in disfavour of the entire authorship of Śaṅkara regarding the SSS. At least, no argument conclusive enough to consider the discussion of the SSS closed. Yet, so far, the case is strong against the possibility that Śaṅkara has anything to do with the SSS. Although unlikely, it remains possible that the authorship of the core of the SSS be, at best, obliquely attributed to the author of the BSB, through an oral transmission perhaps, which he would have initially imparted; or, even less likely, through a basic written template initially forming the core of the SSS – such a basic core would need to be ascertained, at least in the light of the BSB. This being said, one must keep in mind that the text contains numerous passages and expressions which cannot plausibly come from his hand and which would consequently have to be seen as later additions, with the consequence of breaking the text in parts. In order to prove this, one needs to identify the many strata of the text, in the face of the available manuscripts. This is not a task which I have done at this point. Nor do I consider it necessary to my argument. My concern is rather to argue that, even if Śaṅkara would have approved of the structural teleology of the SSS, what I presented as the dialectical propaedeutic of the SSS, the very structure of the text, is a coherent doctrinal product of Advaita ideology, regardless of its author. It deserves attention, because it offers an additional code to read through the intricate dialectic of ancient Indian sectarian faultlines. 3.3.5  On the dates of the SSS The question of the dates of the SSS has seen several hypotheses, which I reviewed in my article. To summarize, the only unchallenged latest date remaining, which is substantiated by an argument, is the one of Liebich: before the tenth-century Bengali scholar Śrīdhara (Nyāyakandalī).128 It has also been brought to my attention,

166  The beginnings of Advaita doxography by Philipp Maas, that the portrayal of Patañjali as an incarnation of the divine serpent Ananta, as mentioned in the first verse of Chapter 10, is first attested in an introductory stanza to Bhoja’s commentary on the sūtra-part of the PYŚ, in the early eleventh century.129 It is possible that the SSS aggregated various materials from different periods. It would seem peculiar, however, to speculate on a date of the SSS coming later than the SDS, since the latter likely borrowed from the former. It would also be prudent to take into consideration the few dated elements of the text when considering a probable theory. Taking these matters into consideration, I suggest that the core of the SSS might have been in place at the earliest around the tenth century, with a more probable later date – allowing for later additions until the fourteenth century. If Śaṅkara had any role in the SSS, in my opinion, it can only have been by establishing a pedagogical methodology making use of a progressive hierarchy of teachings. He needs not be, and can hardly be proven to be, the author of the SSS. Regardless, the SSS is a valuable case study in Indian doxography.

3.4  SSS Chapters 7 and 8: review and reflection My reading of the two SSS chapters has been done by comparing three different versions in addition to the critical edition of Raṅgācārya.130 There need not be an extend commentary and philological analysis of the text presented within these pages, especially since the thesis of Gokulbhai already said most of what needed be said on the matter. However, I will add a few words. The whole SSS has twelve chapters (prakaraṇa-s):   1   2   3   4

  5   6   7   8   9 10 11 12

Introduction (upodghāta-prakaraṇa) The Thesis of the Lokāyatas (lokāyatika-pakṣa-prakaraṇa) The Thesis of the Ārhatas (ārhata-pakṣa-prakaraṇa) The Thesis of the Bauddhas (bauddha-pakṣa-prakaraṇa) a b c d

The The The The

Opinion Opinion Opinion Opinion

The The The The The The The The

Thesis Thesis Thesis Thesis Thesis Thesis Thesis Thesis

of of of of of of of of

of of of of

the the the the

Mādhyamikas (mādhyamika-mata) Yogācāras (yogācāra-mata) Sautrāntikas (sautrāntika-mata) Vaibhāṣikas (vaibhāṣika-mata)

the Vaiśeṣikas (Vaiśeṣika-pakṣa-prakaraṇa) the Naiyāyikas (naiyāyika-pakṣa-prakaraṇa) Prabhākara (prabhākara-pakṣa-prakaraṇa) Bhaṭṭācārya (bhaṭṭācārya-pakṣa-prakaraṇa) Sāṃkhya (sāṃkhya-pakṣa-prakaraṇa) Patañjali (patañjali-pakṣa-prakaraṇa) Vedavyāsa (Vedavyāsa-pakṣa-prakaraṇa) the Vedānta (vedānta-pakṣa-prakaraṇa)

The two Mīmāṃsā chapters which preoccupy us come right after the presentation of the Nyāya system and before the Sāṃkhya one. As noted previously, each

The beginnings of Advaita doxography  167 chapter is dialectically positioned to refute the preceding, at least theoretically, and is thus considered a higher view in the eyes of the author. Thus, the author of the SSS takes the teachings of the two Mīmāṃsā schools as higher than Nyāya and so on, but lower than Saṃkhyā and the following. In a footnote to his article on Dialogue and Genre in Indian Philosophy, Nicholson suggested that the SSS abandons this dialectical teleology from the fifth chapter onwards.131 However, I must admit not knowing on what this claim is supposed to rest. In fact, Chapter 5, on Vaiśeṣika, mentions being a refutation of the previous Buddhist ideas and concludes by pointing out that the following chapter, on Nyāya, will bring the conclusions of the Vaiśeṣika one step further, by resorting to more logical means (SSS 5.37b: tarkaiḥ). Thus, I see no reason to believe that the hierarchy is abandoned, on the contrary. Also, while Nicholson uses the analogy of Hegel’s dialectic to refer the “modern readers” to a similar “progressing order,” we have seen already how it is in fact a pattern common to Upaniṣadic pedagogy and central to Śaṅkara’s teaching method. Additional features worth noting about the hierarchical order of teachings in the SSS is that Patañjali’s system is placed above Sāṃkhya, immediately before the Vedavyāsa thesis, which is the closest to the Vedānta of the author. Perhaps one could see this as an attempt to present the Patañjali system as a combination of the karma-mārga of the Mīmāṃsā, which is hierarchically supplanted by Sāṃkhya, and the jñāna-mārga of the Sāṃkhya, altogether forming a synthesis more acceptable from a Vedāntic perspective. This would go along the line of a tradition at least dating from the time of Bhāviveka where yogic activity, in its performative ritual dimension, was perceived as part of the karma-mārga, and thus as a ritual activity related to the realm of Mīmāṃsā. Moreover, verse SSS 8.36b (related to the thesis of Bhaṭṭa) explicitly refers to the yogic practices of sense withdrawal (pratyāhāra) and so on (ādi), equally mentioned in MHK 9.84. Perhaps one could suggest that when the Mīmāṃsā method of exegesis is applied onto elements from Sāṃkhya, combined with an internalization of ritual practices, as in Patañjali’s YS, the commentary of which was legendarily attributed to a certain Vyāsa, the ‘new’ system adopts a form more palatable for the Brahmanical jñāna-yogins, the Advaitins. This could serve to explain what various scholars perceived as Śaṅkara’s favourable attitude towards yoga and offer an alternative interpretation to Hacker’s hypothesis, according to which the great Advaitin would have been a yogin in his youth. After all, could Advaita itself not be considered as a somewhat insightful combination of Mīmāṃsā and Sāṃkhya, the two Vedic systems with which it argues the most? In this light, perhaps, Patañjali’s YS could also be seen as an early systematizing attempt, meaningful in a Mīmāṃsā-trained milieu, where traditional ritual concerns evolved into a complex theurgy requiring rigorous psychosomatic manipulations. More research on the topic would certainly be needed before jumping to conclusions, but the idea offers new avenues for understanding these simultaneous doctrinal and ritual developments. As for the Vedavyāsa thesis (Vedavyāsa-pakṣa-prakaraṇa), its presence has been succinctly commented upon by Raṅgācārya in his introductory remarks to

168  The beginnings of Advaita doxography the text. In brief, the author of the SSS distinguishes from the opinion of Vyāsa, the author of the Mahābhārata (bhārata-mata), and Bhagavatpāda Govinda’s commentary on Vyāsa, here taken as the author of the Vedāntasūtra-s (sūtramata). Only the latter is taken as the definitive view of the Advaita Vedānta. The division reflects an internal exegetical distinction within Vedānta. It is remarkable that this early Advaita doxographical tradition assigns at its apex two major texts, the MhB, and the VS, attributed to Vyāsa, in a hierarchical order immediately following the thesis of Patañjali, the commentary of which (YSBh) has often been attributed to Vyāsa, as if to claim all three doctrines as coherent progressing stages of a same path. Concerning the two chapters which primarily interest us here, the thesis of Prabhākara (prabhākara-pakṣa-prakaraṇa) and the one of Bhaṭṭācārya (bhaṭṭācārya-pakṣa-prakaraṇa), we can notice at the outset that their twofold division reflects a development within the Mīmāṃsā which was present neither in Bhāviveka’s nor in Haribhadra’s assessments. The latter had mentioned the sūtra-kāra Jaimini, but the two interpretative lineages of Kumārila Bhaṭṭa and of the one whom tradition presents as his dissident pupil, Prabhākara Miśra,132 were not alluded to. The reasons for this omission are likely temporal, but also because it was not necessary within the dialectical settings of the other texts. It is noteworthy that the first verse of SSS Chapter 7 presents a rare occurrence of biographical details within Indian doxographies: “He glorified the pūrva-mīmāṃsā but, yet, with envy towards his teacher” (SSS 7.1b). However, again, the mention appears more rhetorical than historical. The introductory remark offers as a kind of moral justification expediently explaining, in a nutshell, why Prabhākara’s thesis is lower than that of Bhaṭṭa. Envy towards one’s teacher is a reprehensible attitude. Again, the hierarchical order of the SSS carries on. Both chapters are subdivided following a mode of presentation which is repeated throughout the SSS. The division of topics recalls Haribhadra’s ṢDS, while it adds further doctrinal elements. Gokulbhai gives a sixfold nomenclature which well captures the summary: 1 2 3 4 5 6

The The The The The The

world (jagat) means of proof (pramāṇa-s) absolute goal (mokṣa/mukti) practice or method (upāsana/sādhanā) nature of God (deva/īśvara) type of experience (anubhūti) of the state of liberation133

Gokulbhai’s first point (jagat), would be better rendered as tattva, as we can see that it encompasses, for the Prābhākara-s (SSS 7.2), the five properties (pañcaartha) along with potentiality (śakti), similarity (sādṛśya), and quantifiability (saṃkhya); while for the Bhāṭṭa-s (SSS 8.6), they amount to the various types of Vedic sentences which we have discussed before: the injunctive (vidhi), the descriptive (artha-vāda), the mantric formula (mantra), and the appellative (nāmadheya). Thus, the pattern remains that tattva-s are inherent to the formulation of a

The beginnings of Advaita doxography  169 darśana, just as are the pramāṇa-s, the deity (devatā), and the interest in soteriology (mokṣa). What the SSS adds to the discussion, in comparison to the ṢDS, is the method propounded by each darśana to escape bondage and a succinct definition of the nature of that liberated state.

Notes 1 See Halbfass 1979: p. 196. 2 The argument was made by the sole editor of the text in English, Raṅgācārya, in 1909. It can be found in a more recent edition (2006: p. v), which I have been using for the present study. However, one must admit that the argument can go both ways and that the title might not have been given by the author himself. 3 Āryāvarta is a name for Northern India in classical Sanskrit literature. Manu Smriti 2.22 gives the name to the tract of land between the Himalaya and the Vindhya ranges, from the Eastern to the Western Sea. Other texts allude to various locations: Vasistha Dharma Sutra I.8–9 and 12–13; Baudhayana Dharmasutra 1.1.2.10–11, and so on. In brief, the name is used to denote the land of the Aryans, where Brahmanical norms and customs are respected. Its geographical location has never been definitive. The term appears to be a projection of power more than a physical reality. It is sometimes used in contrast with the Uttarāpatha, the land of heterodox believers, as observed by Jason Neelis: “Heterodox beliefs and practices attributed to regional inhabitants who were associated with ritually impure ‘foreigners’ (mlecchas) contributed to the marginalization of the Uttarāpatha. Sanskrit passages often contrast the borderlands of the Uttarāpathawith the heartland of Āryāvarta” (Neelis 2010: p. 195). 4 In reference to How the Brahmins Won: From Alexander to the Guptas, Bronkhorst 2016a. 5 Lorenzen 1983: p. 156. 6 Hirst 2005: p. 14–15. Points v–vii have been left aside since they do not play any significant role for the present analysis. Note that points viii to xi are debated. 7 BAU Bhā 1.1.1. The translation is from Mādhavānanda 1950: p. 9. 8 Researching the history of the term rāja-yoga, Jason Birch (2013: p. 401) found out that “The earliest definition of Rājayoga is found in the twelfth century, Śaiva Yoga text called the Amanaska, which proclaimed Rājayoga to be superior to all other Yogas and soteriologies prevalent in India at that time.” Further on, Birch described the distinction between haṭha-yogaand rāja-yogaas a division which goes along the lines of the one separating the Pūrva-Mīmāṃsā from the Uttara-Mīmāṃsā (Vedānta): “As a foil for Haṭhayoga, it represents the mental, meditative, advanced, and purely soteriological Yoga techniques as opposed to the physical, preparatory, and Siddhi-orientated ones. The fault-line between Haṭhayoga and Rājayoga is an offshoot of the more ancient tension between ritual and gnosis in Indian religions” (Birch 2013: p. 419). This division between an action and gnosis oriented yoga not only characterizes the kind of soteriology promoted by Śaṅkara (identified with Śiva himself) as rāja-yoga, but it justifies the choice of Bhāviveka to discuss action (karma) oriented yoga within his chapter on Mīmāṃsā (MHK 9), as the lowest of all creeds, coming right after the kind of jñāna-yoga promoted by Vedāntins (MHK 8). 9 For the original dissertation, see Bader 1991. For the book format, see Bader 2000. 10 Whether or not Mādhava-Vidyāraṇya is the author of both texts, the rhetoric of conquest permeates both. 11 Hirst 2005: p. 16. 12 Hirst refers to the homage verse 2: yadvāṇīkuliśāvarugṇamatayaḥ . . . tārkikāḥ. See Hirst 2005: p. 12, n. 18. 13 Bader 1991: p. 113.

170  The beginnings of Advaita doxography 14 Bader 1991: p. 114. 15 Bader 1991: p. 116. 16 Hirst refers here to Padmapāda’s commentary on the first chapter of the BSB. See Hirst 2005: p. 16. 17 Bader 1991: p. 118. 18 Remember that, in MHK 9.72, Bhāviveka criticizes the Vedic gods for their lack of compassion towards the demons (asura-s). Would Bhāviveka have also been aware of the Brahmanical tendencies of equating their opponents, particularly Buddhists, with demonic forces? The pattern has often been commented upon. Verardi offered many illustrations of the theme in literature and arts. See Verardi 2011. 19 For the sake of the Tolkien analogy, it is worth noting that the name Sauron (Þauron) or Thauron (θaʊron), the eponymous Lord of the Rings in Tolkien’s mythology, is supposed to mean the ‘The Abhorred,’ a semantic equivalent to the Sanskrit name Rudra (roaring, dreadful, terrific, terrible, horrible) often given to Śiva, of which Śaṅkara is said to be the avatāra. It is exactly this image of Rudra the abhorred that Bhāviveka ridiculed in MHK 9.108–109: “Homage to him, Rudra (the Horrible), whose name the meaning is suiting.” 20 By this, I suggest that Marshall McLuhan’s famous theory that “the medium is the message,” mentioned in his 1964 work Understanding Media: The Extensions of Man, an international classic of communication studies, applies to Indian doxography in general and to the Advaita type in particular. By this sentence, the Canadian professor meant that a medium is inherent in any message it conveys. The consequences are that there is a symbiotic relationship between a medium and the message it creates. The former naturally determines the latter. McLuhan observed that “the medium is the message” “because it is the medium that shapes and controls the scale and form of human association and action. The content or uses of such media are as diverse as they are ineffectual in shaping the form of human association. Indeed, it is only too typical that the ‘content’ of any medium blinds us to the character of the medium” (McLuhan 1964: p. 9). Along these lines of thinking, I argue that it is not exactly the minute arguments presented within doxographies which impact their audience, but the dialectical teleology through which they are presented, the hierarchical structure of the texts. Thus, the classical way of analyzing them solely through a detailed philological examination of each of their consecutive argument runs the risk of blinding the analyst from their actual ‘message.’ 21 Whaling 1979: p. 30. 22 See Isayeva 1993: p. 14. 23 As quoted in King 2003: p. 55. 24 King 2003: p. 65. 25 Whaling 1979: p. 2. 26 Whaling 1979: p. 27. 27 See Whaling 1979: p. 1. 28 This last sentence rewords a previous conclusion of mine. See Bouthillette 2017a: p. 74. 29 Sarvepalli Radhakrishnan (1929: p. 450), for example, in his praiseworthy attempts to defend the concept of ‘Indian philosophy,’ depicted Śaṅkara in the ‘favourable’ garbs of a philosopher. More recently, Ram-Prasad (2002) also insisted on the philosophical framework of Advaita Vedānta. 30 Hirst 2005: p. 86. 31 BAU Bhā 4.4.22: vedānuvacanena = mantrabrāhmaṇādhyayane. 32 Āpastambaparibhāṣāsūtra 1.33: mantrabrāhmanayoḥ vedanāmadheyam. 33 Ṛg V. Bhā, Bhūmikā [Poona, p. 3]: mantrabrāhmaṇātmakaḥ śabdarāśirvedaḥ. 34 Tantra-vārtika 1.3.10: mantrabrāhmaṇayovedeti nāmadheyaṃ śaḍaṅgam eketi. 35 As mentioned by Murty 1974: p. 25–26.

The beginnings of Advaita doxography  171 The SSS, at verse 8.9a, gives its own definition of mantras:



karmāṅgabhūtā mantrāḥ syur anuṣṭheyaprakāśakāḥ | “Mantras, constituting the limbs of ritual performances, elucidate what is to be accomplished.” 36 AS 1.10 apauruṣeyaṃ vākyaṃ vedaḥ | sa ca vidhimantranāmadheyaniṣedhārthavād abhedhāt paṅcavidhaḥ | 37 On the meaning of ‘nitya’ in Mīmāṃsā, see Freschi 2017. 38 Eltschinger 2007: p. 78. 39 Ram-Prasad 2001: p. 216. 40 Ram-Prasad 2001: p. 186. 41 This is one of the four ‘great pronouncements’ (mahā-vākya-s) said to directly designate the self (ātman) and to bring about liberation in one who is ready (adhikārin). It is found in Ch. Up. 6.8.7. 42 Bronkhorst 2007b: p. 38. 43 Bronkhorst 2007b: p. 47. 44 MS 1.1.1 athāto dharmajijñāsā 45 MS 1.1.2 codanālakṣaṇo ’rtho dharmaḥ 46 As examples of what one might sacrifice for, one may think of the various ‘boons’ (vara-s) suggested by Death to Naciketā, in Kaṭha Up. (1.1.23–25), in replacement for his unreasonable desire to know the nature of death: sons, grandsons that will be centenarians, many animals, elephants, gold, horses, to live long, wealth, to become the ruler of a region, otherworldly women with chariots and musical instruments. 47 BSB 1.1.1 athāto brahmajijñāsa 48 Bronkhorst 2007b: p. 38. 49 See Hirst 2005: p. 42. 50 The conditions are listed in BSB 1.1.1. They are here given by Hirst 2005: p. 41. 51 Caste exclusion in Śaṅkara is also discussed in Bronkhorst 2007b: p. 42–43. 52 One might recall here the words of Śantideva’s Bodhicaryāvatāra (3.21): evam ākāśaniṣṭhasya sattvadhātor anekadhā | bhaveyam upajīvyo ’haṃ yāvat sarve na nirvṛtāḥ || So may I in various ways support the whole sphere of life lodged in space, until all be at peace. The translation is from the Bibliotheca Polyglotta of the University of Oslo, visited on 5 February 2018 at https://www2.hf.uio.no/polyglotta/index.php?page=fulltext&vid= 24&view=fulltext&cid=45880&level=1. 53 The place of women within the saṅga, and their ability to attain liberation, however, has been a controversial issue through the ages and among the cultures which adopted Buddhism. On the history of women in Buddhism, one can recommend Engelmajer 2014; Kaushik 2016; Paul and Wilson 1979; Tsomo 2000. Yet, Bhāviveka’s verbose

172  The beginnings of Advaita doxography reaction to the pūrvapakṣa (9.11) requesting their exclusion from the study of the Veda suggests his indignation at the idea. 54 It is to be noticed that MHK 4 is titled Śrāvakatattvaviniścayāvatāraḥ, in reference to Buddhist śravaka-s, whom Bhāviveka refers to in distinction to the followers of Mahāyāna systems. The term śravaka underwent several semantic changes through the history of Buddhism. In Jainism, the term was commonly used to denote the lay followers. 55 To be precise, Bhāviveka’s accusation targeted the Mīmāṃsakas who favoured karma over jñāna. Yet, one could say that by associating with the Mīmāṃsā, Śaṅkara assumed its qualities. 56 On the turbulent relations of Brahmanism with its religious competitors, one can consult Bronkhorst 2007a, 2011, 2013, 2015, 2016a, 2017; Sarao 2012; Singh 2017; Thapar 1975, 1989, 2010; Verardi 2011. 57 On the absence of concern for liberation in Jaimini and Śabara, see Bronkhorst 2007b: p. 1, 58. Also note that the idea of liberation is introduced within the prior Mīmāṃsā by Kumārila. 58 In his book How the Brahmins Won, Bronkhorst made a careful association of the ancient Brahmanical mindset towards large cities and their political structures with the one of the National-Socialists of the Third Reich. The passage is worth quoting: “Brahmanism was not only inward-looking but also extremely conservative. Brahmins did not like the new political structures they encountered, nor the new phenomena that accompanied these. The Maurya Empire had been centred in Magadha, right in the middle of the region where South Asia’s second urbanization was taking place. The Maurya capital, Pāṭaliputra, was a large city, according to some the world’s largest city at its time. Brahmins detested towns and cities. They said so explicitly when obliged to talk about them. More often, they adopted a different strategy: they ignored their existence. Where possible, they depicted themselves in a world that was no longer there. In this respect (and to avoid misunderstanding, let me add: only in this respect), their behaviour was not dissimilar to that of another group that was obsessed with the past: the National-Socialists of the Third Reich. In the House of German Art in Munich, opened by Hitler in 1937, there were hundreds of paintings; not one depicted urban and industrial life” (Bronkhorst 2016a: p. 110–111). Bronkhorst is careful to delimit his association of the Brahmins with this dark period of European history. Yet, following the various lines of reasoning which I have threaded within these pages, and however unpleasant it might be to observe such traits in the character of revered individuals who became central cultural and religious icons, it is no exaggeration to refer to ‘fascist’ tendencies within elements of the Brahmanical supremacist ideology claiming the likes of Kumārila and Śaṅkara. The cultural significance of these figures is such, even today, that the shadow which they cast on a past lying dormant under Gandhi’s beloved myth of ‘the tolerant and peaceful India’ and Nehru’s denial of anti-Buddhist sentiments during the Brahmanical revival of the fifth century onwards (Nehru 1946: p. 188) must be brought to light and duly acknowledged, just as modern Germany acknowledged the mistakes of the past and educates its citizen accordingly. This very myth of tolerance, at the foundations of modern India (Rajan 2011), is now being used by a section of modern Hindutva, an ultra-nationalist movement of Hindu religious leanings (Baber 2000; Banerjee 1991; Chakravarti 1998; Desai 2016; Jaffrelot 1993, 2007; Ram 1997; Reddy 2011; Sharma 2002), as a tool to promote, yet again, the ‘superiority’ of their faith vis-à-vis others (Udayakumar 2005; Uncategorised 1993), and particularly the Abrahamic cults, perceived as a cultural threat, just as Buddhism or Jainism might have been perceived by their ancestors, over a thousand years ago.   What is more, as the ideology of such ancient religious icons spreads worldwide, through the medium of yoga and esoteric thinking, it is relevant to refresh our understanding of the role which they played in shaping their society, besides speculating on

The beginnings of Advaita doxography  173 the philosophical nature of the highest good. Religious authorities are never neutral social actors, even if, or perhaps especially when, they ‘renounced the world’ (Veer 2002). Their thoughts and actions have consequences which can reverberate long after their time. Wouldn’t there be some irony in seeing a young progressive (sexually and politically liberal) and emancipated (unmarried) foreign woman, which would have been little more than a defiled mleccha (a non-Aryan barbarian) in the eyes of the ‘universal guru,’ living by Śaṅkara’s precepts and ideology in hope of an ultimate bliss which he categorically denied her? So far it is perhaps only amusing. But what if the well-meaning devotee starts to assume her moral superiority, on the basis of her exclusive faith, and engages in political activism, bringing Śaṅkara back on the stage of mainstream politics? After considering the factors which I highlighted within these pages, and upon which I suggested an interpretation, it is up to everyone to estimate the worth of such icons and their ideology, in the face of the broader picture of their legacy. This is not to say that Advaita Vedānta did not change through time, for it surely did (Hatcher 2008; Sharma 2013), but simply that the past should be known in its complexity rather than being idealized through mystical or sentimental (spiritual) readings. 59 Parpola 1981, 1994. 60 Bronkhorst 2007b: p. 48. 61 For a comprehensive discussion of pre-Śaṅkara Vedānta and of later Advaitins, see Saraswati and Alston 1997. 62 Bronkhorst 2007b: p. 25. 63 Ram-Prasad 2001: p. 198–199. 64 Ram-Prasad 2001: p. 199. 65 Brackets are mine; Menezes 2017: p. 137. 66 On the relevance of the term ‘praxis’ in this yogic context, see Bouthillette 2017b. 67 Halbfass 1991: p. 226–227. 68 Ram-Prasad 2001: p. 184. 69 Bhāviveka uses the corresponding terms viniścaya and nirṇaya in four chapter titles of the MHK: 4, 5, 8, and 9. 70 Angot (2009: p. 105) noted, “Il est évident que la procédure judiciaire a eu une influence sur la manière dont les débats étaient organisés, ne serait-ce que par la participation de la cour et de la personne du roi. Le jugement, la sentence, l’arbitre, la preuve, etc.: tout ce vocabulaire atteste de la parenté des deux types de débats.” 71 Hirst 2005: p. 64–65. 72 Mu Up 1.1.4 tasmai sa hovāca | dvevidye veditavyeti ha sma yad brahmavido vadanti parā caivāparā ca || As quoted in Hirst 2005: p. 176. 74 Hirst 2005: p. 85. 75 Hirst 2005: p. 67. 76 This section is the abridged version of an article of the same title published in 2019, itself the outcome of the Ninth International Indology Graduate Research Symposium held at Ghent University in September 2017. 77 A partial edition of the SSS had previously been given in Poussin 1902. Other noncritical editions of the text (i.e. editions without any critical apparatus to determine which manuscript version most closely approximates the original version) can be found in Bose 1929; Gokulbhai 2009; Śukla approx. 1930 (undated print). 78 Regarding the learned Indian scholar, Catherine Morice-Singh (2017: p. 1) noted that “M. Raṅgācārya had been appointed in 1901 as Professor of Sanskrit and Comparative Philosophy, as well as Curator of the Library of Oriental Manuscripts (GOML). Even before the recognition he received for this work on the GSS [Gaṇita-sāra-saṃgraha],

174  The beginnings of Advaita doxography he had already been awarded the prestigious ‘Rao Bahadur’ medal and title in 1903, in recognition of his profound scholarship.” 79 Raṅgācārya (2006: p. v–vi) observes that the two best-preserved manuscripts of SSS were kept in Madras, at the Government Oriental Manuscripts Library (GOML), and in Mysore, at the Oriental Research Library (ORI). He mentions that GOML possesses the transcript of another original, which contains the only available commentary composed by Śeṣa Govinda (ca. sixteenth century) and indicates that a partial palm-leaf manuscript is preserved in Cochin, at the Palace Library of H.H. the Maharaja of Cochin, while the Palace Library of Travancore is in possession of another.   However, the recension of Raṅgācārya needs to be updated in the light of the new manuscript catalogues available. Based on the Online Library of Digitized Sanskrit and Prakrit Manuscript Catalogues, maintained by Dominik Wujastyk, at http://­ catalogues.indology.info/, I could retrace a number of manuscripts of the SSS from the locations indicated by Raṅgācārya. I gave the list in the bibliography of my article. A quick look at the list serves to demonstrate that a serious review of the available manuscripts needs to be done, as does perhaps a new critical edition. 80 Raṅgācārya (Raṅgācārya 2006: p.  v) also indicates that the shorter title Sarvasiddhāntasaṅgraha appears in the Tamil Pramāṇattiraṭṭu of Maṇavālamāmuni, the renowned South Indian Vaiṣṇava teacher. It is also known as such in the catalogue of the Sanskrit Manuscripts of the India Office Library, in London. 81 The twelve chapters are (1) Introduction (upodghāta-prakaraṇa); (2) The Thesis of the Lokāyatas (lokāyatika-pakṣa-prakaraṇa); (3) The Thesis of the Ārhatas (ārhatapakṣa-prakaraṇa), Jains; (4) The Thesis of the Bauddhas (bauddha-pakṣa-prakaraṇa); (4a) The opinion of the Mādhyamikas (mādhyamika-mata); (4b) The opinion of the Yogācāras (yogācāra-mata); (4c) The opinion of the Sautrāntikas (sautrāntikamata); (4d) The opinion of the Vaibhāṣikas (vaibhāṣika-mata); (5) The Thesis of the Vaiśeṣikas (vaiśeṣika-pakṣa- prakaraṇa); (6) The Thesis of the Naiyāyikas (naiyāyikapakṣa-prakaraṇa); (7) The Thesis of Prabhākara (prabhākara-pakṣa-prakaraṇa), Mīmāṃsā; (8) The Thesis of Bhaṭṭācārya (bhaṭṭācārya-pakṣa-prakaraṇa) Mīmāṃsā; (9) The Thesis of Sāṅkhya (sāṅkhya-pakṣa-prakaraṇa); (10) The Thesis of Patañjali (patañjali-pakṣa-prakaraṇa) Yoga; (11) The Thesis of Vedavyāsa (vedavyāsa-pakṣaprakaraṇa) as expressed in the Mahābhārata; and (12) The Thesis of the Vedānta (vedānta-pakṣa-prakaraṇa). 82 See Hacker 1978. The same opinion of Hacker on the authorship of the SSS has also been reproduced in Halbfass (1995: p. 51–54). 83 Belvalkar 1929: p. 222. 84 The Sri Vani Vilas Pressedition (1910), in twenty volumes, does not include the SSS, nor does Karl H. Potter’s Encyclopedia of Indian Philosophies, in its volume on Advaita Vedānta Up to Śaṃkara and His Pupils(1981). 85 Hacker regards as certain the Śaṅkara’s authorship of the commentaries on the ten principal Upaniṣads: Īśā, Aitareya, Kaṭha, Kena, Chāndogya, Taittirīya, Praśna, Bṛhadāraṇyaka, Māṇḍūkya (traditionally united with Gaudapāda’s Kārikā), and Muṇḍaka. Śaṅkara’s authorship for the bhāṣya on Śvetāśvatara Upaniṣad is less certain. See Hacker 1978: p. 48–49. 86 On the Bhagavadgītābhāṣya,see Mayeda 1965a. On the plausible attribution of the Pātañjala-yogaśāstra-vivaraṇa to Śaṅkara, see Harimoto 1999: p. 36–62. In his more recent book, Harimoto picks up this issue anew with more reserve. See Harimoto 2014: p. 225–251. 87 On the Upadeśasāhasrī, see Mayeda 1965b. 88 Hacker mentions two German scholars, one against the authorship of Śaṅkara, Hermann Jacobi, and one in favour of it, Bruno Liebich. He refers to the following publications: Jacobi 1921 and Liebich 1923. Liebich argues for the ‘possibility’ of the text being from the bhāṣyakāra. However, Hacker rejects the arguments as inconclusive on

The beginnings of Advaita doxography  175 the ground that they do not discuss the Vedānta chapter of the SSS. As for Jacobi, his comments on the SSS are rather succinct, appearing at the end of a book review, and they have been answered in the discussion of Liebich, as I will point out. Therefore, they need not be treated separately in here. 89 Some of the ‘pieces of evidence’ advanced by Gokulbhai (2009: p. 1–54) may contribute to locating the SSS in the literary and philosophical context of India. For example, he points out that the SSS does not distinguish between the opinions of the śvetāmbara and digambara sects of Jainism, which was already distinguished in 1220 AD by Śivabhūti; this might indeed signify that the text reflects an earlier state of affairs, where no distinction was made between these two sects. However, Gokulbhai assumes that this earlier stage in the development of Jainism was the one prevailing in Śaṅkara’s time (seventh–eighth century CE) and that this fact constitutes a solid evidence of Śaṅkara’s authorship, which is hardly the case. If anything, it can only indicate that the SSS might be earlier than the early thirteenth century. Yet, the information provided in each point is not totally devoid of interest in relation to other issues unrelated to the question of authorship. 90 Bouthillette 2019. 91 See Raṅgācārya 2006: p. x. 92 Hacker 1978. 93 Gokulbhai also discusses several elements found in the Patañjali-pakṣa of the SSS, noticing that some are not found in the YS but are rather related to haṭha-yoga. He nevertheless maintains the authorship of Śaṅkara in every respect. See Gokulbhai 2009: p. 146–147. 94 As suggested by Kengo Harimoto 1999, 2014. However, there are objections to the idea, notably Rukmani 1993. For a summary of the authorship debate and for a discussion on the possible date of the Vivaraṇa, see Maas 2013: p. 73–75. 95 The Prasthānatrayī, literally the “three sources,” refers to the three canonical texts of the Vedānta: Upaniṣads, Brahmasūtra, and Bhagavad Gītā. 96 See Pande 1994: p. 109. 97 However implausible that lifespan might be, let’s observe that, in the case where it would be accurate, one who is supposed to have lived only 32 years, as Śaṅkara, can hardly have had a meaningful ‘early’ career. 98 Hacker 1968. 99 Pande 1994: p. 111. 100 BhN 6.9 vistareṇopadiṣṭānām arthānāṃ sūtrabhāṣyayoḥ | nibandho yo samāsena saṃgrahaṃ taṃ vidurbudhāḥ || 101 102

103 104 105

The translation is mine. For various editions of the BhN, see Vatsyayan 1996 (reprint); Ghosh 1951. See Laks 2007: p. 27–29. In his notes on a series of conferences on Indian doxography which he gave at l’École pratique des hautes études, Gerdi Gerschheimer, beyond the term saṅgraha, for ‘compendium,’ also enumerates other Sanskrit words which indicate a doxographical content. He adds samuccaya and saṃkṣepa, explaining that they designate texts dedicated to exposing in a concise manner views or established doctrines which, in Sanskrit, are designated by the terms darśana, siddhānta, and mata. See Gerschheimer 2000–2001: p. 174. I am prone to believe that if Śaṅkara had composed the original core of the SSS, for the sake of his students, he would have had no need of summarizing his own doctrine. He would simply have explained how it comes as the final word, refuting all others. For a brief overview of the topic, see Laks 1999: p. 49. Halbfass 1988: p. 351.

176  The beginnings of Advaita doxography 106 Gokulbhai 2009: p. 128–131. 107 Ganeri 2012: p. 18–19. 108 Angot 2009: p. 115. My free translation: “What matters is not to tell the truth – it, in the sense that we generally understand it, does not exist at the level of speech – but to tell a truth that extends the one to which we previously had access.” 109 Hirst 2005: p. 84. 110 For further comments on Śaṅkara’s pedagogical method, see also Hirst 2004, 1996. 111 This pedagogy is illustrated in the following Chā Up (7.1.1) passage: adhīhi bhagaveti ha upassāda sanatkumāraṃ nāradas taṃ ha uvāca yad vettha tena mopasīda tatas ta ūrdhvaṃ vakṣyāmi iti sa hovāca|| Do teach, Venerable,” said Nārada as he approached Sanatkumāra. To what he replied: come to me with that which you know, I shall explain from that level on. 112 In his prefatory remarks to the Guide, Maimonides stated his intention as follows: “The object of this treatise is to enlighten a religious man who has been trained to believe in the truth of our holy Law, who conscientiously fulfills his moral and religious duties, and at the same time has been successful in his philosophical studies. Human reason has attracted him to abide within its sphere; and he finds it difficult to accept as correct the teaching based on the literal interpretation of the Law, and especially that which he himself or others derived from those homonymous, metaphorical, or hybrid expressions. Hence he is lost in perplexity and anxiety” (Friedländer 2002 [1st Ed. 1904]). Similarly, the students of Indian doxographies also aim at conscientiously fulfilling their moral and religious duties. This is precisely why they train in philosophical studies. But they need guidance in the face of philosophical opinions going against their cherished religious doctrine. 113 It has been argued that the Advaitins had extracted their main ideas and dialectics from Buddhism and other competing heterodox teachings, their doctrine consisting essentially in a doctrinal agglomeration. For an overview of the subject, see Isayeva 1993: p. 10–18. Others have argued that early Advaita doxographies already expressed a desire for sectarian unification, see Nicholson 2010. But the motive of this conglomeration of views might also be religio-philosophical in nature. I suggest considering this attitude towards ‘the other’ as a dialectical propaedeutic supporting a form of ‘spiritual exercise,’ in addition to the previous readings. 114 Among the Vedānta sources relied on by Bhāviveka can be found the Gauḍapādiyakārikā (GK), a text of substantial importance for the doctrinal development of Śaṅkara. On the relation between Bhāviveka and the GK, see King 1995: p. 35–43. 115 See MHK 8, 86–104. For an English translation of the passages, one is advised to look in Qvarnström (2015: p. 185–190). Obviously, Bhāviveka’s intent is polemical. But he was not the last one to make such observations. It is well known that Rāmānuja, a competing Vedāntin of the Viśiṣṭādvaita filiation, pejoratively labelled Śaṅkara a ‘crypto-Buddhist’ (pracchanna-bauddha), for his apparent adoption of Buddhist ideas and customs. See Isayeva 1993: p. 14. 116 I argue that the ‘soteriological’ significance of doxographical expositions, composed in a manner which varies within each religious system, has been overlooked in scholarship, and that this is why many are at pain to explain the relevance of such writings. Although the suggestion seems to betray a serious anachronism, the tendency is to identify Indian doxographical writings with a similar literature available in modern times: the introductory manuals to philosophy. Gokulbhai, for example, suggests, “The only reason of writing the SSS seems to be the easy grasping of the tenets of all the philosophical systems prevalent, or rather famous in his times in India” (Gokulbhai 2009: p. 26). Halbfass also presented doxographies as an unimpressive “by-product” of the Indian philosophical tradition which have unfortunately not “evolved” to become like the “Western literature on the history of philosophy” (Halbfass 1988:

The beginnings of Advaita doxography  177 p. 355–356). I argue that such opinions betray their holders’ inability to understand Indian doxography’s method and purpose. They rather refer to a modern understanding of ‘doxography’ which is foreign to the Indian context. 117 Hirst 2005: p. 85. 118 Listed together with the Prābhākara-s, it is not clear who are these Tārkika-s referred to in this passage (VS 129), which are supposedly holding that ignorance is the self. 119 The doxographical section begins with the following verse (VS, 122): idānīṃ pratyagātmanīdam idam ayam āropayatīti viśeṣata ucyate | Now will be described the peculiar way so and so imposes this and that upon the individuated self 120 For the definition of parīkṣā within the YBhŚ, see Eltschinger 2014: p. 16–19. 121 Eltschinger 2014: p. 192. 122 On the function of reasoning within Śaṅkara’s work, and Advaita in general, see RamPrasad 2000, 2001, 2007; Comans 2000; King 1995; Murty 1974. 123 For a historical retrospective of the topic, see Isayeva 1993: p. 19–30. See also RamPrasad 2001, 2007; King 1995; Kaplan 1992; Potter et al. 1981; Radhakrishnan 1914, 1929; Whaling 1979. 124 Hugh Nicolson, for example, remarked, “Even a cursory glance at the text shows that refutations of opposing views compose a large portion of the work” (Nicholson 2007: p. 530–531). 125 See Halbfass 1983: p. III–IV. 126 Translation taken from Gambhirananda 1965: p. 368. 127 Eltschinger 2014: p. 18–19. 128 Liebich 1923: p. 127. 129 Maas 2013: p. 60, 66. 130 I base my reading on Bose 1929; Gokulbhai 2009; Raṅgācārya 2006; Śukla approx. 1930 (undated print). 131 Nicholson 2015: p. 164, n. 152. 132 On the unknown dates of Prabhākara (seventh–eighth centuries CE) and his uncertain connection to Kumārila, to whom A. Subrahmany Sastri suggests that he would have been either contemporary or subsequent, see Potter 2014: p. 295–296. 133 Gokulbhai 2009: p. 173–174.

Conclusion

By looking at the earliest doxographies of Madhyamaka Buddhism, Jainism, and Advaita Vedānta, the present study highlighted three dialectical teleologies peculiar to each tradition. By organizing philosophical views in these specific manners, doxographies manage to both level a certain criticism of the doctrinal content of each darśana and establish a dialectical path to ‘right view.’ Each teleology reproduces a dialectical attitude which can be traced back to the narrative forms and pedagogies of its specific tradition. This should come as no surprise since these dialectical methods, by conveying specific propaedeutics to truth, characterize the spiritual identity of each tradition: their own conception of the way to go about cultivating liberative knowledge. It is only natural that Indian doxographies display idiosyncrasies in tune with their respective worldview. Within each separate doxography I reviewed, I have used the view of the Mīmāṃsā as a single dialogical partner, to better contrast the view promoted by the doxographical authors. I considered it skillful to insist on the dialectic of ‘knowledge’ (jñāna) versus ‘ritual action’ (karma). It highlighted the importance the three authors give to knowledge, its nature and means of acquisition. It opened a discussion on the transformative function of this particular form of knowledge transmission, which we call doxographical. It informed my assessment of the strategies adopted by each author to guide their readers. For them, doxography, like philosophy, is not an end in itself, but a means to an end – a literal cessation (nirvāṇa). Their yoga of reasoning is meant to cultivate insight, to see what the awakened sees. Here, doxography becomes a simple intellectual exercise in philosophical positions. There, it is the very grammar of a systematic analysis of the language of reality. In between, it is a political discourse. It must be said that in no way can the depictions of the Mīmāṃsā offered by these three texts pay justice to the school’s profound learning. This is not the point here. Hence I did not consider it necessary to include a separate treatment discussing the tenets of the Mīmāṃsā, but I introduced them nonetheless, with parsimony, within my analysis. The focus was rather on the views of my doxographers, to understand their doxographies. For Madhyamaka Buddhism, it was shown how the dialectic of no-view underpins the thematic progression of Bhāviveka’s MHK, directing its teleology. The first three chapters of the whole text form a single dialectical unit. It establishes

Conclusion  179 the author’s perspective through a negation of Abhidharma categories. Thereafter, every succeeding doctrine is refuted in a movement going from the most acceptable position, from a Mādhyamikan perspective, to the most problematic one, leaving the Mīmāṃsā’s karma-mārga at the extreme end of a downward spiral into absurdity. This reductio ad absurdum of world philosophy is typical of Madhyamaka dialectics. Bhāviveka’s dialectical model, in the MHK, can be said to be a declining teleology, a fall away from truth. The ultimate perspective is first provisionally established through the corrective negation of the most acceptable conventional standpoint. From there on, Bhāviveka continues to grind to pieces all other mundane views. The thematic progression, from system to system, is a plunge into intellectual decay and doctrinal incoherence. Perhaps this ‘dive,’ or ‘descent,’ was also meant to be conveyed by the term avatāra (entry), found in five out of six titles of the MHK chapters dedicated to refuting opposing systems, with the exception of the Vedānta chapter. Wherever one finds oneself along these descending doctrinal stages, one is called upon by Bhāviveka to use reason, in order to stop going further down. To undertake a revolutionary journey backward, step by step, to the original state of reality. For the sake of mental clarity, Madhyamaka reasoning (tarka) serves as the spinning engine of the spiritual vehicle (yāna), the pillar of conceptual wisdom. A common metaphor of Bhāviveka to describe this conceptual stage of understanding the Buddha’s teachings is that of a ‘dream palace,’ beautifully commented upon by Eckel: Bhāviveka starts his analysis of reality with a reference to the steps of a palace. The steps that lead to the top of the palace represent stages in a process of reasoning that leads from the realm of ordinary thought to the realm of ultimate truth. The goal of the philosopher is to “climb” to the top of the palace and see reality as it truly is. When the image of the palace next appears, it represents reality with the ordinary values reversed: the structure that the philosopher once had to climb through a laborious process of conventional study becomes a palace seen in a dream.1 The process of moving up the stairs of the illusory palace of worldly conventions serves as a metaphor for the jñāna-mārga, what I call philosophical yoga. One may ask, what is the result of ultimate sight, once established? Bhāviveka suggests, The learned burn misconduct with knowledge, just as fire [burns] fuel. Hence, because of their peculiar fire, there is no offense produced of them. The learned ones understand this as a knowledge suitable for burning away afflictions. Hence, the learned is not inclined to misconduct, due to the nonarising of its cause.2 The jñāna-mārga is no mere mental game, some infinite accumulation (saṅgraha/samuccaya) of philosophical theories (darśana-s), captivating the

180  Conclusion mind, as in the seducing knowledge promises of the flesh-eating sirens to Ulysses. This wandering mendicant hero could hear their maddening songs without incurring self-destruction only because he restrained himself by the mast of his vessel (yāna). Rather, the jñāna-mārga is a self-transformative study of one’s internal sight and behaviour. In this symbolic context, doxographies appear as conceptual maps. Every view they present is that obtained from a given perspective on the stairway to heaven. However, to reach the top of the palace and obtain the ultimate perspective, one must discern the conventional logic of the map’s worldview, its indications and geography. Then only can one be enthroned as the cakravartin of the whole conceptual realm and establish one’s view. Doxographies are no neutral maps. They do not merely display the blueprints of the edifice of philosophy without indications. Rather, they offer guidance through a carefully designed dialectical teleology. That teleology is the underlying ‘message’ of the ‘medium’ of doxography. Bhāviveka’s own doxographical ‘message’ is articulated as a compassionate descent of the enlightened Bodhisattva through the darkness of the illusory stages of worldly ignorance, somewhat like the philosopher in Plato’s cave; stopping at every station in order to cast light upon the confusion of those who dwell therein. It is through this painstaking dialectical struggle, this grinding (tarka) of every conceptual view, that the healing ‘ambrosia of truth’ (tattva-amṛta) is benevolently brewed and offered to all, whatever their station, to quench their existential ‘thirst’ (tṛṣṇā). Drinking at the cup of knowledge reveals the illusory nature of the world maṇḍala, symbolically divided into six philosophical realms, at the doorway of which stands Bhāviveka’s stance on Abhidharma. As for Haribhadra’s ṢDS, its dialectical model is one of equipoise. Yet, every view is in balance only from the perspective of the absolute. From that perspective alone is one able to perceive the relativizing expression of syāt at the end of each doctrinal pronouncement. In the relative realm, where every logician defends some doctrinal superiority over another, only the way of the Jina, as encapsulated within the anekānta-vāda, able to perceive the limited validity of every perspective, is suited to direct one towards peace, the realization of pure equanimity, where consciousness is freed of polluting karma-s: among which are the ‘wrong views’ (mithyā-darśana-s) of competing systems. When reified as one-sided truth claims, these opposing views must be silenced through intellectual exertion (jñāna-yoga). For the followers of the Jina, the jaina-mārga is the axis mundi of philosophy. It developed all the necessary conceptual tools to abide by the salutary middle way and the purifying ascetic disciplines suited to remove obscuring karma-s. Through the ṢDS, where one learns to pronounce the liberative syād-vāda by oneself, Haribhadra engages his students in a propaedeutic to truth. This dialectical training revolves within the conceptual realm. Nonetheless, it gravitates towards the non-conceptual axis of truth, the ultimate reality, using heterodoxy as a training ground for engaging in a multiplexed (anekānta) soteriological method designed to lay the foundations of omniscience. In the end, views are perspectives. The superior ‘perfect’ vision

Conclusion  181 (samyag-darśana) of the omniscient one (sarvajña) is no longer bound to them. Pure light knows no shade. When it comes to Śaṅkara’s SSS, where every view negates the preceding, ever progressing upwards towards the throning Advaita Vedānta position, the dialectical model is a gradual ascension. Every further step on the stairway to enlightenment broadens the previous perspective. It is a movement towards the overarching truth of the Advaita, the worldly expression of which can only be found within scriptures, acting as fingers pointing at the moon of ultimate truth. The journey is said to culminate in pure brahman, where all views, all singulars, subside in the one general category, the only being (sat), the cosmic axis mundi. Perhaps one could talk of the Advaita methodology as a dialectical movement from the particular to the universal, using doxographical categories as instantiations of necessary particulars to be dissolved as they are outgrown by a wider truth. They are necessary only in the sense that they represent various stages of ignorance found among worldlings and, most importantly, in oneself. Appearing as general models of thought, they serve a skillful pedagogy. As with the MHK, wherever one finds oneself along the doctrinal ladder of a doxography, one is called upon to methodologically proceed from this particular position towards the overarching truth, in an apophatic process of discrimination (neti neti). Yet, in theory, the SSS is the inverted mirror of the MHK. For, it is articulated as a progression towards truth and not as a return to its pure origins. But this might only be a matter of perspective. In practice, the SSS contributes to a rediscovery of one’s ‘true nature,’ pure consciousness. And this ultimate state does not exist among the views listed in texts and debated by scholars. Besides the soteriological dimension of Indian doxography, I have argued that these texts appear to convey a religio-political vision of the status and function of the doxographers’ tradition within the world they knew. I have discussed how the communities involved in doxographical productions were affected by sociohistorical developments in which they had to take a stance in order to secure their survival. Since archaeological evidence and historical data are missing, in order to build a full picture of the social turmoil of the days, it is not a luxury to resort to an analysis of the philosophical production of the time as an additional historical window. Doxographies can also be useful for such endeavours. Yet, one should keep in mind that their primary purpose lies somewhere else. I suggest reading such literature as spiritual exercises. I interpret early Indian doxographies as the locus of interaction between the sacred and the profane, the religious and the political. Embedded in this fertile ideological soil, their purpose takes on another dimension: to provide guidance, protection, and certainty. If one looks at the term darśana in the context of devotion, a completely different environment than the philosophical one which was the focus of this study, the same three purposes are found again. Indeed, darśana of the lord, whether he be a king, an awakened being, or a god, appearing on a coin, an icon, or in contemplation, is a confirmation, a blessing, and a protection. It is the locus of all veneration, the guarantor of truth. Anything conveying that royal

182  Conclusion emblem is an object of power. Consequently, it is clear that no doxography can pretend to be purely objective or neutral in the face of opposition. It is always an authoritative stance. By establishing the only valid darśana, doxographies show where truth and power dwell. Much research remains to be done in order to produce a general account of all Indian doxographies. So far, we do not even have a complete picture of the extent of the Sanskrit doxographical materials available or known to have existed. Only a few texts are published in translation. Would it be relevant to formulate a genealogy of each dialectical structure found within different sectarian filiations, including a listing of the various topics included or absent in every compendium? Such genealogies would considerably help to reflect on the genre as a whole, for example to appraise the evolution of a tradition’s self-perception through time, its gradual shifts in focus and methods in regard to doctrinal debates. Besides the adoption of prose and the more detailed discussions found in later doxographies, did their attitude to heterodoxy evolve at all? Perhaps not. But, could there be a shift in the sectarian perceptions of competing views observable through time, perhaps affected by known sociohistorical developments? What other surprises are there to be seen? The field of Indian doxography is fertile. When read in the context of the interplay of intellectual exertion and spiritual exercises, doxographical texts also provide a valuable insight into the praxis of philosophy as an intellectual form of yoga. On a related note, but now going backward in time, the phenomenon of ‘list making’ in Indian philosophical literature – be it in the numbered lists of the old Sāṃkhya system, in the mātikā-s of Buddhist Abhidharma,3 or in the broader lexes of an emerging pan-Indian scholasticism, listing various sets of padārtha-s and tattva-s to be put to memory – suggests a mode of knowledge transmission connected to contemplative practices shared across the sectarian faultlines of ancient India. The phenomenon requires more scrutiny, if only to retrace the origins of Indian doxography further back within the oral culture of the subcontinent and to better envision the interdependence of intellectual learning, narratives, and spiritual exercises in India. It is possible to imagine that, progressively, through a process of ongoing systematization, lists of disparate and fragmented representations of the world, different stories, shared by words of mouth, were synchronized, either in recitation practices, while memorizing texts, or by other symbolic means. One can read this scholastic process as a ritualization of knowledge transmission. In this context, the three stages of acquiring wisdom mentioned by our three doxographers display a logical, quasi-organic, progression, a learning process similar to traditional pedagogy. In order to gain the fruit of religious life, the student must first hear the words (śabda-s) of scriptures and store them in his or her mind. Out of this primary seed, cultivated by further reasoning and logical enquiry, the meaning of these words is eventually revealed. Only then, when the seed grows, can reality (tattva), which scriptures point at, be apprehended. At this crucial point, however, whatever is the chosen name given to ‘knowledge,’ be it ātman, brahman, puruṣa, jīva, or śūnya, it is but a concept, reappearing in the conventional wheel of doxography.

Conclusion  183

Notes 1 Eckel 1992: p. 23. 2 MHK 9.60–61 jñānena jñāninaḥ pāpaṃ dahaty agnir ivendhanam | atas tejoviśeṣāc ca na teṣāṃ pratyavāyitā || yat kleśadahanāyālaṃ tajjñānaṃ jñānino viduḥ | nātaḥ prakurute pāpaṃ jñānī taddhetvasaṃbhavāt || 3 Reflecting on the mnemonic use of lists in early Buddhism and in the broader Indian cultural context, Rupert Gethin (1992: p. 149) noticed that: “a penchant for analyzing something in terms of a neat categorized list is characteristic of much of traditional Indian learning, and the oral origins of Indian learning continued to inform its structure long after its exponents had begun to commit it to writing.” Further on, Gethin suggests that “The lists actually inform and to some extent govern the structure of the literature [p. 155). This last feature seems to foretell the dialectical teleology of Indian doxographies.   More importantly, in relation to Abhidharma, Gethin added that: “the act of reducing suttas to lists was seen, I think, as laying the Dhamma bare and revealing its inner workings. Thus the various composite lists might be viewed as different ways of getting at the structure lying at the very heart of the Dhamma [p. 157). Gethin argues that Buddhist lists are born of vibhaṅga (analysis), breaking the world of experience into parts, and that they are meant to be performed, chanted and memorized, to support mindfulness and meditation practices. In the same way, I suggest that later doxographies, which likely evolved out of such lists making practices, were contemplative literary technologies meant to dissolve heterodoxy, to break competing views into their constitutive parts, to assess their logic, in order to establish ‘right view’ (samyag-darśana) in meditation. For further readings on list making in early Buddhism and Abhidharma, see Bronkhorst 1985; Ronkin 2005.

Bibliography

Abhedananda, S. 1936. ‘Hindu Philosophy in India’, in S. Radhakrishnan and J. H. Muirhead (eds.), Contemporary Indian Philosophy, 47–62 (New York: MacMillan). Aklujkar, Ashok. 2017. ‘History and Doxography of the Philosophical Schools’, in Jonardon Ganeri (ed.), The Oxford Handbook of Indian Philosophy (Oxford University Press: New York). Ames, William Longstreet. 1985. ‘Bhāviveka’s Prajñāpradīpa: Six Chapters’, University of Washington. Angot, Michel. 2009. Le Nyāya-Sūtra de Gautama Akṣapāda, Le Nyāya-Bhāṣya d’Akṣapāda Pakṣilasvāmin, L’art de conduire la pensée en Inde ancienne (Les Belles Lettres: Paris). ———. 2012. Le Yoga-Sūtra de Patañjali, le Yoga-Bhāṣya de Vyāsa, avec des extraits du Yoga- Vārttika de Vijṇāna-Bhikṣu, La parole sur le silence (Les Belles Lettres: Paris). Baber, Zaheer. 2000. ‘Religious Nationalism, Violence and the Hindutva Movement in India’, Dialectical Anthropology, 25: 61–76. Bader, Jonathan. 1991. ‘Conquest of the Four Quarters. Traditional Accounts of the Life of Śaṅkara’, The Australian National University. ———. 2000. Conquest of the Four Quarters: Traditional Accounts of the Life of Śaṅkara (Aditya Prakashan: New Delhi). Bahulkar, S. S. 1994. ‘The Madhyamaka-Hṛdaya-Kārikā of Bhāvaviveka: A Photographic Reproduction of Prof. V.V. Gokhale’s Copy’, Nagoya Studies in Indian Culture and Buddhism Saṃbhāṣā, 15: 1–49. Balcerowicz, Piotr. 2003. Historia, klasycznej filozofii indyjskiej, Czêœæ pierwsza: poczatki, nurty analityczne i filozofia przyrody (Wydawnictwo Akademickie Dialog: Warszawa). ———. 2015. ‘Introduction’, in Piotr Balcerowicz and Karl H. Potter (eds.), Encyclopedia of Indian Philosophies, Jain Philosophy (Part Three) (Motilal Banarsidass: Delhi). ———. 2017. ‘Jainism. Disambiguate the Ambiguous’, in Joerg Tuske (ed.), Indian Epistemology and Metaphysics (Bloomsbury Academic: London, Oxford and New York). Ballanfat, Marc. 1997. Les matérialistes dans l’Inde ancienne, Préface de Pierre-Sylvain Filliozat (L’Harmattan: Paris). Baltussen, Han. 2005. ‘The Presocratics in the Doxographical Tradition. Sources, Controversies, and Current Research’, Studia Humaniora Tartuensia, A.6: 1–26. Banchoff, Thomas (ed.). 2008. Religious Pluralism, Globalization, and World Politics (Oxford University Press: New York). Basham, A. L. 1951. History and Doctrines of the Ājīvikas: A Vanished Indian Religion (Motilal Banarsidass: Delhi). repr. 1981.

Bibliography  185 Banerjee, Sumanta. 1991. ‘“Hindutva”: Ideology and Social Psychology’, Economic and Political Weekly, 26: 97–101. Barbato, Melanie. 2017. Jain Approaches to Plurality: Identity as Dialogue (Brill: Leiden). Belvalkar, S.K. 1929. Sri Gopal Basu Mallik Lectures on Vedānta Philosophy (Bilvakunja Publishing House: Poona). Bergson, Henri. 1899. Le rire: essai sur la signification du comique (Revue de Paris: Paris). Bhatla, Neeraj, Tapan Mukherjee, and Gian Singh. 1984. ‘Plants: Traditional Worshipping’, Indian Journal of History of Science, 19: 37–42. Bhaṭṭācārya, Ramkrishna. 2011. Studies on the Carvaka/Lokayata (Anthem Press: London and New York). Biardeau, Madeleine. 1976. ‘IV. Études de mythologie hindoue (IV) [II. Bhakti et avatàra] II. Bhakti et avatàra’, Bulletin de l’Ecole française d’Extrême-Orient, 63: 111–263. Birch, Jason. 2013. ‘Rājayoga: The Reincarnations of the King of All Yogas’, International Journal of Hindu Studies, 17: 401–444. Blom, Philipp. 2004. Encyclopédie: The Triumph of Reason in an Unreasonable Age (Fourth Estate: London and New York). ———. 2005. Enlightening the World: Encyclopédie, The Book that Changed the Course of History (Palgrave Macmillan: New York). Bose, Prem Sundar (ed.). 1929. Sarva-Siddhanta-Samgraha. Critically edited, translated and annotated by Prem Sundar Bose (Navavidhan Press: Calcutta). Bouthillette, Karl-Stéphan. 2015. ‘Taking the Enemy as Medicine: Dialectic and Therapy in the Work of Two Early Indian Doxographers’, Religious Inquiries, 4: 51–64. ———. 2017a. ‘Battle for Middle Way, Bhāviveka’s Dialectical Strategy in Context’, Distant Worlds Journal, 3: 67–79. ———. 2017b. ‘Hermeneutic Praxis: The Yoga of Reason(ing)’, Studia Religiologica, 50: 103–115. ———. 2019. ‘Reconsidering the Sarva-Siddhānta-Saṅgraha: Authorship, Doxography and Pedagogy.’ in Heleene De Jonckheere, Marie-Hélène Gorisse and Agnieszka Rostalska (eds.), Puṣpikā V: Tracing Ancient India, through Texts and Traditions: Contributions to Current Research in Indology (Oxbow Books: Oxford). Braarvig, Jens. 1997. ‘Bhavya on Mantras: Apologetics Endeavours on Behalf of the Mahāyāna’, Studia Indologiczne, 4: 31–39. Bronkhorst, Johannes. 1985. ‘Dharma and Abhidharma’, Bulletin of the School of Oriental and African Studies, 48: 305–320. ———. 2007a. Greater Magadha: Studies in the Culture of Early India (Brill: Leiden and Boston, MA). ———. 2007b. ‘Vedānta as Mīmāṃsā’, in Johannes Bronkhorst (ed.), Mīmāṃsā and Vedānta: Interaction and Continuity (Motilal Banarsidass: Delhi). ———. 2010a. ‘Reflections on the history of Jainism’, in International Conference on Jainism Through the Ages: An Historical Perspective, 8–10 October, 1–17. Mysore. ———. 2010b. ‘What Did Indian Philosophers Believe?’, in Piotr Balcerowicz (ed.), Logic and Belief in Indian Philosophy (Motilal Banarsidass: Delhi). ———. 2011. Buddhism in the Shadow of Brahmanism (Brill: Leiden and Boston, MA). ———. 2012. ‘Ājīvika’, in Knut A. Jacobsen et al. (eds.), Brill’s Encyclopedia of Hinduism (Brill: Leiden and Boston, MA). ———. 2013. ‘Who Is a Brahmin?’, in Collaborative Research Centre “Ritual Dynamics” in cooperation with the Cluster “Asia and Europe”, 1–34 (Heidelberg Universität, Karl Jaspers Centre: Heidelberg).

186  Bibliography ———. 2015. ‘The Historiography of Brahmanism’, in Bernd-Christian Otto, Susanne Rau and Jörg Rüpke (eds.), History and Religion: Narrating a Religious Past (Walter de Gruyter: Berlin and Boston, MA). ———. 2016a. How the Brahmins Won: From Alexander to the Guptas (Brill: Leiden). ———. 2016b. ‘Kundakunda versus Sāṃkhya on the Soul’, in Christopher Key Chapple (ed.), Yoga in Jainism (Routledge: London and New York). ———. 2017. ‘Brahmanism: Its Place in Ancient Indian Society’, Contributions to Indian Sociology, 51: 361–369. Brummett, Barry. 1996. ‘Rhetoric of Silence’, in Theresa Enos (ed.), Encyclopedia of Rhetoric and Composition: Communication from Ancient Times to the Information Age (Routledge: New York and London). Bunnin, Nicholas, and Jiyuan Yu. 2004. The Blackwell Dictionary of Western Philosophy (Blackwell Publishing: Malden,MA). Burnouf, Émile-Louis. 1861. La Bhagavad-Gîtâ ou Le chant du Bienheureux, Poème indien (L’académis de Stanislas: Paris). Burnouf, Eugène. 1844. Introduction à l’histoire du Bouddhisme Indien (Imprimerie Royale: Paris). Chakravarti, Uma. 1998. ‘Saffroning the Past: Of Myths, Histories and Right-Wing Agendas’, Economic and Political Weekly, 33: 225–232. Chandra, K.R. 1988. ‘Haribhadra Suri (Apabhramsha/Sanskrit)’, in Amaresh Datta (ed.), Encyclopaedia of Indian Literature: Devraj to Jyoti (Sahitya Akademi: New Delhi). Chapple, Christopher Key, and John Thomas Casey. 2003. Reconciling Yogas: Haribhadra’s Collection of Views on Yoga With a New Translation of Haribhadra’s Yogadrstisamuccaya by Christopher Key Chapple and John Thomas Casey (State University of New York Press: Albany, NY). Clayton, John. 2006. Religions, Reasons and Gods: Essays in the Cross-Cultural Philosophy of Religion, Prepared for Publication by Anne M. Blackburn and Thomas D. Carroll (Cambridge University Press: Cambridge). Coccia, Emanuele. 2016. La vie des plantes, Une métaphysique du mélange (Éditions Payot & Rivages: Paris). Collins, Randall. 2000. The Sociology of Philosophies: A Global Theory of Intellectual Change (Belknap Press of Harvard University Press: Boston, MA). Comans, Michael. 2000. The Method of Early Advaita Vedānta, A Study of Gauḍapāda, Śaṅkara, Sureśvara and Padmapāda (Motilal Banarsidass: Delhi). Cort, John E. 2000. ‘“Intellectual Ahiṃsā” Revisited: Jain Tolerance and Intolerance of Others’, Philosophy East and West, 50: 324–347. Couture, André. 2001. ‘From Viṣṇu’s Deeds to Viṣṇu’s Play, or Observations on the Word Avatāra as a Designation for the Manifestations of Viṣṇu’, Journal of Indian Philosophy, 29: 313–326. Cush, Denise, Catherine Robinson, and Michael York (eds.). 2008. Encyclopedia of Hinduism (Routledge: London and New York). Davidson, Ronald M. 2002. Indian Esoteric Buddhism: A Social History of the Tantric Movement (Columbia University Press: New York). Desai, Radhika. 2016. ‘Hindutva and Fascism’, Economic and Political Weekly, LI: 20–24. Diels, Hermann Alexander. 1879. Doxographi Graeci. Collegit recensuit prolegomenis indicibusque instruxit Hermannus Diels Opus adademiae litterarum regiae Borussicae praemio ornatum. (G. Reimer, Lge 8vo: Berlin). Doniger, Wendy. 2018. Against Dharma, Dissent in Ancient Indian Sciences of Sex and Politics (Yale University Press: New Haven and London).

Bibliography  187 Dreyfus, G. 2003. The Sound of Two Hands Clapping: The Education of a Tibetan Buddhist Monk (University of California Press: Berkeley, CA). Dundas, Paul. 1996. ‘Jain Attitudes Towards the Sanskrit Language’, in J.E.M. Houben (ed.), Ideology and Status of Sanskrit: Contributions to the History of the Sanskrit Language (Brill: Leiden). ———. 2006a. ‘The Later Fortunes of Jamāli’, in Peter Flügel (ed.), Studies in Jaina History and Culture, Disputes and Dialogues (Routledge: London and New York). ———. 2006b. ‘A Non Imperial Religion? Jainism in Its “Dark Age”’, in Patrick Olivelle (ed.), Between the Empires: Society in India 300 BCE to 400 CE (Oxford University Press: New York). Eckel, Malcolm David. 1992. To See the Buddha: A Philosopher’s Quest for the Meaning of Emptiness (Princeton University Press: Princeton). ———. 2008. Bhāviveka and His Buddhist Opponents (Harvard University Press: Cambridge, MA). Edwards, S.M. 1922. ‘Tree-Worship in India’, Empire Forestry Journal, 1: 78–86. Ejima, Yasunori 江島恵教. 1980a. ‘Bhāvaviveka/Bhavya/Bhāviveka [in Japanese]’, Indogaku Bukkyōgaku Kenkyū 印度學佛教學研究, XXXVIII: 98–106. ———. 1980b. Chūgan shisō no tenkai – Bhāvaviveka kenkyū 中観思想の展開: Bhāvaviveka 研究 [An Exposition of Madhyamaka Thought. A Study of Bhāvaviveka] (Shunjūsha 春秋社: Tokyo). Eltschinger, Vincent. 1998. ‘Bhāvaviveka et Dharmakīrti sur āgama et contre la Mīmāṁsā’, Asiatische Studien/Études Asiatiques, 52: 57–84. ———. 2007. Penser l’autorité des Écritures: la polémique de Dharmakīrti contre la notion brahmanique orthodoxe d’un Veda sans auteur; autour de Pramāṇavārttika I.213–268 et Svavṛtti (Verlag der Österreichischen Akademie der Wissenschaften: Wien). ———. 2008. ‘Pierre Hadot et les “Exercices Spirituels”: Quel Modèle pour la philosophie Bouddhique Tardive?’, Asiatische Studien Études Asiatiques, LXII: 485–544. ———. 2010. ‘Studies in Dharmakīrti’s Religious Philosophy: The Cintā-mayī-prajñā’, in Piotr Balcerowicz (ed.), Logic and Belief in Indian Philosophy (Motilal Banarsidass: Delhi). ———. 2014. Buddhist Epistemology as Apologetics: Studies on the History, Self-understanding and Dogmatic Foundations of Late Indian Buddhist Philosophy (Ôsterreichisch Akademie Der Wissenschaften: Wien). Eltschinger, Vincent, H. Krasser, and J. Taber. 2012. Can the Veda Speak? Dharmakīrti against Mīmāṃsā Exegetics and Vedic Authority: An Annotated Translation of PVSV 164,24–176,16 (Verlag der Österreichischen Akademie der Wissenschaften: Wien). Emilsson, Eyjólfur Kjalar. 2007. Plotinus on Intellect (Clarendon Press: Oxford). Engelmajer, Pascal. 2014. Women in Pāli Buddhism: Walking the Spiritual Paths in Mutual Dependence (Routledge: New York). Fiordalis, David V. (ed.). 2018. Buddhist Spiritual Practices. Thinking with Pierre Hadot on Buddhism, Philosophy, and the Path (Mangalam Press: Berkeley, CA). Flood, Gavin, and Charles Martin. 2015. Bhagavad Gītā: A New Translation (W.W. Norton & Company: London). Flügel, Peter. 2009. ‘Power and Insight in Jain Discourse’, in Piotr Balcerowicz (ed.), Logic and Belief in Indian Philosophy (Warsaw Indological Studies 3) (Motilal Banarsidas: Delhi). Folkert, Kendall Wayne. 1975. ‘Two Jaina Approaches to Non-Jainas: Patterns and Implications’, Harvard University. ———. 1993. Scripture and Community: Collected Essays on the Jains (Scholars Press: Atlanta).

188  Bibliography Foucault, Michel. 2001. L’herméneutique du sujet, Cours au Collège de France. 1981– 1982 (EHESS, Gallimard, Seuil: Paris). ———. 2013. L’origine de l’herméneutique de soi, Conférences prononcées à Dartmouth College, 1980 (Vrin: Paris). Freschi, Elisa. 2017. ‘What Does Nitya Mean in Mīmāṃsā?’, Journal of UA Foundation for Indological Studies, 3: 1–14. Friedländer, Michael. 2002 (1st Ed. 1904). The Guide for the Perplexed by Moses Maimonides (Varda Books: Skokie). Fujinaga, Sin. 1999. ‘Kundakunda on Sarvajna’, Journal of Indian and Buddhist Studies, 47: 23–26. Fuller, Paul. 2005. The Notion of Diṭṭhi in Theravāda Buddhism. The Point of View (RoutledgeCurzon: Oxon). Gambhirananda, Swami. 1965. Brahma-Sūtra-Bhāṣya of Śri Śaṅkarācārya (Advaita Ashrama: Delhi). Ganeri, Jonardon. 2001a. Indian Logic: A Reader (Routledge: Abingdon and New York). ———. 2001b. Philosophy in Classical India, The Proper Work of Reason (Routledge: London). ———. 2012. The Concealed Art of the Soul, Theories of Self and Practices of Truth in Indian Ethics and Epistemology (Oxford University Press: Oxford). Ganeri, Jonardon, and Clare Carlisle (eds.). 2010. Philosophy as Therapeia (Cambridge University Press: Cambridge). Gerety, Finnian McKean Moore. 2015. ‘This Whole World Is OM: Song, Soteriology, and the Emergence of the Sacred Syllable’, Harvard University. Gerschheimer, Gerdi. 2000–2001. ‘Systèmes “philosophico-religieux ” (darsana) indiens et doxographies’, École pratique des hautes études, Section des sciences religieuses. Annuaire, Tome 109: 173–189. ———. 2007. ‘Les “Six doctrines de spéculation” (ṣaṭtarkī). Sur la catégorisation variable des systèmes philosophiques dans l’Inde classique’, in Karin Preisendanz (ed.), Expanding and Merging Horizons: Contributions to South Asian and Cross-Cultural Studies in Commemoration of Wilhelm Halbfass (Philosophish-Historische Klasse, Denkschriften 351; Beiträge zur Kultur- und Geistesgeschichte Asiens 53) (Österreichische Akademie der Wissenschaften: Wien). Gethin, Rupert. 1992. ‘The Matikas: Memorization, Mindfulness, and the List’, in Janet Gyatso (ed.), In the Mirror of Memory, Reflections on Mindfulness and Remembrance in Indian and Tibetan Buddhism (State University of New York Press: Albany, NY). Ghosh, Manomohan. 1951. The Nāṭyaśāstra, A Treatise on Hindu Dramaturgy and Histrionics, Ascribed to Bharata-Muni, Vol. I (Chapters I–XXVII), Completely Translated for the First Time from the Original Sanskrit with an Introduction and Various Notes (The Royal Asiatic Society of Bengal: Calcutta). Ghosh, Suresh Chandra. 2002. Civilisation, Education, and School in Ancient and Medieval India, 1550 B.C.-1757 A.D.: Ancient India, 1500 B.C.-1192 A.D., Vedic Schools and Buddhist Viharas and Medieval India, 1192 A.D.-1757 A.D., Maktabs and Madrashas [i.e. Madrasahs], Mosques and Khanqahs (Peter Lang: Frankfurt). Gokhale, V.V. 1958. ‘The Vedānta-Philosophy Described by Bhavya in the Madhyamakahṛdaya’, Indo-Iranian Journal, II: 165–180. ———. 1972. ‘The Second Chapter of Bhavya’s Madhyamakahṛdaya’, Indo-Iranian Journal, XIV: 40–45.

Bibliography  189 Gokhale, V.V. and S.S. Bahulkar. 1985. ‘Madhyamakahṛdayakārikā Tarkajvālā, Chapter I’, in Chr. Lindtner (ed.), Miscellanea Buddhica, 37–62. (Akademisk Forlag: Copenhagen). Gokulbhai, Patel Maheshkumar. 2009. ‘Sarvasiddhāntasaṅgraha – A Study’, Maharaja Sayajirao University of Baroda. Gombrich, Richard. 1971. ‘Merit Transference in Sinhalese Buddhism: A Case Study of the Interaction between Doctrine and Practice’, History of Religions, 11: 203–219. Gracia, Jorge J.E. 1992. Philosophy and Its History, Issues in Philosophical Historiography (State University of New York Press: Albany, NY). Haberman, David L. 2013. People Trees: Worship of Trees in Northern India (Oxford University Press: New York). Hacker, Paul. 1968. ‘Śaṅkara der Yogin und Śaṅkara der Advaitin, Einige Beobachtungen’, Wiener ZeitschriJt für die Kunde Süd-und Ostasiens, 12: 119–148. ———. 1978. ‘Śaṅkārācarya and Śaṅkarabhagavatpāda, Preliminary Remarks Concerning the Authorship Problem’, in Lambert Schmithausen (ed.), Paul Hacker, Kleine Schriften (Franz Steiner Verlag: Wiesbaden). Hadot, Pierre. 1993. Exercices spirituels et philosophie antique (Albin Michel: Paris). ———. 2001. La Philosophie comme manière de vivre, Entretiens avec Jeannie Carlier et Arnold I. Davidson (Albin Michel: Paris). Halbfass, Wilhelm. 1979. ‘Observations on Darśana’, Wiener Zeitschrift für die Kunde Südasiens, 23: 195–203. ———. 1983. Studies in Kumārila and Śaṅkara (Dr. Inge Wezler Verlag für Orientalistische Fachpublikationen: Reinbeck). ———. 1988. India and Europe: An Essay in Understanding (State University of New York Press: New York). ———. 1991. Tradition and Reflection: Explorations in Indian Thought (State University of New York Press: Albany, NY). ———. 1995. Philology and Confrontation, Paul Hacker on Traditional and Modern Vedānta (State University of New York Press: Albany, NY). Ham, Hyoung Seok. 2016. ‘Buddhist Critiques of the Veda and Vedic Sacrifice: A Study of Bhāviveka’s Mīmāṃsā Chapter of the Madhyamakahṛdayakārikā and Tarkajvālā’, University of Michigan. Harimoto, Kengo. 1999. ‘A Critical Edition of the Pātāñjalayogaśāstravivaraṇa, First Pāda, Samādhipāda, with an Introduction’, University of Pennsylvania. ———. 2014. God, Reason and Yoga, A Critical Edition and Translation of the Commentary Ascribed to Śaṅkara on Pātañjalayogaśāstra 1.23–28 (Universität Hamburg – AAIAbt. für Kultur u. Geschichte Indiens und Tibets: Hamburg). Hartmann, Jens-Uwe. 1998. ‘The Ranjana Script, Translated from the German by Prof. Dr. Ursula Heise’, in Andreas Kretschmar (ed.), The Fifth Seal Calligraphic-Icons/Kalligraphikons, 37–39 (Arun K. Saraf: The Radheshyam Saraf Art Collection at Hotel Yak & Yeti, Kathmandu, Nepal). Hatcher, Brian A. 2008. Bourgeois Hinduism, or the Faith of the Modern Vedantists. Rare Discourses from Early Colonial Bengal (Oxford University Press: Oxford). He, Huanhuan. 2011. ‘Bhavya’s Critique of the Vaiśeṣika Theory of Liberation in the Tarkajvālā’, Studies in Indian Philosophy and Buddhism (インド哲学仏教学研究), 18: 23–38. ———. 2014. ‘Further Notes on Bhāviveka’s Principal Oeuvre’, Indo-Iranian Journal, 57: 299–352.

190  Bibliography Heidegger, Martin. 1982. The Basic Problems of Phenomenology, Translation, Introduction and Lexicon by Albert Hofstadter, Studies in Phenomenology and Existential Philosophy (Indiana University Press: Bloomington, IN). Heitmann, Annette L. 1995. ‘Momentane (ekakṣaṇa) Gnosis (jñāna) im Sinne der Madhyamakahṛdayakārikā und der in tibetischer Übersetzung vorliegenden Tarkajvālā I–IV’, Asiatische Studien/Études Asiatiques, 49: 391–427. ———. 2004. Nektarder Erkenntnis. Buddhistische Philosophie des 6. Jh.: Bhavyas Tarkajvālā I– III.26 (Shaker Verlag: Aachen). ———. 2009. Textedition und -kritik von Bhavyas Madhyamakahṛdayakārikā I-III (Verlag Dr. Kovač: Hamburg). Hirabayashi, Jay, and Shotaro Iida. 1978. ‘Another Look at the Mādhyamika vs. Yogācāra Controversy Concerning Existence and Non-Existence’, Edward Conze Festschrift (Berkeley Buddhist Studies Series), 1: 341–360. Hirakawa, Akira. 1990. A History of Indian Buddhism, From Sakyamuni to Early Mahayana, Translated and Edited by Paul Groner (University of Hawaii Press: Honolulu, HI). Hirst, Jacqueline Suthren. 1996. ‘Strategies of Interpretation: Śaṃkara’s Commentary on Bṛhadāraṇyakopaniṣad’, Journal of the American Oriental Society, 116: 58–75. ———. 2004. ‘Images of Śaṃkara: Understanding the Other’, International Journal of Hindu Studies, 8: 157–181. ———. 2005. Śaṃkara’s Advaita Vedānta, A Way of Teaching (RoutledgeCurzon: London and New York). Holquist, Michael. 1990. Dialogism, Bakhtin and His World (Routledge: Oxford). Honda, Megumu. 1967. ‘Sāṃkhya Philosophy Described by His Opponent Bhavya’, IBK, XVI: 33–38. Hoornaert, Paul. 1999. ‘An Annotated Translation of Madhyamakahṛdayakārikā/Tarkajvālā V.1–7’, Kanazawa Daigaku Bungakubu Ronshū Kōdō Kagaku.Tetsugaku hen 金沢大学 文学部論集行動科学・哲学篇沢大学文学部論集行動科学・哲学篇 [Bulletin of the Faculty of Letters of Kanazawa University, Behavioral Sciences and Philosophy], 19: 127–159. ———. 2000. ‘An Annotated Translation of Madhyamakahṛdayakārikā /Tarkajvālā V.8– 26’, KURA, 20: 75–111. ———. 2001. ‘An Annotated Translation of Madhyamakahṛdayakārikā/Tarka-jvālā V.27– 54’, Kanazawa Daigaku Bungakubu Ronshū Kōdō Kagaku.Tetsugaku hen 金沢大学文 学部論集行動科学・哲学篇, 21: 149–190. ———. 2002. ‘An Annotated Translation of Madhyamakahrdayakarika/Tarkajvala V.69– 84’, Kura, 22: 113–137. ———. 2003. ‘An Annotated Translation of Madhyamakahṛdayakārikā/Tarka-jvālā V.85– 114’, Kanazawa Daigaku Bungakubu Ronshū Kōdō Kagaku.Tetsugaku hen 金沢大学文 学部論集行動科学・哲学篇, 23: 139–170. ———. 2004. ‘The Dharmapāla-Bhāvaviveka Debate as Presented in Dharmapāla’s Commentary to Catuḥśataka XVI.23’, Kanazawa Daigaku Bungakubu Ronshū Kōdō Kagaku. Tetsugaku hen 金沢大学文学部論集行動科学・哲学篇, 24: 119–149. Hopkins, Jasper. 1985. Nicholas of Cusa’s Dialectical Mysticism, Text, Translation, and Interpretive Study of De Visione Dei (The Arthur J. Banning Press: Minneapolis, MN). Hopkins, Jeffrey. 1996. ‘The Tibetan Genre of Doxography: Structuring a Worldview’, in José Ignacio Cabezón and Roger R. Jackson (eds.), Tibetan Literature: Studies in Genre (Snow Lion Publications: Ithaca, NY). Hotta, Kazuyoshi. 2009. ‘Fasting Unto Death, Holy Ritual or Suicide?’, in The 3rd BESETO Conference of Philosophy, Philosophy in East Asian Context: Knowledge, Action, Death, and Life, 216–219. The University of Tokyo.

Bibliography  191 Hsu, Chien Yuan. 2011. ‘Bhāvaviveka’s Syllogism as an Initial Step to Enlightenment (正觀第五十九期/二Ο一一年十二月二十五日)’, available online at http://enlight.lib. ntu.edu.tw/FULLTEXT/JR-AN/an381604.pdf, 收稿: 2011/5/30, 通過審查: 2011/8/14 57–92. ———. 2013. ‘Bhāviveka’s Jewel in the Hand Treatise: Elucidating a Path to Awakening Utilizing Formal Inference’, University of Calgary. Hulin, Michel. 1994. Qu’est-ce que l’ignorance métaphysique (dans la pensée hindoue)? Śaṅkara (Librairie Philosophique Vrin: Paris). Iida, Shotaro. 1966. ‘Āgama (Scripture) and Yukti (Reason) in Bhāvaviveka’, Kanakura Kinenronbunshû (Kanakura Festschrift): 77–96. ———. 1980. Reason and Emptiness: A Study in Logic and Mysticism (Hokuseido Press: Tokyo). Inada, Kenneth K. 1993. Nāgārjuna, A Translation of His Mūlamadhyamakārikā with an Introductory Essay (Sri Satguru Publications: Delhi). Ingalls, Daniel H.H. 2001. ‘Logic in India (1955)’, in Jonardon Ganeri (ed.), Indian Logic, A Reader (Routledge: New York). Isayeva, Natalia. 1993. Shankara and Indian Philosophy (State University of New York Press: Albany, NY). Jacobi, Hermann. 1921. ‘“Die Bhagavadgītā”, Review of Die Bhagavadgītā, by Richard Garbe’, Deutsche Literaturzeitung, 42: 715–724. ——— (ed.). 1926. Samarāicca Kahā. A Jaina Prākṛta Work. Volume I: Text and Introduction (Asiatic Society of Bengal, Bibliotheca Indica, 169: Calcutta). Jaffrelot, Christophe. 1993. ‘Hindu Nationalism: Strategic Syncretism in Ideology Building’, Economic and Political Weekly, 28: 517–524. ——— (ed.). 2007. Hindu Nationalism. A Reader (Princeton University Press: Princeton and Oxford). Jain, Nemi C., and Anuradha Matukumalli. 2014. ‘The Functions of Silence in India: Implications for Intercultural Communication Research’, in M.K. Asante, Y. Miike and J. Yin (eds.), The Global Intercultural Communication Reader, Second edition (Routledge: New York and London). Jaini, Padmanabh S. 1979. The Jaina Path of Purification (University of California Press: Berkeley). ——— 2000. Collected Papers on Jaina Studies (Motilal Banarsidass: Delhi). Jakubczak, Marzenna. 2017. ‘The Woman-and-Tree Motif in the Ancient and Contemporary India’, in Zoltán Somhegyi (ed.), Retracing The Past, Historical Continuity in Aesthetics from a Global Perspective (International Association for Aesthetics: Santa Cruz). Johnson, William J. 1999. ‘Kundakunda: Two-Standpoints and the Socio-Religious Functions of AnekāntaVāda’, in N.K. Wagle and Olle Qvarnström (eds.), Approach to Jaina Studies: Philosophy, Logic, Rituals and Symbols (University of Toronto, Center for South Asian Studies: Toronto). Jones, Richard H. 2011. Indian Madhyamaka Buddhist Philosophy After Nagarjuna, Volume I, Plain English Translations and Summaries of the Essential Works of Aryadeva, Rahulabhadra, Buddhapalita, and Bhaviveka (Jackson Square Books: New York). Kanchi, Vijay Srinath. 2016. ‘Was the Original Lokāyata a “Live-in-the-Present” Spiritualistic School?: An Alternative Perspective’, Journal of Indian Council of Philosophical Research, 33: 385–405. Kant, Immanuel. 1987. Critique of Judgment: Including the First Introduction, Translated and Introduction by Werner S. Pluhar, with a Foreword by Mary J. Gregor (Hackett: Indianapolis, IN).

192  Bibliography Kapadia, Hiralal R. 1933–34. ‘Historico-Religious Classifications of Mankind as Embodied in the Jaina Canon’, Annals of the Bhandarkar Oriental Research Institute, 15: 97–108. ———. 1947. Haribhadra’s Anekāntajayapatākā (Oriental Institute: Baroda). Kaplan, Stephen. 1992. ‘The Yogācāra Roots of Advaita Idealism? Noting a Similarity Between Vasubhandu and Gauḍapāda’, Journal of Indian Philosophy, 20: 191–218. Kapstein, Matthew T. 2003. Reason’s Traces. Identity and Interpretation in Indian and Tibetan Buddhist Thought (Oxford University Press: Delhi). ———. 2013a. ‘“Spiritual Exercise” and Buddhist Epistemologists in India and Tibet’, in Steven M. Emmanuel (ed.), A Companion to Buddhist Philosophy (John Wiley & Sons: Chichester, UK). ———. 2013b. ‘Stoics and Bodhisattvas: Spiritual Exercises and Faith in Two Philosophical Traditions’, in Michael Chase, Stephen R.L. Clark and Michael McGhee (eds.), Philosophy as a Way of Life: Ancients and Moderns – Essays in Honor of Pierre Hadot (Wiley-Blackwell: Malden, MA, and Oxford, UK). Kaushik, Garima. 2016. Women and Monastic Buddhism in Early South Asia: Rediscovering the Invisible Believers (Routledge: New York). Kawasaki, Shinjo. 1973. ‘Bhavyano tsutaeru Mīmāṃsā shisô (The Mīmāṃsā thought transmitted by Bhavya)’, in Indo Shisô to Bukkyô (Nakamura Festschrift) (Shunjûsha: Tokyo). ———. 1974. ‘Quotations in the Mīmāṃsā Chapter of Bhavya’s Madhyamaka-hṛdayakārikā’, Journal of the Japanese Association of Indian and Buddhist Studies, 22: 1127–1120. ———. 1976. ‘The Mīmāṃsā Chapter of Bhavya’s MHK – Text and Translation (1) Pūrvapakṣa’, Studies 1976, Institute of Philosophy, The University of Tsukuba: 1–15. ———. 1985. ‘肉食と Bhāvaviveka [Flesh-eating and Bhāvaviveka]’, 東方, 1: 174–184 (L). ———. 1992a. ‘Discrepancies in the Sanskrit and Tibetan Texts of Bhavya’s Madhyamakahṛdaya-Tarkajvālā (the 9th and 10th Chapters)’, in Ihara Shōren and Yamaguchi Zuihō (eds.), Buddhist Philosophy and Literature: Tibetan Studies: Proceedings of the 5th Seminar of the International Association for Tibetan Studies, Narita 1989 (Naritasan Shinshoji: Narita). ———. 1992b. 切智思想の研究 [A Study of the Idea of Omniscience] (Shunjūsha: Tokyo). Kierkegaard, Søren. 1855. ‘On the Concept of Irony with Continual Reference to Socrates’, in The Concept of Irony with Continual Reference to Socrates, Kierkegaard’s Writings, II, Together with Notes of Schelling’s Berlin Lectures (1989), edited by Howard V. Hong and Edna H. Hong (Princeton University Press: Princeton). King, Karen L. 2003. What Is Gnosticism? (Belknap Press of Harvard University Press: Cambridge, MA). King, Richard. 1995. Early Advaita Vedānta and Buddhism, The Mahāyāna Context of the Gauḍapādīya-kārikā (State University of New York Press: Albany, NY). ———. 1999. Indian Philosophy: An Introduction to Hindu and Buddhist Thought (Edinburgh University Press: Edinburgh). Klostermaier, Klaus K. 2007. A Survey of Hinduism (State University of New York Press: Albany, NY). Kramrisch, Stella. 1988. The Presence of Siva (Motilal Banarsidass: Delhi). Krasser, Helmut. 2011. ‘How to Teach a Buddhist Monk to Refute the Outsiders – Text Critical Remarks on Some Works by Bhāviveka’, Dhīḥ, 51: 49–76. ———. 2012. ‘Bhāviveka, Dharmakīrti and Kumārila’, in F. Voegeli et al. (eds.), Devadattīyam. Johannes Bronkhorst Felicitation Volume (Peter Lang: Bern).

Bibliography  193 Laks, André 1999. ‘Histoire critique et doxographie, Pour une histoire de l’historiographie de la philosophie’, Presses Universitaires de France, Les Études philosophiques, No. 4, La philosophie et ses histoires: 465–477. ———. 2007. Histoire, doxographie, vérité, Études sur Aristote, Théophraste et la philosophie présocratique (Éditions Peeters: Louvain-la-Neuve). Larson, Gerald James. 1969. Classical Sāṃkhya: An Interpretation of its History and Meaning (Motilal Banarsidass: New Delhi). La Vallée Poussin, Louis de. Tr. 1903. ‘Nanjio’s 1185: Bhāvaviveka’, The Journal of the Royal Asiatic Society of Great Britain and Ireland, July: 581–583. ———. 1903–1913. Mūlamadhyamakakarikas de Nagarjuna avec la Prasannapadā Commentaire de Candrakīrti (l’Académie Impériale des Sciences: St. Petersburg). ———. 1932–1933. ‘Madhyamaka, L’auteur du Joyau dans la main; 3, traduction du Joyau dans la main’, MCR, 2: 60–138. ———. 1933. ‘Madhyamaka’, Mélanges Chinois et Bouddhiques, II. Lehtonen, Tommi. 2000. ‘The Notion of Merit in Indian Religions’, Asian Philosophy, 10: 189–204. Liebich, Bruno. 1923. ‘Bemerkungen zu Śaṅkara’s Sarva-Siddhānta-Saṅgraha’, Zeitschrift für Indologie und Iranistik, 2: 124–132. Lindtner, Christian. 1982. ‘Adversaria Buddhica’, Wiener Zeitschrift für die Kunde Südasiens, XXVI: 167–194. ———. 1986. ‘Bhavya, the Logician’, Adyar Library Bulletin, Golden Jubilee Volume: 58–84. ———. 1995. ‘Bhavya’s Madhyamakahṛdaya (Pariccheda Five) Yogācāratattvaviniścay āvatāra’, Adyar Library Bulletin, 59: 37–65. ———. 2001a. Bhavya on Mīmāṃsā, Mīmāṃsātattvanirṇayāvatāraḥ, With English Translation by Christian Lindtner (The Adyar Library and Research Centre: Adyar). ———. 2001b. Madhyamakahrdayam of Bhavya (The Adyar Library and Research Center: Adyar). Lorenzen, David N. (ed.). 1983. The Life of Śaṅkarācārya (Manohar: New Delhi). Maas, Philipp A. 2013. ‘A Concise Historiography of Classical Yoga Philosophy’, in Eli Franco (ed.), Periodization and Historiography of Indian Philosophy (Sammlung De Nobili: Vienna). Mādhavānanda, Swāmī. 1950. The Bṛhadāraṇyaka Upaniṣad, with the commentary of Śankarācārya, with an Introduction by Mahāmahopādhyāya Prof. S. Kuppuswāmī Śāstrī, M.A., I.E.S. Vidyā-vācaspati, Darśana-kalānidhi, Kulapati (Advaita Ashrama: Almora). Mansfeld, Jaap. 1999a. ‘Sources’, in A.A. Long (ed.), The Cambridge Companion to Early Greek Philosophy, 22–44 (Cambridge University Press: Cambridge). ———. 1999b. ‘Sources’, in A.A. Long (ed.), The Cambridge Companion to Early Greek Philosophy (Cambridge University Press: Cambridge). ———. 2002. ‘Deconstructing Doxography’, Philologus, 146: 277–286. ———. 2016. ‘Doxography of Ancient Philosophy’, in Edward N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Metaphysics Research Lab, Stanford University: Stanford). Mansfeld, Jaap, and D.T. Runia. 1997. Aëtiana. 1. The Sources (E.J. Brill: Oxford). Matilal, Bimal Krishna. 1977. Nyāya-Vaiśeṣika (Otto Harrassowitz Verlag: Wiesbaden). ———. 1981. The Central Philosophy of Jainism (Anekānta-vāda) (L.D. Institute of Indology: Ahmedabad).

194  Bibliography ———. 1990. Logic, Language and Reality: Indian Philosophy and Contemporary Issues (Motilal Banarsidass: New Delhi). Matilal, Bimal Krishna, J. Ganeri, and H. Tiwari. 1998. The Character of Logic in India (State University of New York Press: New York). Mayeda, Sengaku. 1965a. ‘The Authenticity of the Bhagavadgītābhāṣya Ascribed to Śaṅkara’, Wiener Zeitschrift für die Kunde Süd-und Östasiens, IX: 155–194. ———. 1965b. ‘The Authenticity of the Upadeśasāhasrī Ascribed to Saṅkara’, Journal of the American Oriental Society, 85(2): 178–196. McClintock, Sara L. 2002. ‘Omniscience and the Rhetoric of Reason in the Tattvasaṅgraha and the Tattvasaṅgrahapañjikā’, Harvard University, Cambridge, MA. ———. 2010. Omniscience and the Rhetoric of Reason: Śantarakṣita and Kamalaśīla on Rationality, Argumentation, and Religious Authority (Wisdom Publications: Boston, MA). ———. 2018. ‘Schools, Schools, Schools-Or, Must a Philosopher be Like a Fish?’, in David V. Fiordalis (ed.), Buddhist Spiritual Practices. Thinking with Pierre Hadot on Buddhism, Philosophy, and the Path (Mangalam Press: Berkeley, CA). McLuhan, Marshall. 1964. Understanding Media: The Extensions of Man (McGraw-Hill: New York). Menezes, Walter. 2017. Exploring Ātman from the Perspective of the Vivekacūḍāmaṇi (Palgrave Macmillan: Cham). Mimaki, Katsumi. 1982. Blo Gsal Grub Mtha’: Chapitres IX (Vaibhasika) et XI (Yogacara) édités et Chapitre XII (Madyamika) édité et traduit (Zinbun Kagaku Kenkyusyo – ­Université de Kyoto: Kyoto). Mims, Cedric. 2014. When We Die: The Science, Culture, and Rituals of Death (St. Martin’s Press: New York). Miyasaka, Yûshô. 1954a. ‘Bhavya to Vaiśeṣika gatusetsu (Bhavya and the Vaiśeṣika doctrines)’, Shûkyô Kenkyû, 133: 173–177. ———. 1954b. ‘Shôben inyô no Vaiśeṣika tetsugakusetsu (The Vaiśeṣika Philosophy Quoted by Bhavya)’, Bunka, 18: 24–40. ———. 1958. ‘Ronri no honoho ni okeru Vaiśeṣika Tesugaku (Vaiśeṣika Philosophy in the Tarkajvālā)’, Kôyasan Daigaku Ronsô, 1: 1–37. Mohan, A., and K. Sivaramalingam. 2014. ‘Agnipraveṣa in the Rāmāyaṇa at Vaduvur’, Asia Pacific Journal of Research, 1: 171–177. Motegi, Shujun. 2010. ‘Early Concept of Logic in Sāṃkhya’, in Piotr Balcerowicz (ed.), Logic and Belief in Indian Philosophy (Motilal Banarsidas: Delhi). Muller, Albert Charles. 2011. ‘An Inquiry into Views: Lessons from Buddhism, Behavioral Psychology, and Constructivist Epistemology’, in Center for International Affairs (ed.), 2010 Civilization and Peace Beyond National Boundaries: Building a World Without Walls (Academy of Korean Studies Press: Korea). Müller, Friedrich Max. 1879. The Upanishads, Part 2 (Clarendon Press: Oxford). ———. 1899. The Six Systems of Indian Philosophy (Longmans, Green and Co.: New York, London and Bombay). Mun, Chanju. 2006. The History of Doctrinal Classification in Chinese Buddhism: A Study of the Panjiao Systems (University Press of America, Incorporated: Lanham, MD). Murty, K. Satchidananda. 1957. Ṣaḍ-darśana-samuccaya, A Compendium of Six Philosophies, by Haribhadra (Eastern Book Linkers: Delhi). ———. 1974. Revelation and Reason in Advaita Vedānta (Motilal Banarsidass: Delhi). Nagao, Gajin M. 1991. ‘The Silence of the Buddha and Its Madhyamic Interpretation [Japanese, Chūkan to Yuishiki, 1991],’ in L.S. Kawamura with the collaboration of G.M.

Bibliography  195 Nagao (eds.), Mādhyamika and Yogācāra: A Study of Mahāyāna Philosophy: Collected Papers of G M Nagao (State University of New York: Albany, NY). Nakada, Naomichi. 1972. ‘The Sanskrit Text of the MHK and the Tibetan Text of the MHKvṛtti Tarkajvālā Ṣaṣṭaḥ-paricchedaḥ/Sāṃkhya-tattvāvatāraḥ // (Part 1. Pūrvapakṣa)’, Tsurumi Jyoshidaigakubu Kiyô, 6: 145–146. Nakamura, Hajime. 1950. ‘Shoki no Vedānta Tetsu-gaku (The Early Vedānta Philosophy)’, Tôkyô: 238–332. ———. 1955. ‘Upaniṣadic Tradition and the Early School of Vedānta, as Noticed in Buddhist Scripture’, HJAS, 18: 74–104. ———. 1958. ‘The Tibetan Text of the Madhyamaka-hṛdaya-vṛtti-tarkajvālā’, IIJ, 2: 181–190. ———. 1968. ‘Shoki Vedānta-tetsugaku kenkyû no hatten (The Progress Report on the Study of the Early Vedānta)’, IBK, XVI: 10–22. Naugle, David K. 2002. Worldview: The History of a Concept (W.B. Eerdmans Publishing Company: Grand Rapids, MI). Neelis, Jason. 2010. Early Buddhist Transmission and Trade Networks: Mobility and Exchange Within and Beyond the Northwestern Borderlands of South Asia (Brill: Leiden). Nehru, Jawaharlal. 1946. The Discovery of India (The Signet Press: Calcutta). Nicholson, Andrew J. 2010. Unifying Hinduism: Philosophy and Identity in Indian Intellectual History (Columbia University Press: New York). ———. 2012. ‘Doxography and Boundary-Formation in Late Medieval India’, in Piotr Balcerowicz (ed.), World View and Theory in Indian Philosophy Warsaw Indological Studies (Manohar: New Delhi). ———. 2015. ‘Dialogue and Genre in Indian Philosophy’, in Brian Black and Laurie Patton (eds.), Dialogue in Early South Asian Religions: Hindu, Buddhist, an Jain Traditions (Ashgate: Farnham, Burlington). Nicholson, Hugh. 2007. ‘Two Apologetic Moments in Śaṅkara’s Concept of Brahman’, The Journal of Religion, 87: 528–555. Olivelle, Patrick. 1998. The Early Upaniṣads, Annotated Text and Translation (Oxford University Press: New York). ———. 2007. ‘The Date and Provenance of the Viṣṇu Smṛti’, Indologica Taurinensia, 33: 149–163. Organ, Troy Wilson. 1954. ‘The Silence of the Buddha’, Philosophy East and West, 4: 125–140. Pahlajrai, Prem. 2004. ‘Doxographies – Why Six darśanas? Which Six?’, in Asian Studies Graduate Student Colloquium, 1–14 (University of Washington: Seattle). http://faculty. washington.edu/prem/Colloquium04-Doxographies.pdf (accessed January 12, 2018). Pande, Govind Chandra. 1994. Life and Thought of Śaṅkarācārya (Motilal Banarsidass Publishers: Delhi). Park, Kwangsoo. 2006. ‘An Analysis of the Buddha’s Paradoxical Silence: Neither the Positive nor Nihilistic View’, International Journal of Buddhist Thought & Culture, February, 6: 243–264. Parpola, Asko. 1981. ‘On the Formation of the Mīmāṃsā and the Problems Concerning Jaimini, with Particular Reference to the Teacher Quotations and the Vedic Schools’, Wiener Zeitschrift für die Kunde Südasiens, 25: 145–177. ———. 1994. ‘On the Formation of the Mīmāṃsā and the Problems Concerning Jaimini, with Particular Reference to the Teacher Quotations and the Vedic Schools (Part II)’, Wiener Zeitschrift für die Kunde Südasiens, 38: 293–308.

196  Bibliography Paul, Diana Y., and Frances Wilson. 1979. Women in Buddhism: Images of the Feminine in the Mahayana Tradition (Asian Humanities Press: Berkeley, CA). Petit, Jérôme. 2014. ‘Absolute and Conventional Points of View in Jainism: A Historical Perspective’, in J. Soni, M. Pahlke and C. Cüppers (eds.), Buddhist and Jaina Studies, Proceedings of the Conference in Lumbini, February 2013 (Lumbini International Research Institute: Lumbini). Petzold, Bruno, S. Hanayama, and S. Ichimura. 1995. The Classification of Buddhism (Otto Harrassowitz Verlag: Wiesbaden). Potter, Karl H. (ed.). 2003. Buddhist Philosophy from 350 to 600 A.D. (Motilal Banarsidas Publishers: Delhi). ——— (ed.). 2014. Encyclopedia of Indian Philosophies. Philosophy of Pūrva-Mīmāṃsā (Motilal Banarsidass: Delhi). Potter, Karl H., Edeltraud Harzer, S. Subrahmanya Sastri, and Allen W. Thrasher. 1981. ‘Advaita Vedānta up to Śaṃkara and His Pupils’, in Karl H. Potter (ed.), Encyclopedia of Indian Philosophies (Motilal Banarsidass: Delhi). Poussin, Louis de La Vallée. 1902. ‘Sarvasiddhāntasaṁgraha’, Le Muséon, III: 402–412. Prebish, Charles S. 1996. Buddhist Monastic Discipline, The Sanskrit Prātimokṣa Sūtras of the Mahāsāṃghikas and Mūlasarvāstivādins (Motilal Banarsidass: Delhi). Qvarnström, Olle. 1988. ‘Space and Substance. A Theme in Madhyamaka-Vedānta Polemics’, Studies in Central and East Asian Religions, 1: 3–34. ———. 1989. Hindu Philosophy in Buddhist Perspective. The Vedāntatattvaviniścaya Chapter of Bhavya’s Madhyamakahṛdayakārikā (PlusUltra: Lund). ———. 1999. ‘Haribhadra and the Beginnings of Doxography in India’, in N.K. Wagle and Olle Qvarnström (eds.), Approaches to Jaina Studies: Philosophy, Logic, Rituals and Symbols (Center for South Asian Studies: Toronto). ———. 2012. ‘Sāṃkhya as Portrayed by Bhāviveka and Haribhadrasūri’, Journal of Indian Philosophy, 40: 395–409. ———. 2015. Bhāviveka on Sāṃkhya and Vedānta, The Sāṃkhya and Vedānta Chapters of the Madhyamakahṛdayakārikā and Tarkajvālā (Harvard University Press: Cambridge, MA). Radhakrishnan, Sarvepalli. 1914. ‘The Vedanta Philosophy and the Doctrine of Maya’, International Journal of Ethics, 24: 431–451. ———. 1929. Indian Philosophy, Volume 2 (HarperCollins Publishers: London). Rajan, Rajeswari Sunder. 2011. ‘The Politics of Hindu “Tolerance”’, Boundary 2, 38: 67–86. Raju, Poola Tirupati. 1985. Structural Depths of Indian Thought (State University of New York Press: Albany, NY). Ram, P.R. 1997. ‘Left Ideology, Ends and Means and Hindutva’, Economic and Political Weekly, 32: 1426–1428. Rambachan, Anantanand. 1991. Accomplishing the Accomplished: The Vedas as Source of Valid Knowledge in Śaṅkara (University of Hawaii Press: Hawaii). Ram-Prasad, Chakravarthi. 2000. ‘Knowledge and Action I: Means to the Human End in Bhāṭṭa Mīmāṃsā and Advaita Vedānta’, Journal of Indian Philosophy, 28: 1–24. ———. 2001. Knowledge and Liberation in Classical Indian Thought (Palgrave: Hampshire). ———. 2002. Advaita Epistemology and Metaphysics: An Outline of Indian Non-Realism (Routledge: London and New York). ———. 2007. Indian Philosophy and the Consequences of Knowledge: Themes in Ethics, Metaphysics and Soteriology (Ashgate: Hampshire).

Bibliography  197 Raṅgācārya, M. 2006. The Sarva Siddhānta Saṅgraha of Śaṅkarācārya: Text with English Translation, Introduction, Notes, Glossary & Index (Eastern Books Linkers: Delhi). Raskin, Victor. 1984. Semantic Mechanisms of Humor (D. Reidel Publishing Company: Dordrecht). Reddy, Deepa S. 2011. ‘Hindutva: Formative Assertions’, Religion Compass, 5: 439–451. Renfrew, Alastair. 2015. Mikhail Bakhtin (Routledge: London and New York). Rhi, Juhyung. 2006 (2008). ‘Some Textual Parallels for Gandhāran Art: Fasting Buddhas, Lalitavistara, and Karuṇāpuṇḍarīka’, Journal of the International Association of Buddhist Studies, 29: 125–153. Ronkin, Noa. 2005. Early Buddhist Metaphysics, The Making of a Philosophical Tradition (RoutledgeCurzon: London and New York). Ruegg, David Seyfort. 1981. The Literature of the Madhyamaka School of Philosophy in India (Otto Harrassowitz: Wiesbaden). ———. 1990. ‘On the Authorship of Some Works Ascribed to Bhāvaviveka/Bhavya’, in Seyfort Ruegg and L. Schmithausen (eds.), Earliest Buddhism and Madhyamaka (E.J. Brill: Leiden). ———. 2004. ‘Aspects of the Study of the (Earlier) Indian Mahāyāna’, Journal of the International Association of Buddhist Studies, 27: 3–62. Rukmani, T.S. 1993. ‘Śaṅkara’s Views on Yoga in the Brahmasūtrabhāṣya in the Light of the Authorship of the Yogasūtrabhāṣya-Vivaraṇa’, Journal of Indian Philosophy, 21: 395–404. Runia, David Theunis. 1997. ‘Lucretius and Doxography’, in K.A. Algra et al. (eds.), Lucretius and His Intellectual Background (Verh. KNAW Afd. Lett. N.R. 172: Amsterdam etc.). Russell, Bertrand. 2014. Mysticism and Logic, and Other Essays (1919 Compilation) (Peruse Press: Los Angeles). Saitō, Akira 斎藤明. 2005. ‘Chūgan shinron’ no shōmei to sono seiritsu o meguru shōmondai 『中観心論』 の書名とその成立をめぐる諸問題 [Some Problems with the Title of the ‘Madhyamakahṛdayaśāstra’ and Its Formation]’, Indogaku Bukkyōgaku Kenkyū 印度學佛教學研究, LIII: 832–838. ———. 2011. ‘Bhavya’s Critique of the Sāṃkhya Theory of pratibimba’, 哲学仏教学 研究, 18: 13–23. Sarao, K.T.S. 2012. The Decline of Buddhism in India: A Fresh Perspective (Munshirm Manoharlal Publishers: New Delhi). Saraswati, Sri Swami Satchidanandendra, and A.J. Alston. 1997. The Method of the Vedanta: A Critical Account of the Advaita Tradition (Motilal Banarsidass: Delhi). Schaffer, Simon. 1998. ‘The Leviathan of Parsonstown: Literary Technology and Scientific Representation’, in Timothy Lenoir (ed.), Inscribing Science: Scientific Texts and the Materiality of Communication, Writing Science (Stanford University Press: Stanford). Schmithausen, Lambert. 1991. The Problem of the Sentience of Plants in Early Buddhism (The International Institute for Buddhist Studies: Tokyo). ———. 2009. Plants in Early Buddhism and the Far Easter Idea of the Buddha-Nature of Grasses and Trees (Lumbini International Research Institute: Lumbini). Schubring, W. 2000. The Doctrine of the Jainas Described After the Old Sources (LSJRS 15: Delhi). Sen, Amartya. 2006. The Argumentative Indian: Writings on Indian History, Culture and Identity (Penguin Adult: London). Shapin, Steven, and Simon Schaffer. 1985. Leviathan and the Air-Pump: Hobbes, Boyle, and the Experimental Life. Including a Translation of Thomas Hobbes Dialogus Physicus De Natura Aeris (Princeton University Press: Princeton).

198  Bibliography Sharma, Arvind. 2002. ‘On Hindu, Hindustān, Hinduism and Hindutva’, Numen, 49: 1–36. Sharma, Jyotirmaya. 2013. A Restatement of Religion. Swami Vivekananda and the Making of Hindu Nationalism (Yale University Press: New Haven and London). Shukla, Ram Sajiwan. 1989. India as Known to Haribhadra Sūri (Kusumanjali Prakashan: Meerut). Singh, Upinder. 2017. Political Violence in Ancient India (Harvard University Press: Cambridge, MA). Sopa, Geshe. 1984. ‘The Tibetan “Wheel of Life”: Iconography and Doxography’, Journal of the International Association of Buddhist Studies, 7: 125–145. Stcherbatsky, Fedor Ippolitovich. 1970. Buddhist Logic (Biblio-Verlag: Osnabrück). Steinkellner, Ernst. 1980. ‘Some Sanskrit-Fragments of Jinandrabuddhi’s Viśālāmalavatī’, in Ernst Steinkellner and A.L. Basham (eds.), A Corpus of Indian Studies: Essays in Honour of Professor Gaurinath Sastri (Sanskrit Pustak Bhandar: Calcutta). Stewart, James John. 2015. Vegetarianism and Animal Ethics in Contemporary Buddhism (Taylor & Francis: Oxford). Stuart, Daniel. 2015. A Less Traveled Path: Saddharmasmṛtyupasthānansūtra Chapter 2. Critically Edited with A Study on Its Structure and Significance for the Development of Buddhist Meditation. Volume I and Volume II (VÖAW: Vienna). Śukla, Śrī Sūryanārāyaṇa (ed.). approx. 1930 (undated print). Sarvasiddhāntasaṃgraha, Śrīmacchaṅkarācāryaviracitaḥ (Kāśī Viśvanātha Press: Kāśī). Sypher, Wylie (ed.). 1956. Comedy (Doubleday: Garden City, NY). Tatia, Nathmal. 1951. Studies in Jaina Philosophy (P. V. Research Institute, Jainashram, Hindu University: Varanasi). Thapar, Romila. 1975. ‘Ethics, Religion, and Social Protest in the First Millennium B.C. in Northern India’, Daedalus, 104: 119–132. ———. 1989. ‘Epic and History: Tradition, Dissent and Politics in India’, Past & Present: 3–26. ———. 2010. Ancient Indian Social History, Some Interpretations (Orient BlackSwan Private Limited: Chennai). Thomas, F.W. 1960. The Flower-Spray of the Quodammodo Doctrin, Śrī Malliṣeṇasūri, Syād-Vāda-Mañjarī (Akademie-Verlag: Berlin). Triplett, Katja. 2017. ‘Using the Golden Needle: Nāgārjuna Bodhisattva Ophthalmological Treatise and Other Sources in the Essentials of Medical Treatment’, in C. Pierce Salguero (ed.), Buddhism and Medicine, An Anthology of Premodern Sources (Columbia University Press: New York). Tsomo, Karma Lekshe (ed.). 2000. Innovative Buddhist Women: Swimming Against the Stream (Curzon Press: Richmond). Udayakumar, S.P. 2005. “Presenting” the Past: Anxious History and Ancient Future in Hindutva India (Praeger: Westport, CT). Uncategorised. 1993. ‘Politics – Hindutva’s Other Victims’, Economic and Political Weekly, 28: 192. Vatsyayan, Kapila. 1996 (reprint). Bharata, the Nāṭyaśāstra (Sahitya Akademi: New Delhi). Veer, Peter van der. 2002. ‘Religion in South Asia’, Annual Review of Anthropology, 31: 173–187. Verardi, Giovanni. 2011. Hardships and Downfall of Buddhism in India (Institute of Southeast Asian Studies: New Delhi). Verpoorten, Jean-Marie. 2014. ‘5. Jaimini (25 A.D.?)’, in Karl H. Potter (ed.), Encyclopedia of Indian Philosophies: Vol. XVI: Philosophy of Pūrva-Mīmāṃsā, 58. New Delhi: Motilal Banarsidass.

Bibliography  199 Vidhyabhusana, Satis Chandra. 1913. The Nyāya Sūtras of Gotama (Munshiram Manoharlal Publishers: New Delhi). ———. 1920. A History of Indian Logic: Ancient, Mediaeval, and Modern Schools (Motilal Banarsidass: Delhi). Virupakshananda, Swami. 1995. Sāṃkhya Kārikā of Īśvara Kṛṣṇa with the Tattva Kaumudī of Śrī Vācaspati Miśra (Sri Ramakrishna Math: Madras). Warder, Anthony Kennedy. 1970a. ‘The Description of Indian Philosophy’, Journal of Indian Philosophy, 1: 4–12. ———. 1970b. Indian Buddhism (Motilal Banarsidas: Delhi). Watanabe, Chikafumi. 1994. ‘Bhā-viveka (A.D. c. 490–570)’s Madhyamaka-hṛdayakārikā, Tattvajñānaiṣanā, verses 137–266, An English Translation and Explanation’, University of British Columbia. ———. 1998. ‘A Translation of the Madhyamakahrdayakarikd with the Tarkajvala III. 137–146J’, Journal of the International Association of Buddhist Studies, 21: 125–155. ———. 2012. ‘Madhyamakahṛdayakārikā III. 147–158’, in Chikafumi Watanabe, Michele Desmarais and Yoshichika Honda (eds.), Saṁskṛta-sādhutā: Goodness of Sanskrit. Studies in Honour of Professor Ashok N. Aklujkar (D. K. Printworld: New Delhi). Westerhoff, Jan. 2014. ‘Madhyamaka: Conventional Categories in Madhyamaka Philosophy’, in Jessica Frazier (ed.), Categorisation in Indian Philosophy. Thinking Inside the Box. Dialogues in South Asian Traditions: Religion, Philosophy, Literature and History, 115–131 (Ashgate: Surrey and Burlington). Wezler, A. 1984. ‘On the Quadruple Division of the Yogaśāstra, the Caturvyūhatva of the Cikitsāśāstra and the “Four Noble Truths” of the Buddha (Studies in the Pātañjalayogaśāstravivaraṇa II)’, Indologica Taurinensia, 12: 289–337. Whaling, Frank. 1979. ‘Śaṅkara and Buddhism’, Journal of Indian Philosophy, 7: 1–42. White, David Gordon. 1996. The Alchemical Body: Siddha Traditions in Medieval India (University of Chicago Press: Chicago). Winternitz, Moriz. 1920. Geschichte der indischen litteratur (C. F. Ameland Verlag: Leipzig). Wolters, Albert M. 1983. On The Idea of Worldview and Its Relation to Philosophy (Stained Glass University Press of America: Lanham). Yamaguchi, Susumu. 1941 & 1964. Bukkhyô ni okeru Mu to U no Tairon. (Buddhist Controversies Concerning Existence and Non-Existence) (Sankibô: Tokyo). Yamashita, Jon M. 1998. ‘A Translation and Study of the Pāṇinidarśana Chapter of the Sarvadarśanasaṅgraha’, University of Pennsylvania. Zhmud, Leonid. 2001. ‘Revising Doxography: Hermann Diels and His Critics’, Philologus, 145: 219–243. Zin, Monika. 2014. ‘Imagery of Hell in South, South East and Central Asia’, Rocznik Orientalistyczny, LXVII: 269–296. Zydenbos, Robert J. 1983. Mokṣa in Jainism, According to Umāsvāti (Franz Steiner Verlag: Wiesbaden).

Index

Note: only the authors mentioned in the main body of the text are listed. abhāva 126 Abhayadeva Sūri 93 Abhidānacintāmani 111 Abhidharma 26, 27, 26, 38, 39, 40, 82n71, 83n81, 98, 114, 179, 181, 183; Abhidharmic 39, 40; Ābhidharmika 26 adhikārin 147, 151, 171n41 adhyāropa 15, 134, 136, 145, 151, 154, 155, 163 Advaita 13, 14, 15, 17, 40, 41, 98, 104, 114, 115, 123, 133, 134, 135, 137, 138, 139, 140, 141, 142, 145, 148, 149, 151, 152, 153, 154, 155, 158, 159, 160, 161, 162, 163, 164, 163, 164, 165, 167, 168, 170n20, 176n113, 177n122, 178, 181 āgama 17, 36, 37, 38, 55, 57, 70, 81n56, 93, 110, 111, 121, 123, 124, 143, 144, 152, 153, 161 ahiṃsā 121 ahistorical 2, 9 Ājīvika 108, 129n75; Ājīvikism 108, 129n75 Amarakośa 110 amṛta 17, 37, 70, 109, 180 Ānandagiri 135 anapatrapa 41, 49, 50, 148 anapatrapatva 51 anekaṃsika 95 anekānta 15, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 113, 116, 117, 127n29, 130n90, 162, 180 Anekāntajayapatākā 95 Anekāntasiddhi 95 Anekāntavādapraveśa 95 Angot, Michel 87n134, 161, 173n70, 176n108 anubhava 145, 151, 154, 155 anumāna 17, 27, 31, 36, 37, 49, 55, 57, 59, 68, 70, 71, 109, 110, 123, 126, 142, 143

Anuradha Matukumalli 120 ānvīkṣikī 38, 106, 110, 122 aparāvidyā 17, 161 Aparokṣānubhūti 156 Āpastambaparibhāṣāsūtra 143 apauruṣeya 71, 144, 146; apauruṣeyatva 56 apavāda 15, 105, 134, 136, 145, 154 apologetic(s) 11, 19, 31, 32, 40, 72, 75, 101, 102, 128n47, 134, 135, 146, 163, 164 āpta 17, 121, 123, 124, 152, 162; āptatva 63, 121 apūrva 36, 37, 47, 85n111 Āraṇyaka 62 Aristotle 3, 4, 6, 22n3, 23n56, 24n56 arthāpatti 126 Arthasaṅgraha (AS) 144 Arthaśāstra (AŚ) 110 arthavāda 144 Ārya 108, 136 Āryadeva 14, 29 Āryāvarta 134, 169n3 āsana 61, 151, 158 asceticism 24n57, 67, 69, 102, 108, 120, 128n45, 129n67, 140 āstika 112, 126 asura 37, 139 Aśvaghosa 39 aśva-medha 136 Ātmaboddha 156 ātman 16, 44, 46, 64, 118, 125, 134, 136, 145, 150, 151, 153, 154, 161, 163, 171n41, 182 ātma-vidyā 106, 110 austerity 65, 101, 102, 103, 106, 128n44 avatāra 50, 60, 61, 84n106, 90n167, 109, 137, 170n19, 179 avidyā 16, 145, 151, 154, 163 avīta 125

Index  201 Bakhtin, Mikhail 21 Balcerowicz, Piotr ix, 98, 99, 103, 128n36, 128n37, 128n39, 128n40, 128n41, 128n42, 128n54, 128n56, 128n57, 130n90 Baltussen, Han 3, 12, 22n9, 23n43, 23n56, 24n56 bandhakī 35, 61, 64 Barbato, Melanie 94, 101, 127n21, 128n48, 130n90, 131n103 Basham, A.L. 129n75 Belvakar, S.K. 156 Bhadreśvara 92 Bhagavad Gītā (BhG) 20, 61, 62, 67, 101, 128n43, 135, 143, 148, 156, 175n95 Bhagavadgītābhāṣya 174n86 Bhagavatpāda 168 bhākta 65 bhakti 16, 30 Bhāmatī 151 Bharata 159 Bhartṛhari 94 Bhāskara 144, 149 bhāvana-mayī-prajñā 34, 123 Bhāvasena 122, 128n55 Bhāviveka xi, 1, 7, 10, 12, 17, 19, 23n54, 24n57, 30, 58, 59, 76n2, 76n7, 76n9, 76n13, 78n39, 78n41, 79n43, 80n48, 81n57, 82n72, 83n81, 84n106, 85n112, 85n117, 86n128, 88n145, 88n147, 89n154, 90n170, 90n178, 90n181, 91, 96, 102, 103, 105, 107, 109, 110, 111, 113, 118, 123, 127n3, 127n10, 129n59, 130n77, 132n119, 133, 141, 143, 144, 147, 148, 151, 154, 160, 163, 168, 169n8, 170n18, 170n19, 171n53, 172n54, 172n55, 173n69, 176n114, 176n115, 179, 180; Bhāvaviveka 17, 25; Bhavya 9, 25, 76n6 Birch, Jason 169n8 Bodhicaryāvatāra 82n70, 171n52 bodhicitta 39, 90n167 Bodhisattva 37, 38, 39, 42, 45, 69, 74, 75, 82n71, 180 Brahmā 53, 59, 60, 64, 65, 66, 147 Brahmajālasutta 27, 64 brahman 17, 62, 120, 134, 137, 147, 148, 150, 151, 153, 154, 162, 181, 182 Brāhmaṇa 51, 54, 55, 66, 70, 74, 92, 127n30, 144 Brahmanism 27, 30, 111, 112, 136, 139, 141, 172n56, 172n58 brahmanization 97 Brahmasūtra 135, 143, 147, 148, 164, 175n95

Brahmasūtrabhāṣya (BSB) 147, 153, 156, 158, 159, 163, 164, 165, 170n16, 171n47, 171n50 Bṛhadāraṇyaka Upaniṣad (BAU) 14, 136, 153 Bronkhorst, Johannes ix, 28, 77n22, 90n170, 98, 127n31, 127n33, 128n34, 129n75, 131n102, 134, 146, 147, 149, 169n4, 171nn42–43, 171n48, 171n51, 172nn56–58, 173n60, 173n62 Brummett, Barry 114, 117, 119, 131n93, 131n100, 131n104 Buddha 37, 39, 41, 45, 48, 67, 69, 71, 89n155, 95, 96, 97, 115, 119, 148; Sugata 121 Buddhacarita 39 Buddhapālita 25, 31 Buddhism 1, 8, 13, 14, 15, 26, 27, 28, 29, 31, 41, 68, 69, 75, 78n38, 79n44, 80n48, 83n81, 85n117, 89n162, 90n167, 97, 108, 109, 111, 112, 119, 120, 138, 141, 142, 145, 148, 149, 150, 162, 171n53, 172n54, 172n58, 176n113, 178, 183n3 Burnouf, Eugène 25, 76n1 cakravartin 41, 136, 180 Candrakīrti 31 cāritra 15, 106, 118, 120, 123, 124, 152 Cārvāka 29, 112, 163 Catherine Morice-Singh 173n78 Catuḥśataka 26 catuṣkoti 60 causation 63, 65, 102 Chāndogya Upaniṣad 87n135, 160 Chapple, Christopher Key 88n143, 92, 127n13, 130n90, 132n116 cintā-mayī-prajñā 34, 123 Citrakūṭa 93 classification 3, 5, 6, 8, 21, 22, 39, 67, 105, 140, 109, 120, 122, 160 Clayton, John 28, 77n23 codanā 145, 146, 147 Collins, Randall 31, 77n31 conceptual xii, 3, 34, 35, 80n48, 80n49, 102, 105, 123, 152, 154, 162, 179, 180; conceptuality 34, 35, 80n48, 125 contemplation 35, 62, 79n48, 103, 120, 122, 123, 124, 132n113, 136, 150, 155, 161, 181 conventional 16, 17, 19, 20, 21, 33, 34, 35, 36, 38, 39, 40, 41, 57, 59, 71, 79n46, 80n48, 99, 100, 102, 109, 110, 117, 120, 121, 125, 142, 152, 153, 154, 179, 180, 182; conventions 35, 36, 37,

202 Index 40, 71, 73, 80n53, 110, 117, 118, 120, 152, 154, 179 conversion 12, 18, 19, 50, 92, 138; converting 19, 33, 92, 100, 117, 128n47, 137 Dasgupta, Surendranath 3 darśana 6, 7, 14, 15, 16, 17, 19, 28, 29, 40, 42, 47, 51, 63, 65, 76, 80n48, 81n56, 81n57, 82n71, 102, 103, 105, 106, 107, 109, 111, 112, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 128n59, 129n59, 130n76, 131n112, 132n113, 133, 140, 152, 153, 154, 155, 160, 169, 175n102, 178, 179, 180, 181, 182, 183; darśanatva 107; darśanic 29, 103, 107; mithyā-darśana 16, 106, 152, 153, 180; samyag-darśana 14, 17, 19, 103, 106, 107, 121, 124, 152, 153, 154, 181, 183n3; six 43, 105, 107, 112, 113, 115, 121, 122, 126, 130n76, 180 debate 12, 16, 20, 24n63, 26, 27, 28, 31, 32, 38, 39, 49, 50, 53, 57, 72, 74, 75, 80n48, 90n178, 91, 92, 93, 94, 101, 103, 110, 111, 112, 113, 118, 119, 122, 126, 129n66, 134, 135, 155, 156, 161, 162, 163, 165, 175n94, 182; debater 12, 26, 31, 39, 50, 70, 93, 94 dependent origination 33 deva 37, 121, 122, 168 devatā 42, 83n87, 113, 115, 120, 121, 122, 123, 124, 125, 125, 126, 220 dharma 44, 47, 51, 58, 60, 64, 71, 85n111, 95, 98, 100, 101, 110, 122, 123, 124, 126, 137, 146, 147, 148, 152 Dharmakīrti 35, 72, 77n37, 80n51, 94 Dharmapāla 25 Dharmarāja 150 dialectic 3, 9, 11, 13, 14, 19, 21, 23n56, 38, 75, 81n69, 96, 99, 104, 115, 116, 118, 124, 125, 134, 135, 139, 140, 142, 152, 160, 161, 163, 165, 167, 178; dialectically 6, 73, 74, 104, 124, 125, 167 dialectical xii, 1, 2, 4, 5, 6, 7, 8, 10, 11, 12, 14, 15, 19, 20, 21, 24n56, 26, 28, 29, 31, 32, 33, 34, 38, 40, 41, 42, 43, 48, 49, 50, 54, 55, 56, 59, 71, 72, 74, 83n81, 84n105, 90n178, 92, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 109, 110, 111, 113, 114, 115, 116, 118, 122, 124, 125, 127n3, 133, 139, 152, 154, 160, 161, 162, 164, 165, 167, 168, 170n20, 176n113, 178, 179, 180, 181, 182, 183n3

dialogue xii, 3, 7, 9, 19, 21, 22, 34, 49, 50, 51, 72, 76, 95, 96, 98, 119, 167; dialogic 1, 19, 21; dialogical 3, 17, 19, 20, 29, 95, 96, 97, 109, 111, 114, 119, 178; dialogism 21; dialogues ix, 6, 18, 19, 50, 73, 96 Diels, Hermann Alexander 2, 3, 5, 11, 22n2, 22nn4–5, 22n10–11 Dīghanikāya 27, 64, 80n50, 108 digvijaya 138, 139 Diṅnāga 25, 27, 34, 50, 55 disācara 108 Dravyasaṅgraha 118 dṛṣṭi 14, 15, 16, 17, 27, 45, 81n56 duḥkha 16, 45, 100, 101, 116, 124 Dundas, Paul 127n15, 127n30, 127n32, 128n35 Dvātriṃśikā 91 Eckel, Malcolm David 8, 12, 23n41, 31, 43, 76n2, 77n18, 77n35, 79n45, 81, 83n90, 83n94, 128n59, 179, 183n1 Ejima, Yasunori 26 ekaṃsika 95 Eltschinger, Vincent 18, 19, 28, 29, 31, 36, 39, 76, 145, 163, 164 endoxa 17, 23n56 epistemology 17, 31, 99, 101, 102, 118, 126; epistemological 20, 28, 34, 36, 38, 55, 77n36, 94, 97 exegesis: exegetical 4, 49, 111, 134, 142, 144, 148, 149, 154, 163, 168 Falin 25 Flügel, Peter 127n32 Folkert, Kendall Wayne 8, 22n25, 91, 103, 105, 106, 107, 109, 113, 121, 122, 123, 127nn6–9, 128n51, 128n55, 129n63, 129n68, 130n91, 131n91, 131n92, 131n109, 131n112, 132n115 freedom xii, 16, 17, 28, 37, 67, 74, 101, 148, 149 Frisch, Max Rudolf 141 Fuller, Paul 14, 15, 23n50 gaṇadhara 120 Gandhāra 98 Ganeri, Jonardon 24n62, 77n34, 161, 162 Gārgī 162 Gauḍapāda 135, 174n85 Gauḍapādiyakārikā (GK) 176n114 Gautama 31, 50, 106 genre xii, 1, 2, 4, 5, 7, 8, 10, 11, 13, 40, 91, 103, 104, 113, 133, 156, 159, 182

Index  203 Gerety, Finnian McKean Moore 62, 87n136, 87n138 Gerschheimer, Gerdi 38, 67, 109, 130n80, 130n83, 130n84, 130n85, 131n109, 175n102 Gethin, Rupert 183n3 Ghosh, Suresh Chandra 30, 77n29 gnostic 21, 74, 125, 141, 145, 151 Gnosticism 141 Gokhale, V.V. 30, 37, 39, 76n2, 76n11, 78n41, 81n6, 139 Gokulbhai, Patel Maheshkumar 156, 160, 166, 168, 173n77, 175n88, 175n93, 176n106, 176n116, 177n130, 177n133 Gosāla 108 Govinda 135, 174n79 Gracia, Jorge J. E. 104, 128n58 Guṇaratna-Sūri 103 Gupta 29, 30 Gurvāvalī 111 Hacker, Paul 146, 155, 156, 157, 158, 159, 160, 165, 167, 174n82, 174n85, 174n88, 175n98 Hadot, Pierre 15, 16, 18, 19, 20 Halbfass, Wilhelm 7, 8, 11, 22nn20–23, 23n49, 38, 40, 63, 81n65, 83n79, 88n141, 91, 107, 114, 127n2, 129n66, 131n94, 133, 151, 156, 160, 169n1, 173n67, 174n82, 175n105, 176n116, 177n125; India and Europe 7, 8, 156 Ham, Hyoung Seok 52, 53, 57, 59, 63, 72, 76n2, 81n56, 83n99, 84n107, 84nn109– 110, 90n171 Hara 61, 65 Hari 59, 60, 61, 62, 65, 86n124 Haribhadra 1, 3, 17, 18, 21, 26, 42, 63, 83n87, 91, 92, 93, 94, 95, 97, 100, 101, 102, 103, 105, 106, 107, 109, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 127n10, 127n12, 128n53, 128n55, 130n87, 130n90, 132n116, 133, 137, 139, 143, 154, 162, 168, 180 Harimoto, Kengo 174n86 Hastāmalaka 135 Haṭhayogaprādīpika (HYP) 61 Hemacandra 111, 122 hermeneutic/hermeneutical/hermeneutics xii, 1, 14, 20, 24n58, 35, 38, 39, 49, 51, 72, 123, 124, 148, 150, 151, 152, 154 heterodoxy 21, 94, 95, 97, 109, 110, 113, 117, 118, 136, 152, 180, 182, 183n3 hetu 27, 29, 33, 36, 40, 37, 86, 75

hetu-vidyā 27, 28, 35, 75, 77n36 hīnayāna 138 Hindu 59, 62, 120, 139, 140, 141, 172n58 Hirst, Jacqueline Suthren 135, 137, 138, 139, 143, 147, 153, 154, 155, 162, 163, 169n6, 169nn11–12, 170n16, 170n30, 171nn49–50, 173nn71–75, 176n109, 177n117 historiographical 3, 9, 11, 12, 14, 19, 32, 33, 43, 51, 53, 103, 104, 112, 161, 165 Holquist, Michael 21, 24n69 Hopkins, Jeffrey 8, 22n30 Hsu, Chien Y. 25, 26, 34, 76n2, 76n3, 76n7, 76n9, 76nn13–14, 79n47 humour 32, 65, 72, 73, 74, 107 ignorance 16, 23n52, 37, 38, 85n112, 107, 115, 119, 124, 145, 150, 151, 154, 163, 177n118, 180, 181 Iida, Shotaro 30, 32, 76n1, 76nn5–8, 76n13, 77n30, 78n38, 83n93, 83n99 Indra 37, 161 inference 31, 35, 36, 37, 38, 50, 57, 70, 81n67, 109, 110, 122, 123, 126 intellectual exertion 17, 23n55, 100, 101, 102, 150, 180, 182 irony 7, 19, 50, 53, 58, 64, 65, 71, 73, 74, 75, 90n181, 119, 140, 173n58; ironic 59, 64, 66, 68, 73, 74, 75, 76, 127n15 īśvara 62, 64, 168 Īśvara 45, 121, 124 Jacobi, Hermann Georg 92, 127n11, 127n17, 155, 174n88 Jaimini 125, 138, 143, 146, 148, 168 Jaiminīya 92, 112, 121, 125 Jain, Nemi C. 120, 131n107 Jaina 8, 9, 10, 11, 17, 26, 27, 29, 38, 40, 42, 63, 83n79, 83n89, 89n154, 89n156, 90n162, 91, 92, 93, 94, 95, 96, 97, 98, 101, 102, 103, 104, 105, 106, 107, 109, 112, 113, 114, 115, 118, 121, 122, 123, 124, 125, 127n15, 127n29, 128n55, 129n75, 130n76, 130n77, 130n90, 152, 162 Jaini, Padmanabh S. 127n32, 129n75, 131n102 Jainism 13, 14, 15, 23n55, 24n57, 41, 67, 92, 94, 102, 103, 105, 106, 107, 108, 109, 111, 114, 117, 118, 119, 120, 148, 149, 150, 164, 178 Jayanta 111, 143 Jayanta Bhaṭṭa 111 Jayatilleke, K.N. 95

204 Index Jean François Pons 43 Jina 17, 96, 97, 99, 102, 115, 116, 117, 119, 121, 134, 180 Jinabhaṭṭa 92 Jinadatta 92, 128n55 jñāna 20, 36, 39, 49, 55, 101, 102, 103, 106, 107, 110, 115, 118, 120, 121, 122, 124, 136, 142, 145, 148, 151, 152, 154, 163, 172n55, 178 jñāna-kaṇḍa 55, 81n56, 138, 148 jñāna-mārga 16, 20, 37, 41, 42, 52, 106, 110, 122, 125, 149, 151, 152, 153, 154, 162, 167, 179, 180 jñāna-mārgin 37, 42, 52 jñāna-yoga 16, 17, 20, 23n55, 24n57, 100, 101, 102, 106, 109, 116, 118, 120, 123, 125, 150, 154, 180 jñāna-yogin 16, 150, 151, 155, 167 Johnson, W.J. 101, 127n29, 128nn45–46 Jonathan, Bader 137 Kahāvalī 92 kaivalya 106 Kalyanavijaya 93 Kamalaśīla 18, 35, 75, 76n13 Kant, Immanuel 79n48 Kapadia, H.R. 129n71, 129n72, 129n73 kāraṇa 63 Kapstein, Matthew T. ix, 18, 24n61, 90n179 Karatalaratna (KR) 26, 76n10 karma 15, 16, 22, 41, 49, 52, 55, 57, 63, 64, 65, 66, 67, 68, 69, 75, 94, 99, 100, 101, 102, 106, 108, 112, 115, 118, 119, 122, 136, 138, 142, 144, 147, 148, 149, 151, 152, 169n8, 172n55, 178, 179, 180 Karṇa 86n128 Kaṭha Upaniṣad 162 Kātyāyana 144 Kauṭilya 106, 110 Kāvyamīmāṃsā 110

Kawasaki, Shinjo 51, 84n107

Keśava 59 kevala: kevalin 118 King, Karen 141, 170nn23–24 klṛptā 61 knowledge 12, 15, 16, 17, 20, 23n55, 30, 32, 33, 34, 35, 36, 37, 38, 39, 41, 42, 44, 48, 49, 50, 51, 52, 54, 55, 57, 59, 69, 70, 71, 74, 77n36, 81n60, 82n71, 84n105, 85n111, 92, 95, 98, 100, 101, 102, 103, 105, 106, 107, 110, 111, 115, 118, 119, 120, 121, 122, 123, 124, 126, 129n67, 130n81, 131n111, 136, 137, 138, 140, 142, 143, 145, 146, 147, 148, 149, 150,

151, 153, 154, 158, 162, 163, 178, 179, 180, 182 kriyā 47, 51, 55, 58, 158; mānasī kriyā 63 kriyā-kaṇḍa 55, 81n56 Kṛṣṇa 41, 59, 61, 65, 84n106, 124 kṣaṇika 98 Kumārila Bhaṭṭa 29, 168; Kumārila 50, 94, 135, 138, 144, 148, 164, 172n57, 172n58, 177n132 Kundakunda 118 kūrma 61 Kuṣānas 97 kutarka 110 Laertios, Diogenes 7 Laghuvṛtti 128n53 Laks, André 3, 6, 159 Lal Sanghavi 93 Larson, Gerald James 124 Laugakṣi Bhāskara 144 liberation 13, 16, 21, 30, 31, 34, 37, 38, 41, 49, 52, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 65, 66, 68, 81n60, 88n140, 100, 101, 106, 110, 116, 121, 124, 125, 145, 146, 147, 148, 149, 151, 152, 153, 154, 162, 168, 171n53, 172n57 Liebich, Bruno 155, 156, 157, 158, 159, 165 Lindtner, Christian 26, 37, 39, 60, 61, 62, 64, 65, 106 literary technology(ies) 13, 23n44, 32, 75, 77n36, 96, 103, 120, 162, 183n3 Lokatattvanirṇaya 122 Lokātītastava 39 Lokāyata 26, 45, 109, 111, 114, 121, 122, 126, 129n67, 130n77, 130n82, 131nn111–112; Lokāyatas 12, 98, 112, 166, 174n81

Lorenzen, David N. 135, 169n5

Mādhava 91, 137 Madhyamaka 13, 14, 26, 27, 31, 32, 33, 39, 40, 41, 51, 68, 80n48, 83n81, 96, 104, 113, 115, 119, 151, 154, 162, 178, 179 Madhyamakahṛdayakārikā (MHK) xi, 1, 7, 10, 12, 19, 25, 26, 27, 30, 31, 32, 33, 37, 38, 39, 40, 41, 42, 43, 48, 49, 50, 51, 54, 55, 60, 62, 63, 69, 70, 72, 73, 75, 76, 76n2, 76n11, 77n18, 77n37, 78nn38–40, 78n42, 79n45, 80n53, 80nn55–56, 81n58, 81nn62–64, 82n71, 83n74, 83n81, 83n85, 84n103, 84nn106–109, 85n112, 85n114, 85n116,

Index  205 85n118, 85n120, 86nn121–123, 86n126, 86n128, 87nn129–131, 87n140, 88n144, 88nn146–148, 89nn149–150, 89n161, 90n178, 91, 96, 103, 107, 110, 113, 114, 119, 121, 122, 125, 127n3, 128n59, 129n59, 130n77, 133, 144, 147, 148, 153, 160, 162, 163, 167, 169n8, 170nn18–19, 172n54, 173n69, 176n115, 178, 179, 181, 183n2 Mādhyamika 26, 27, 41, 166, 174n81 Mahābhārata 67, 84n106, 86n128, 136, 139, 168, 174n81 mahāvākya 151, 154 Mahāvīra 95, 96, 97, 115 Mahāyāna 26, 35, 45, 71, 123, 138, 148 Maimonides 162, 176n112 makṣikā 61 Māṇavaka 95 Maṇavālamāmuni 174n80 maṇḍala 19, 180 Maṇḍana Miśra 135, 138 Maṇibhadra 121, 128n53 Maṅkha 111 Mansfeld, Jaap 2, 3, 4, 5, 6, 10, 11, 12, 13, 22n1, 22nn4–5, 22n8, 22nn10–11, 22n13, 22n15, 22n18, 23n40, 160 mantra 14, 47, 49, 87, 88, 143, 144, 220 Manusmṛti 110 mārga 14, 16, 20, 37, 41, 48, 54, 57, 58, 59, 63, 69, 75, 105, 106, 110, 116, 121, 122, 123, 124, 125, 145, 148, 149, 151, 152, 153, 154, 162, 167, 179, 180 mata 14, 15, 168, 175n102 Matilal, B.K. 85n113, 85n115, 95, 96, 97, 115, 119, 127n23, 127n25, 127n26, 127n27, 127n28, 128n49, 130n90, 131n95 māyā 143, 145 Mayeda, Sengaku 156, 174n86, 174n87 McClintock, Sara 18, 24n59, 24n63, 24n65 McLuhan, Marshall 170n20 medicine 3, 14, 37, 50, 103, 107, 118, 152 meditation 33, 41, 49, 63, 150, 151, 183n3 Menezes, Walter 150, 151, 173n65 Merutuñga 128n55 middle way 32, 95, 96, 115, 116, 181 Mimaki, Katsumi 9 Mīmāṃsā 1, 7, 11, 20, 27, 30, 36, 41, 42, 43, 49, 50, 51, 52, 53, 54, 55, 56, 57, 59, 63, 69, 74, 84n106, 87n134, 92, 105, 110, 111, 112, 114, 122, 125, 126, 130n77, 134, 135, 137, 138, 143, 144, 145, 146, 147, 148, 149, 151, 152, 154, 155, 163, 164, 166, 167, 168, 169n8, 171n37, 172n55, 172n57, 174n81, 178;

Mīmāṃsaka 20, 29, 41, 49, 50, 51, 52, 53, 58, 59, 106, 143, 144, 146, 147, 172n55 Mīmāṃsāsūtra (MS) 51, 109, 125, 143, 146, 147, 149 mithyā 15, 16, 62, 107, 107, 116, 117, 152, 153, 181 mithyātva 105, 116 mokṣa 16, 21, 44, 57, 58, 66, 101, 105, 106, 121, 122, 123, 124, 145, 147, 148, 149, 168, 169 Mount Meru 38, 109, 115 mukti 168 Mūlamadhyamakakārikā (MMK) 26, 32, 36, 37, 153 Muller, Albert Charles 15, 23nn47–48 Müller, Max 3, 22n6, 87n137 Mun, Chanju 23n35 Muṇḍaka Upaniṣad 17, 153, 161 Municandra Sūri 93 Munisundara 111 Nāgārjuna 26, 29, 31, 32, 33, 39, 79n46, 80n54, 82n71, 102, 154 Nālandā 25 nāmadheya 144 narrative 11, 13, 34, 50, 134, 135, 137, 139, 145, 160, 164, 178 nāstika 30, 112, 126, 136 Nāṭyaśāstra (BhN) 159, 175n100 Naugle, David K. 79n48 naya 14, 15, 100, 117, 124 neti neti 145, 151, 181 Nicholas of Cusa 99 Nicholson, Andrew J. 8, 10, 11, 23n38, 91, 103, 167 Nicolson, Hugh 177n124 nididhyāsana 150, 151, 155 niḥśreyasa 16, 106 nikṣepa 100 nimitta 62 nirvāṇa 15, 16, 36, 123, 148, 178 niṣedha 144 nodanā 126 no-view 33, 40, 80n48, 104, 178 Nyāya 27, 55, 109, 111, 112, 114, 121, 144, 145, 153, 166, 167; Naiyāyikas 112, 166, 174n81 Nyāyamañjarī 111 Nyāyasūtra 31, 32, 50, 106 Oṃ 62, 63, 65, 87n136, 150 omniscience 41, 42, 48, 71, 116, 117, 118, 125, 130n77, 139, 148, 180; omniscient

206 Index 48, 52, 53, 71, 98, 99, 101, 102, 120, 126, 140, 181 orthodoxy 28, 30, 118 orthopraxy 118 Padmapāda 135, 170n16 Pahlajrai, Prem 107, 131n99 paramārtha 33, 124 pāramārthika 21, 118, 142, 152 parampara 133 parāvidyā 17, 162 parīkṣā 17, 27, 28, 36, 77n20, 163, 164, 177n120 Parpola, Asko 173n59 Pātañjalayogaśāstravivaraṇa (PYŚV) 62, 156, 158 Patañjali 158, 159, 166, 167, 168 pedagogic 19, 164 pedagogy 34, 72, 134, 135, 155, 156, 162, 167, 176n111, 181, 182; pedagogical 16, 34, 35, 143, 160, 161, 162, 165, 166, 176n110 performative 16, 125, 136, 155, 167 Petzold, Bruno 8, 22n29 Plato 7, 134, 180; Platonic 18, 50 political 1, 3, 9, 21, 27, 28, 29, 30, 72, 90n168, 111, 114, 119, 137, 138, 139, 153, 172n58, 173n58, 178, 181 Potter, Karl H. 128n53, 156 Poṭṭhapāda 95 Prabandhakośa 92 Prabhācandra Sūri 92 Prabhākara Miśra 168 Prabhāvakacaritra 92 pracchanna-bauddha 140, 176n115 Prajāpati 136, 161 prajñā 34, 39, 61, 80n49, 123 Prajñāpradīpa (PP) 25, 26, 31, 90n178 prakaraṇa 93, 158, 166 Prākrit 93, 127n32 Prakṛti 16, 124, 125 pramāṇa 17, 21, 27, 29, 33, 48, 57, 71, 77n36, 110, 111, 121, 122, 126, 143, 145, 146, 168, 169; pramāṇatva 38, 47, 51, 57, 110 Pramāṇasamuccaya (PS) 27 Pramāṇattiraṭṭu 174n80 praṇava 62 prāṇāyāma 61, 158 prapañca 34, 104 prasaṅga 31, 153 Prasthānatrayī 133, 148, 150, 158, 175n95 pratipatti 36, 110 Pratītyasamutpādahṛdayakārikā 39

pratyāhāra 61, 167 pratyakṣa 34, 65, 110, 122, 126 praxis 21, 35, 38, 75, 101, 123, 124, 145, 151, 154, 163, 164, 173n66, 182 propaedeutic 76, 77n36, 102, 154, 159, 160, 162, 164, 176n113; propaedeutical 39; propaedeutics 17, 118, 163, 178 psychagogic 19 pudgala 44, 98 Purāṇa 84n106, 136 puruṣa 17, 60, 61, 71, 87n133, 106, 124, 125, 136, 184 Pūrva-Mīmāṃsā 63, 138, 168 Pūrvamīmāṃsāsūtra 149 pūrvapakṣa 9, 10, 45, 46, 47, 49, 50, 53, 54, 57, 58, 59, 69, 70, 84n100, 84n107, 132n119, 147, 172n53 quodammodo 95, 100, 116 Qvarnström, Olle 7, 8, 9, 10, 11, 23n55, 27, 41, 63, 91, 100, 103, 107, 176n115 Radhakrishnan, Sarvepalli 3, 170n29 Rāhula Sāṅkṛtyāna 32 Rājaśekhara 92, 110, 111, 128n55 rāja-sūya 136, 137 rākṣasa 139 Rāmānuja 140, 149, 176n115 Ram-Prasad, Chakravarthi ix, 145, 149, 150, 151, 152, 170n29, 171nn39–40, 173nn63–64, 173n68 Rañjanā 32 Raṅgācārya, M. 155, 156, 157, 158, 166, 167, 169n2, 173n78, 174n79, 174n80, 175n91, 177n130 reason 5, 15, 17, 34, 35, 36, 42, 43, 49, 52, 55, 56, 62, 67, 68, 70, 72, 74, 83n92, 86n127, 93, 95, 97, 110, 111, 116, 125, 130n81, 131n97, 142, 143, 145, 146, 152, 155, 164, 167, 176n112, 176n116, 179; reasonability 55, 59; reasoning 10, 21, 23n55, 24n57, 28, 30, 32, 33, 35, 36, 37, 38, 49, 50, 51, 55, 56, 60, 67, 69, 70, 71, 76, 80n49, 85nn112–113, 88n145, 97, 98, 100, 109, 110, 123, 125, 128n47, 143, 145, 146, 147, 153, 164, 172n58, 177n122, 178, 179, 182 reductio ad absurdum 27, 72, 179 religious xii, 1, 2, 3, 9, 11, 12, 13, 14, 15, 17, 18, 21, 26, 27, 30, 35, 48, 55, 72, 74, 89n154, 90n168, 93, 94, 95, 96, 97, 109, 110, 116, 119, 120, 123, 127n29, 127n32, 134, 135, 136, 142, 143, 145, 148, 151, 152, 153, 162, 163, 164, 165,

Index  207 172n56, 172n58, 173n58, 176n112, 176n116, 181, 182 rhetoric 19, 40, 43, 54, 55, 72, 88n147, 105, 113, 114, 115, 119, 142, 169n10; rhetorical 8, 11, 19, 49, 50, 51, 53, 54, 57, 59, 72, 73, 95, 113, 114, 115, 117, 118, 119, 134, 168 ritual 20, 41, 42, 49, 53, 54, 55, 57, 58, 59, 63, 65, 66, 67, 68, 73, 74, 85n116, 89n154, 109, 111, 112, 126, 136, 138, 139, 140, 147, 148, 149, 151, 152, 167, 169n8, 171n35, 178 Rudra 65, 139 170n19 Ruegg, David Seyfort 8, 23n32, 76n2 Runia, David Theunis 3, 4, 5, 12, 13, 22n4, 22n8, 22n14, 160 Russell, Bertrand 16, 23n53 Śabara 51, 144, 172n57 śabda 47, 51, 57, 110, 126, 140, 142, 143, 144, 145, 146, 161, 182 Sadānanda 154, 163 Ṣaḍdarśananirṇaya 128n55 Ṣaḍdarśanasamuccaya (SDS) 1, 22n6, 42, 56, 63, 77n27, 91, 92, 94, 100, 103, 104, 105, 107, 109, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 124, 125, 126n1, 128n53, 128n55, 130n86, 131n96, 131n98, 131n108, 131nn110–111, 133, 137, 157, 162, 166, 168, 169, 180 sādhanā 168 Śaiva 12, 27, 60, 137 Śaivism 30 sallekhanā 67, 89n154, 89n156, 119 salvation 41, 122, 136 samādhi 44, 62, 151 Samantabadhra 122 Samarādityakathā 93 Samayasāra 118 Saṃhitā 62 saṃjñāyoga 101 saṃketa 57 saṃketasyānuvāda 109 Sāṃkhya 17, 26, 41, 42, 63, 98, 106, 109, 110, 111, 112, 114, 121, 124, 130n81, 152, 153, 154, 158, 164, 167, 168, 182 Sāṃkhyakārikā (SK) 106, 124 Sāmmaññaphalasutta 108 saṃnyāsa 136; saṃnyāsin 135, 147 saṃsāra 16, 24n57, 36, 38, 63, 68, 70, 80n54, 85n112, 110, 148, 152; saṃsāric 36, 37, 100 saṃśaya 152

saṃvṛti 33 samyag-cāritra 103, 120 samyag-darśana: right view 14, 17, 19, 103, 106, 107, 121, 124, 152, 153, 154, 181 samyag-jñāna 103, 120, 124 samyak 17, 35, 107, 116, 118, 123, 124 samyaktva 105, 108, 121, 122, 123 Śaṅkara 1, 17, 20, 24n57, 62, 63, 77n28, 118, 123, 128n43, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 145, 146, 147, 148, 149, 151, 153, 154, 155, 156, 157, 158, 159, 160, 162, 163, 164, 165, 166, 167, 169n8, 170n19, 170n29, 171n51, 172n55, 172n58, 173n58, 174nn85–88, 175n89, 175n93, 175n97, 175n103, 176n110, 176nn114–115, 177n122, 181 Śaṅkara-Bhagavatpāda 158 Śaṅkaradigvijaya 137 Sanskrit xi, 1, 14, 23n49, 23n54, 25, 26, 32, 41, 43, 58, 61, 64, 69, 77n37, 78n40, 80n48, 84n109, 86n128, 93, 97, 98, 100, 116, 127n15, 127n32, 132n113, 155, 156, 159, 160, 169n3, 170n19, 173n78, 174n80, 175n102, 182 Śāntarakṣita 18, 50, 75 Śantideva 82n70, 171n52 sapta-bhaṅgī 100, 117 Sarvadarśanasaṅgraha (SDS) 63, 91, 133, 136, 165 sarvajña 42, 98, 101, 117, 118, 120, 131n101, 140, 181 Sarvasiddhāntapraveśaka 128n55 Sarvasiddhāntasaṅgraha (SSS) 1, 63, 77n28, 87n134, 133, 134, 138, 142, 153, 155, 156, 157, 158, 159, 160, 161, 162, 164, 165, 166, 167, 168, 169, 171n35, 173n77, 174n79, 174n80, 174n82, 174n84, 175n88, 175n89, 175n93, 175n103, 176n116, 181 Sarvāstivāda 98 śāstra 17, 26, 50, 54, 154, 158; śāstric 26, 28, 50 Śāstravārtāsamuccaya 100 śāśvata 98 Śataśāstra 26 Śatāślokī 156 ṣaṭ-tarkī 109, 110 Sautrāntika 31, 166, 174n81 Sāvitrī 33 scholasticism xii, 12, 182; scholastic 12, 27, 28, 77n20, 77n34, 80n49, 103, 182 school 4, 6, 8, 10, 13, 16, 18, 26, 27, 28, 32, 50, 94, 96, 98, 103, 104, 106, 110, 111,

208 Index 112, 115, 119, 121, 125, 126, 128n59, 129n59, 129n67, 130n77, 130n81, 131n112, 133, 134, 139, 153, 167 scriptures 9, 17, 34, 35, 36, 38, 54, 55, 56, 57, 58, 67, 70, 71, 93, 106, 109, 110, 111, 121, 123, 124, 133, 134, 135, 136, 142, 143, 144, 145, 146, 152, 154, 155, 161, 163, 164, 181, 182 sectarian xii, 1, 3, 18, 19, 29, 42, 67, 112, 138, 139, 152, 165, 176n113, 182 self-transformation 38, 180 Sen, Amartya 21, 24n70 Shastra, Nemicandra 93 Shujun Motegi 125 Shukla, Ram Sajiwan 92, 94, 127n14, 127nn18–20 siddhānta 9, 17, 32, 82n71, 129n59, 142, 152, 155, 161, 175n102 Siddhasena Divākara 98, 127n30 Sin Fujinaga 131n101 Singh, Upinder 38, 82n72, 172n56 Sitz im Leben 11, 160 Śiva 41, 53, 59, 60, 65, 78n41, 86n124, 86n127, 121, 137, 169n8, 170n19 Ślokavārttika (ŚV) 50 smṛti 62, 65, 67 Socrates/Socratic 19, 50, 73 Somatilakaka Sūri 103 Sopa, Geshe 8, 23n34 soteriology: soteriological 11, 17, 35, 38, 40, 48, 53, 63, 66, 71, 80n49, 94, 95, 97, 101, 120, 123, 124, 163, 164, 165, 169n8, 176n116, 180, 181; soteriologically 37, 49, 52, 53, 57, 58, 59, 64, 67, 71 spiritual exercise 15, 19, 155, 176n113; exercice spirituel 16, 23n51; spiritual exercises xii, 16, 18, 19, 24n65, 120, 182; technique du soi 16, 23n51; yogic exercise 18, 20 śramaṇa 30 śravaka 148 śrāvakas 41 Śrīdhara 165 Śrīkaṇṭhacarita 111 śruta 35 śruta-mayī-prajñā 34, 123 śruti 110, 136, 150, 153, 154 Sthiramati 25 strategy 27, 31, 32, 39, 40, 42, 48, 50, 52, 53, 54, 59, 71, 72, 75, 96, 99, 100, 101, 103, 104, 113, 115, 116, 117, 128n47, 148, 154, 160, 172n58 Stcherbatsky, Fedor Ippolitovich 8

Śubhagupta 94 śūdra 47, 53, 58, 85n120, 147 sukha 100, 116, 124 śūnyatā 45, 76, 102, 125, 154; śūnya 74, 184 Śūrapāla 93 Sureśvara 135, 137 Sūtrakṛtāṅga 95 svabhāva 34, 39, 41 svarga 145, 146, 148 svatantra 27, 31, 68 Svātantrika 31 Śvetāṃbara 91, 92, 93, 97, 98, 175n89 Śvetāśvatara Upaniṣad (ŚU) 61, 174n85 syād-vāda 95, 98, 100, 115, 116, 117, 118, 124, 180 Syādvādakucodyaparihāra 95 syāt 100, 117, 118, 119, 180 Taittirīya Upaniṣad 137 Tantra 48, 57, 108 Tāranātha 25 tarka 17, 38, 56, 109, 110, 111, 112, 116, 122, 137, 150, 152, 154, 179, 180; ṣaṭtarkī 77n27, 105, 110, 111; tarka-vidyā 110, 129n67; tarkī 109; tārkika 16, 105, 108, 111, 122, 137, 143, 146, 154, 163, 177n118 Tarkajvālā (TJ) 9, 26, 31, 66 Tarkarahasyadīpikā 103 tat tvam asi 145, 151, 154 Tathāgata 96 tattva 17, 36, 37, 39, 42, 51, 55, 81n57, 83n75, 106, 107, 118, 119, 120, 121, 122, 123, 124, 125, 126, 129n59, 130n89, 131n112, 132n119, 146, 155, 163, 168, 180, 182 tattva-amṛta 17, 37, 180 tattvajñānaiṣaṇā 39 Tattvāmṛtāvatāra 37, 39 Tattvārthasūtra (TAS) 98, 106, 107, 121, 123 Tattvasaṅgraha (TS) 18, 50 teleology 4, 9, 19, 40, 43, 114, 124, 160, 162, 165, 167, 170n20, 178, 179, 180, 183n3 testimonies 6, 7, 27, 160 Theophrastus 3, 4 theōría 122, 131n113 therapeutic 11, 16, 18, 20, 34, 35, 37, 51, 75, 81n60, 102, 105, 122, 124, 145, 152 therapy 14, 38, 75, 76, 81n69, 121, 123, 124, 152 Theravada 14 Tīrthaṅkara 115, 120

Index  209 tīrthika 97, 111 Toṭakācārya 135 Traividyaviracita 128n55 transformative 2, 14, 19, 21, 178 Trimūrtī 59 udgitha 62 ultimate xii, 15, 16, 17, 19, 20, 21, 33, 34, 35, 36, 39, 40, 45, 70, 79n46, 80n48, 83n81, 101, 102, 110, 114, 117, 118, 120, 124, 125, 137, 138, 142, 145, 147, 149, 152, 153, 154, 162, 164, 173n58, 179, 180, 181 Upadeśasāhasrī (US) 123, 147, 156, 174n87 upamāna 126 Upaniṣad 14, 17, 38, 61, 137, 153, 160, 161, 162 upāsana 168 Uttara-Mīmāṃsā 122, 138, 152, 169n8 Uttaramīmāṃsāsūtra 149 uttarapakṣa 9, 10, 19, 36, 45, 46, 47, 49, 52, 53, 54, 57, 58, 69, 70, 71, 113, 132n119, 147, 160 Vācaspati Miśra 158 vāda 15, 27, 28, 29, 31, 49, 50, 56, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 113, 115, 116, 117, 118, 124, 162, 163, 168, 180 vādin 56, 95, 96 Vādinadeva Sūri 93 Vaiśeṣika 12, 26, 27, 41, 42, 46, 109, 111, 112, 114, 121, 122, 157, 164, 166, 167, 174n81 Vaiṣṇava 27, 137, 174n80 Vaiṣṇavism 30 vākyārtha-vidyā 111 Vasunandin 122 Veda 17, 41, 47, 48, 51, 52, 53, 54, 55, 56, 57, 58, 59, 62, 63, 64, 66, 69, 70, 71, 72, 74, 77n27, 78n41, 81n56, 85n111, 85n119, 85n120, 87n134, 108, 110, 112, 121, 126, 130n87, 135, 136, 137, 138, 139, 140, 143, 144, 145, 146, 147, 148, 149, 153, 154, 161, 172n53 Vedānta 9, 10, 13, 14, 15, 40, 41, 42, 46, 55, 90n181, 98, 110, 122, 130n77, 138, 140, 141, 142, 148, 149, 150, 157, 160, 163, 164, 166, 167, 168, 169n8, 170n29, 173n58, 173n61, 174n81, 175n88, 175n95, 176n114, 178, 179, 181 Vedāntasāra (VS) 154, 155, 163, 168 Vedāntasūtra 168

Verardi, Giovanni 30, 31, 76n4, 77n26, 77n34, 170n18, 172n56 Vetter, Tillman 156 vibhajya 95, 110 vibhajya-vāda: vibhajya-vādin 95, 96, 97 vibhajya-vādin 95, 96 vidhi 126, 144, 147, 168 vidyā 27, 28, 35, 75, 77n36, 106, 110, 111, 129n66, 129n67, 138, 143 Vidyabhusana, Satis Chandra 8 Vidyādharagaccha 92 view (s) xii, 2, 5, 6, 13, 14, 15, 17, 19, 20, 21, 22, 23n53, 33, 34, 38, 40, 43, 50, 53, 65, 70, 75, 76, 79n48, 80n48, 83n81, 83n87, 90n162, 98, 102, 104, 105, 106, 107, 113, 114, 115, 116, 118, 126, 129n72, 130n91, 142, 151, 154, 155, 160, 163, 164, 167, 168, 178, 180, 181 vijaya 137 vikalpa 62, 65 virahāṃka 93 Vīraśaivism 30 Virocana 161 Viśiṣṭādvaita 140, 176n115 Viṣṇu 53, 59, 60, 61, 66, 67, 84n106, 86n127, 137 Viśvarūpa 135 Viśvatattvaprakāśa 122 vīta 125 vitaṇḍā 32 Vivaraṇa 151 viveka 16, 20, 25, 55, 70, 121, 124, 152, 154 Vivekacūḍāmaṇi 150, 156 Vivekavilāsa 128n55 von Harnack, Adolf 141 vṛtti 15, 93 vyavahāra 117, 152 vyāvahārika 21, 118, 142, 152 Warder, A.K. 8, 22n24, 76n13, 76n15, 130n91 Westerhoff, Jan 40, 82n73, 83n80, 83n82 Wezler, Albrecht 27, 81n60 Whaling, Frank 140, 141, 170n21, 170n25, 170n26, 170n27, 177n123 Winternitz, Moriz 8, 22n26 women 47, 53, 58, 59, 85n120, 86n128, 147, 171n46, 171n53 worldview 8, 10, 11, 14, 17, 19, 23n45, 28, 34, 38, 39, 40, 42, 49, 50, 72, 80n48, 84n106, 98, 99, 106, 113, 114, 115, 130n82, 131n111, 132n113, 149, 152, 163, 178, 180

210 Index Xuanzang 12, 25 yajña 78n41, 151 Yājñavalkya 162 Yākinī Mahattarā 92 Yāmuna 149 yati 92 yoga 9, 20, 34, 41, 48, 52, 53, 56, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 87n134, 87n139, 93, 100, 108, 110, 112, 120, 123, 130n90, 145, 151, 154, 158, 164, 167, 169n8, 172n58, 174n81, 178, 179, 182; dhyāna-yoga 63; rāja-yoga 136,

169n8; śaiva-yoga 65; yogic 16, 18, 20, 61, 62, 63, 123, 151, 155, 167, 173n66 Yogācāra 42, 45, 119 Yogācārabhūmi 35, 123 Yogācārabhūmiśāstra (YBhŚ) 27, 163 yogācārins 41 Yogadṛṣṭisamuccaya (YDS) 63, 123, 132n116 Yogasūtra (YS) 61, 62, 87n134, 158, 167, 175n93 yogic exercise 18, 20 yogin 16, 17, 20, 98, 158, 167 yukti 17, 36, 55, 79n44, 116, 118, 125, 146 Yuktidīpikā (YDī) 125